At work we have an email account set up for this. We replace the user's email with deleted+<uuid>@<company>. Any emails sent end up in one of our inboxes.
My brain dislikes justified paragraphs. Can't put my finger on why. I just find them less pleasant to read. Even if the word spacing is mostly uniform.
I believe the ragged-right whitespace pattern offers visual landmarks that make it easier to keep track of the line you're reading. No such cues in justified paragraphs.
Do you dislike them only on the web, or in print, too? It might be worth looking at some books published between about 1880 and 1950 to see what justified paragraphs are really supposed to look like, if you haven't already.
I created this stylesheet which hides everything but the videos and makes videos smaller so I can fit more on the screen, and I've been very happy with the results.
I'm browsing from a low-ish speed connection today, so it's nice to load the informative details before the pretty pictures.
The text lets me decide if it's worth waiting for the pictures to load or not.
When the screenshots are inline, the other parts of the README move around as they finish loading, making it harder to read as I wait for them to load.
Also, it's a nice informative README. No notes from me. If you want the screenshots, they make them very easy to find.
It's a link to their git and the readme, the home page itself has screenshots (https://git-cola.github.io/index.html). Maybe that would be a better link.
Sorry, I've thought that direct source code link will be better than redirection page with little info and images, and didn't realised there were no screenshots included in the README file.
Personally, I don't like images embedded in README file, that's because I'm expecting to provide only the necessary and important instructions on how to use and build the project. No docs, code snippets, images, and all that.
I'm with you. Their README is not a low-effort README; I find it quite informative, and I'm glad you linked it instead of their marketing site. They included a link to screenshots right at the top of the README, and anyone who couldn't be bothered to click that but calls the README "low-effort" puzzles me.
Last time somebody shared here a link to WASM-focused language. You land on the marketing page with two lines of text, “Getting Started” prompting info about being not written yet, broken GitHub button below, and the only way to get into its page was another GH link in the navbar that actually worked…
We don’t want a European Google Maps! We want our institutions at all levels to contribute to OpenStreetMap (which was created by a British citizen, by the way).
Well, I still would like to see a navigation app based on OpenStreetMap, which is as good as Google Maps. Here We Go is considered the European alternative, but it is not as reliable and lacks some convenient integrations. I don't see how Open Source could help here.
The biggest take is that it is impossible to have fine-grained traffic data like Google without having billions of spyware enabled smartphone on the road. You just cannot do it.
2. I would like to stop comparing solutions on a pure technical basis. "I would like to quite Google Thing but the open source solution I tried doesn’t have the shiny color I like"
There’s so many problems with that kind of answers.
Firstly, it means you value "shiny coloured featured" more than political independence and ethical value. That’s a huge problem. So, basically, you don’t want to change! You would only migrate to a perfect clone (and, guess what? even for a perfect clone, you would not do it because "the clone doesn’t add value)
Secondly: You are not even trying. When you try alternative such as Comaps, you discover that there are many many ways in which Google Maps sucks: ever tried planning bikepacking trips or walking treks with Google Maps? Ever used Google Maps in a forest? It sucks. Every solutions shines in a given context. Saying "but Google Maps is superior" is simply false. The fact that most people, including those in tech, cannot see it is a huge problem.
Thirdly: The whole point is to advocate for institutions to invest into Open Source. We don’t care that it is not perfect yet. Nobody cares that you still use Google Maps on a personal level. That’s not the point. The point is : how can we make alternatives that first allows to survive without monopolies. If those alternative get better than monopolies, well, that would be the cherry on the cake. But that’s not the goal!
Thank you for arguing clearly exactly what has been in my head for the longest time! You first and second points under "2." really drive home the underlying problems. Open source (or any European alternatives) are having to fight an uphill battle. The deck is stacked against them.
I also think it will stay that way, probably forever, because of how easy it is to build a product and become Really Big in the US: few pesky consumer laws to comply with, a single unified culture and language with many states to grow into and solidify the product, before expanding into the European market. And of course, a big pool of VC investor money to get yourself bootstrapped. All of these mean the competition from closed source US products is fierce. We cannot compete with them on sheer features, and where we can (as you point out with Comaps etc), we cannot compete on marketing. Because this also boils down to simple brand loyalty and awareness.
All of the above are also big reasons why I advocate for the EU to attack the problem on multiple fronts. Education on the impact of software choices being a very important pillar, and requiring by law that government institutions use open source software being another.
Otherwise, this pressure to stick with the status quo is too strong to overcome, even with the very best software alternatives. And even if the alternatives come, people will just jump ship the moment the next product with a "shiny new color" comes along.
impossible to have fine-grained traffic data like Google without having billions of spyware enabled smartphone on the road
One solution to this is to pass legislation that mandates Google/others share live traffic data via a public API at minimal or no cost. This could apply to other similarly useful kinds of data as well.
In Czechia, phone operators share anonymized phone locations with public transportation companies and urban planners so that those can figure out how to make the trips more efficient.
Firstly, it means you value "shiny coloured featured" more than political independence and ethical value. That’s a huge problem. So, basically, you don’t want to change! You would only migrate to a perfect clone (and, guess what? even for a perfect clone, you would not do it because "the clone doesn’t add value)
I think this is a very important part that I fail to completely reason about.
In any area where there is profit to be made, a commercial company will have a huge incentive to make a shiny and even truly very user-friendly product. They have money to spend wisely on a good user experience.
Most open source projects cannot compete- they simply won't have comparable resources.
Plus, they will invest in marketing.
I think you are right that many times, the open source will be superior in some senses. For example, just by virtue of not being oriented towards a profit, it can be better for the user.
But, esp. in software that benefits from network effects, you're not in a good position because simply most people do not care. And why would they? I only have bandwidth to care about a finite number of things- and the things I care about mostly gravitate around things I know. I do not expect most people to care about which instant messaging platform to use!
And honestly, for example, in instant messaging, the popular choice comes and goes- we've been in a long stretch of WhatsApp dominance in many places, but the leader has nearly always been a proprietary IM platform. So WhatsApp will die and be replaced with something else, people will move their chats to the new shiny, and move on. While we the nerds appreciate that it would have been great if XMPP would have won in the early 2000s and had evolved through these days, it's hard to articulate to people who do not care about instant messaging about what basically they'll be using for a number of years and then change...
Maybe the answer is governments. I agree with what someone else said in the thread- just trying to be independent of the US does not seem like a good motivator for me. I think we just need to realize that some services are mandatory for modern life, and large companies should not be able to control them.
I mean, we could insist on standards for mass email providers. Even limit their bundling options, move to a similar model we have for mobile operators... even advance the standard to enable address e.g. portability.
And IM providers could be forced to agree on a set of live standards they will follow to enable interoperability and again, account portability.
We managed with phone numbers, so...
It's a policy choice. Either we allow private roads that can only be used with road owner rental vehicles or we force everyone to be compatible.
The federated XMPP/Jabber network has not won in terms of market share, but it sure has evolved and is still doing so. I am taking the train tomorrow to go the XMPP Summit 26, (some of) my family members and friends have been using Conversations and Monal for several years, so I would know. ;-)
Yes, it's moving forward, but their objectives don't really align with mine. I would like to set fire to WhatsApp and use any IM system that allows for third-party clients without encumbrances.
I decided to play again with XMPP very recently to see where it was at, esp. because I had played with Delta Chat and I was very impressed with that.
So I went with Conversations, that has the nicest onboarding I've ever seen for an XMPP client. (Slightly inferior to Delta Chat, but good enough.) But nowadays, even non-technical people expect to be able to join their IM to their computer.
Which should precisely be the forte of XMPP!
However, in WhatsApp (and Delta Chat, Telegram, Signal...) the mobile app guides you through this process, but I found no guide on Conversations to do this. Which means, if I recommend XMPP/Conversations to people I know, they'll likely won't be able to figure this out on their own.
(I filed a bug. Someone said this was not something that Conversations should do. It's still not closed, but it's been idle for over a month...)
I use Gajim on my Linuxes, so does my wife on her Windows, I used to use Beagle on my mac work laptop, I am not sure what you mean? You have to type a username@server and password, and that's it. There are also web clients like movim or converse. Maybe you mean the "flash this QR on your phone to login" flow? I believe Fast Authentication Streamlining Tokens could be used as a building block for such feature, but this is really outside my "expertise".
nowadays, even non-technical people expect to be able to join their IM to their computer.
I was surprised to realize this, but among friends and family that have migrated -some- of their comms to XMPP, most do not care the slightest about using it outside their phone. I contributed some stuff to Gajim and would have been pleased to get some of my friends to use it, but the answer was 99% of the time: "oh no, I don't want messages on my computer, that's what the phone is for". The "nowadays" is what surprises me most about your sentence, I'm old enough to remember my non-technical friends using MSN Messenger on their 1024×768 CRT monitors, but "nowadays" they just want to use their phone.
in WhatsApp (and Delta Chat, Telegram, Signal...)
Delta Chat and Telegram have proper non-mobile clients indeed, but for Signal and WhatsApp, the mobile client is the "main device" and the desktop/web client is a second-class citizen. Close the app on your phone and it will eventually stop working until you start the app again (and in the case of WhatsApp, you'll have to re-auth, using your phone).
their objectives don't really align with mine
Unlike any other alternatives you mentioned, there is no such thing as "their" objectives. The Conversations maintainer has his own objectives, sure, but XMPP has a democratic process for the protocol development. You don't even have to be involved in this process to just start building stuff that uses the protocol. It is a common good. If you want to get involved, it can just be participating in mailing lists and/or chat rooms. Relevant recent thread in the mailing list: Participation of civil society in the standardization process. Come back and let's "set fire to WhatsApp" together. :-)
I mean any process where someone follows the onboarding in Conversations, then taps on "add a device", and they get a client on a different device. I don't care the process as long as it's straightforward and it offers a "golden path" that works.
Then I can recommend people "get Conversations and follow the instructions to do a second device".
I'm all for there being choice, for a XEP standardizing the QRs, but there must be a clear simple process for people who want to do things without having to do research.
You might be right in that many, perhaps most, just care about having their IM on their phone, but I meet enough people who use a second device that for me it's one of the clearest obstacles for me to recommend XMPP. (In contrast, Delta Chat works and provides everything I need to recommend it.)
I know it's open source, and I like the idea of XMPP. But I need a mobile client to recommend that has a good onboarding that doesn't require research, including a second device. Conversations seems to be the closest (although the costs on the Play Store are a turn-off for me, although I understand it's likely needed), so that's why I filed the bug- because the more decent options available, the better. At the moment, Delta Chat ticks my boxes better- I hope some XMPP combo does too at some point.
Maybe we can learn from XMPP not winning the IM world at the time. And from email being more and more centralized. And for the web itself (which is now gatekeeped by two browsers).
I think there's another lesson to be learned. Beyond user experience and marketing, private companies trying to dominate a market will collaborate with existing open standards etc. exactly as long as it benefits them. But not a second longer.
So yes, Meta's Threads is supposed to federate with ActivityPub. No one should base any decision on that, because if (when) Threads achieves sufficient marketshare, ActivityPub will disappear.
I use ActivityPub and I refuse to engage with BlueSky or Threads, because it's a trap.
However, a lot of people on ActivityPub- even those who are escaping from proprietary networks and aware of their tactics, will still engage and feed BlueSky and Threads.
I don't disagree with you here, but I wonder if it answers the right question?
The question is not if the EU should invest more into political independence or its values. The question is how. Why should the EU rather invest into OSS than into proprietary EU companies?
Also, I would assume they hope for details. Like, should the EU invest more into the OSS social media or OSS operating systems or OSS smartphones or ...?
The most important question though, since this is a "call for evidence": What is the evidence that such investments will pay off (quickly and bigly)?
Your article is a well-written call for action. When I look at this call for evidence, specifically the section "Consultation strategy", I don't think it addresses their questions.
Coming back to the example of a "map app": It is probably not hard to find evidence how OSM has made mapping data cheaper and easily available without downsides if you get the data from Google. Would investing a few more millions there be a good investment? How would it pay off? Probably not in the navigation business because "it is impossible to have fine-grained traffic data like Google without having billions of spyware enabled smartphone on the road".
Maybe there is something else related to OSM? The data is already great and while maintenance is important, what could there be as a next big step? Thousands of municipalities in the EU deal with mapping data. Maybe OSM could help there and save billions of tax money within ten years? I don't know and neither does the EU commission.
OSM has severe limitations for building a modern mapping data model, ignoring the well-known data license and graffiti issues for the moment.
It is impossible to use human crowdsourcing to detect and process the number of changes per day that actually occur in cutting edge mapping data models. Even with fully automated pipelines processing petabytes of real-time telemetry per day, it is generally recognized that there are still large gaps in accuracy and freshness that limit usability for many applications. The large closed source mapping data models only use the OSM data model for parallel construction, cueing their systems to look for potential changes they can find in other data sources.
This is the crux of the matter. Maps are no longer slowly changing display artifacts created by humans. Today, they are real-time petabyte-to-exabyte scale data models that are being continuously updated from data flowing in at rates most software devs can barely imagine. The data informing the data model is often noisy, erroneous, corrupted, or manipulated so you need sophisticated and compute-intensive inference pipelines to clean this up. Put this way, it is obvious why Google is so much better at this than the traditional mapping companies.
There is a large opportunity here. Industry has an enormous desire for a competitive non-Google map infrastructure, so if the EU were to build one it would be immediately relevant globally. The large caveat is that it is impossible to build a state-of-the-art map infrastructure on current OSS. Creating it is a commitment to building a completely bespoke data infrastructure with a lot of computer science that is not in the literature. The reason this doesn't already exist isn't because many companies have not tried to build it, it is because it requires extreme technical sophistication to have a prayer of implementing it successfully.
The barrier to an MVP in mapping is very, very high and there is no way to get around that fact. In principle the EU could acquire the necessary data sources to build these data models, that isn't the real limitation.
It is impossible to use human crowdsourcing to detect and process the number of changes per day that actually occur in cutting edge mapping data models
My house showed up on OSM as under construction when the work started. When I moved in, OSM showed the building and had the unfinished half of the estate marked as under construction. It took Google Maps over a year to notice that the estate existed.
Some roads around here became one-way streets a while ago. Again, OSM had up-to-date data within a week. It took Google Maps and the major SatNav systems over a year to stop sending people down the roads.
One of the roads nearby was being used as a cut through so they put a barrier in the middle and turned it into two dead ends. OSM had that marked before the construction work finished. I helped some folks who were struggling to turn around towing a caravan a few months later. Why had they gone down a dead-end street? Because Google Maps had told them to.
So, while I agree that updating maps is hard, I have seen plenty of evidence that OSM is better at it than proprietary map data sets.
The quality and freshness of all base maps vary widely based on where you are. OSM is good in some areas but is much more uneven globally than the closed source base maps.
There are only about half a dozen global base maps, of which OSM is one. Every company that builds a base map regularly compares their coverage and quality against their competitors. All of the base maps, including Google, have regions where they perform well and areas where they perform poorly.
No current base map is a clear winner globally. Beyond requiring more consistent access to data globally, it requires technical infrastructure at a scale that OSM and most closed source base maps will never be able to build.
I guess it's locality-dependent. Waze (which I assume shares data with Google Maps) picked up newly constructed roundabouts around here 6 months or so before they were added to Open Street Map.
Ultimately someone has to sit down and edit OSM, and if the people who know how to edit it aren't evenly distributed then some places will lag behind.
Ultimately someone has to sit down and edit OSM, and if the people who know how to edit it aren't evenly distributed then some places will lag behind.
The same thing is true for Google Maps, and it's much harder to scale a thing where a single company has to maintain the data thna it is a distributed environment.
In Cambridge, the University of Cambridge used to maintain their own maps. These were very detailed for university / college sites, but also needed to be kept up to date for surrounding roads because they were used to show people how to get to various university places. Once OSM reached a certain level of quality, they realised it was cheaper to contribute their map data to OSM than maintain their own, so they wrote an OSM renderer that matched their old style and switched over. They still have people updating the areas in and around university sites, but now other people mostly handle the rest of the city. Everyone wins. With Google, you can put in requests for them to add / fix things in the maps, but they can't easily do the things like footpaths inside colleges because they're private property. And you can't then use their data for all purposes, you need to have a legal agreement to use them for any commercial purpose and using them also comes with a load of GDPR obligations.
It's quite possible that the mapping of hiking and biking trails is better in OSM than Google for where I live. I think there's a certain disunion between people who commute regularly by car - thus providing input to Waze/Google - and those that hike and bike who might be more familiar with navigation aids that rely more on OSM.
Possibly. But when travelling I've found that there's a very sharp drop off in Google Maps quality when you get out of populous areas. My mother lives in rural France and Google maps doesn't have half the roads near here, whereas OSM has them with metadata about the kind of road, and also has a load of the walking paths and most of the buildings marked. In the middle of Paris, Google Maps data is quite good.
It is impossible to use human crowdsourcing to detect and process the number of changes per day that actually occur in cutting edge mapping data models
I'm ready to believe that, but the fact is OSM is very often more right than Google Maps here in Belgium, a country that should have more than enough data available to display a 120 km/h speed limit on highways instead of the random 90, 110 and even 130 (that exists nowhere in Belgium) limits that Google Maps often shows.
After years of taking my own street the right way, Google Maps still doesn't know that that street goes that way. And I do use Google Maps, and so do probably a sizeable amount of the cars that travel through that street every day. It's not a low-traffic road.
It's probably different in other places. But in Belgium and France at least, when I want to be sure the data on my map is correct, I use OpenStreetMap. The rest of the time I use Google Maps for traffic information.
Maps are no longer slowly changing display artifacts created by humans.
This is true, and true of more than just maps.
I've worked at a few health care companies, for example, and payment in that world is generally based on a combination of factors like name, billing address, license ID number, etc. But when I last worked in the industry, the measured half-life of that data was 18 months.
Meaning, if you compiled a list of all the doctors, hospitals, etc. you have a payment relationship with, and their names, addresses, license numbers, etc., then 18 months later half of them would have changed in ways that invalidate your data set. Which is a huge source of unnecessary friction and anger, since to the patient "it's just Dr. Smith, the same Dr. Smith I've been going to for years!" but in that time Dr. Smith has, for payment purposes, operated under four or five different names, used half a dozen different billing addresses... and so, no, the payment systems can't just automatically verify that "it's still Dr. Smith".
The reason this doesn't already exist isn't because many companies have not tried to build it, it is because it requires extreme technical sophistication to have a prayer of implementing it successfully.
I disagree. The reason it hasn't been built is because there is no straightforward way to pay for it.
This is the kind of unprofitable baseline infrastructural thing that is absolutely the bailiwick of government. Sure, the government won't do a great job, but it can provide a publicly accessible baseline that decreases the entry cost of other people building on it.
To be competitive, the base map needs to be consistent and global. Many governments, including those in Europe, actively resist the notion of giving their mapping data to an organization not based in their country. To the extent they are willing to pool their data in a common system, most want absolute veto power over any mapping edits that concerns their country in that system. The intransigence of governments on these issues has been a major roadblock despite many efforts to build a common base map.
It would be great to have a global base map operated as a public utility. Most companies and people would prefer that. The base maps that we have now are a reaction to governments being unwilling to participate in a public map, but most heavy users of them recognize that they are not really fit for modern use cases. Perhaps ironically, the US government is by far one of the friendliest to this idea and always has been.
The direction the industry has been headed is the creation of a new commercial mapping entity that is deemed sufficiently neutral by design that all the major tech companies (except Google) are okay with them controlling the One True Map. There are plenty of ways for this to be paid for, that isn't a material issue.
As someone from the the US, the idea that the government would try to hoard and control map data is a bit weird to me since the USGS (and most government entities) has had that kind of thing available for general purchase for decades.
One difference is that in most countries the mapping data models were originally created for national defense. Those organizations often still view that data through a national security lens, or at the very least, control of that data for national security purposes is an input to the process.
In the US, these concerns have long been separated. The USGS is a purely civil organization, essentially a public service. For national defense, there is a large and entirely separate US government organization (NGA) that is responsible for creating mapping data models in the national security context.
I travel a lot in the UK and Europe and I use Organic Maps (OSM), google maps and CityMapper. OSM has always been at least as good for road maps and always better for walking and pedestrian paths (often much much better).
I use CityMapper and Google maps for turn-by-turn directions, and google maps for finding restaurants (OSM has pretty out of date info on restaurants and businesses, in my experience)
My article is a translation from an old article and I agree that it doesn’t really answer the call for evidence.
And the main reason is that I believe the very first thing we need to do is to shift our priorities. To change our very definition of success. As long as we play by US-monopolies rules, we will not be able to make anything but sub-par cheap clones. It simply cannot work that way. Even if it start to work, it will be easily subverted (see my history of the Nokia Meego debacle)
Well, I still would like to see a navigation app based on OpenStreetMap, which is as good as Google Maps
I use Organic Maps on my phone and, unlike Google Maps it:
Knows about cycle paths and doesn't send me on long detours because the Google mapping cars can't see a small path that lets my skip a big chunk of the trip.
Doesn't send me the wrong way down one-way streets.
It doesn't have live traffic data, but it also isn't connected to a massive surveillance system that it uses to build the traffic data so I consider that a win overall.
mapy.cz (now, mapy.com) is one of these I believe.
They are in extremely common use in Czechia, much more common than Google because Google Maps seems to have a difficult time mapping and pathfinding in Czechia.
I was recommended this because I wanted to cycle to work.
(I also learned that “cycling to work” means totally different things to Scandinavians and Czech people).
Oh, I live in Malmö- people in the first picture basically don’t exist.
To be fair with you, the few dozen times I have been to Stockholm it didn’t strike me as a very friendly place to cycle, and I didn’t see many cyclists. But the ones I did see didn’t see to stand out as being especially spandex laden with clip-ins.
I don’t think I even saw a helmet, but; your experience likely trumps mine as I’ve only lived in Sweden for 11 years and most of that was spent in Malmö.
I will not let this random dunk and affront to my city of residence to stand unchallenged. GBG cyclist here, both for leisure and work commuting.
The road traffic is kind of random with lane changes and trams and all, but the proportion of a'holes and psychos at the wheel is not higher than elsewhere.
I figured there would be even more salt-of-the-earth Volvo car builders and owners there, than the Beemer and Merc drivers up here. Not saying they're any less assholish about cyclists, though.
I would never bike without a helmet. I know of too many people who have been seriously injured or killed when biking.
That begs the question of whether or not helmets significantly reduce serious injuries and death on net. Obviously, all other things being equal it’s not worse to have a helmet on than to be bareheaded when struck on the head. But the helmet can not be effective at all in certain head strikes (e.g. when struck by a bullet or fast vehicle). And of course it is not effective if not hit on the head: helmets do nothing against hand, wrist and forearm injuries.
Helmets are not costless, either. Besides the actual cost, they make cycling less spontaneous and more inconvenient (one has to lug the helmet around after parking the bike). The number-one threat to cyclists is not themselves but drivers: studies have shown that cyclist safety increases with the number of cyclists on the road: by decreasing the number of cyclists, helmets reduce safety. Worse, other studies have shown that drives drive more closely to helmeted cyclists, increasing the risk of collisions. So helmets lead to drivers less familiar with cyclists, and more prone to driving dangerously close.
I believe it is true that helmets on net are bad for individuals, bad for cycling and bad for society. I don’t believe that they should be banned, but I definitely don’t believe that they should be encouraged or mandated.
I think using a helmet is a deterrent to cycling for a bunch of reasons:
Makes it seem less safe
They're uncomfortable
They mess up your hair (a gendered issue that men often don't think about)
They're a big hassle if you want to use hire bikes
They're something else to take with you or lock up when you park
Personally, I cycle without a helmet unless I am going on a long cycling holiday, then I sometimes wear one because I am more likely to be cycling in poor conditions and on steep roads.
I think data on hire-bike users pretty clearly suggests that helmets are not necessary for cycling in cities.
If you want more rationale for not mandating helmets, I recommend Ruwen Ogien's works, a French philosopher who strongly advocated against any laws meant to prevent self-inflicted harm or risks (while at the same time absolutely not being a "libertarian" in the US sense). I don't know how many of his books have been translated in English though.
Back to the subject, I do think it's more reasonable to wear a helmet, but I generally agree with Ogien's "minimal ethics" and don't think mandating them would count as "progress".
Yeah that second picture is from the Netherlands (it even says NL on it and in the URL). Perfectly normal here, even though there's heavy campaigning from the government to try and change things...
That first pic is from London, and spandex wearing commuters are fairly rare - usually going long distances. Most other people cycle commuting are wearing a helmet and some outside clothes. Maybe clips and cycling gloves.
Appreciate the pictures but how did you mix up the countries so badly? LOL. The first one is from London. The second is from NL. But thanks for the pics.
You should probably read the bit of my comment where I explicitly mention that these are london and netherlands.
The point was to illustrate the difference, not to specifically show a Czech cyclist (they are somewhat rare) and a Copenhagen/Malmö cyclist (common, but they just look the same as the dutch photo and sourcing photos is more difficult).
Munich has both, I'd say - although pic1 would more be "full bike touring kit (rain gear or leggings+gloves+rainjacket/jersey) on a touring bike with panniers" instead of spandex. And I'm kinda equating p1 because it's specialized cycling kit, just not the performance aspect :P
(Me, I'm probably in between. If the weather is bad I'll use different pants than I'm wearing in the office and because I own rain gear and panniers, I'll use that if needed - but you'd 90% see me commuting in normal clothes and a backpack on a reasonable bike with fenders)
Not only for cycling, but also for hiking. I have never been disappointed by a hiking route suggested by Mapy.cz. It's much better than Komoot because mapy.cz can suggest a hiking route based on your current position and the distance you want to hike. However, the app tends to crash frequently and some time (years?) ago they started advertising premium features.
Have you tried OsmAnd (developed by OsmAnd BV in the Netherlands, fully open source)? I have been using it for years, and they also have a very nice funding model: If you download from F-Droid, you get all the features, but if you download from Google Play or the iOS App Store, some non-essential functionalities are in-app-purchases, e.g. unlimited offline map packs.
Yes, you can download the entire world as an offline map, if you wanted. Device storage has become so large, there is little reason not to download your entire home country or state. More than once have I found myself in the forest with no cell reception with friends, who then struggled to navigate on their always-online apps. In contrast, OsmAnd just worked for me. It has dozens of modes (walking, bike, car, airplane, etc.).
Many weaknesses have been resolved in recent years, including sometimes choppy map rendering and slow route calculation. Given the excellent OSM data, you get lane-level navigation hints, which Apple Maps and Google Maps often get wrong.
I found OsmAnd to be super clunky on iOS, and I didn't love the style in which the maps were rendered. I've heard it's a better experience on Android. If you say the choppy rendering has improved, maybe I'll try it again.
I personally like Organic Maps – also open source, Estonian. Also has offline maps.
Haven't used navigation very much in either of them, TBH. When on foot, I usually need public transit timetable, and when driving, I prefer a dedicated device over a phone app (and want some live traffic info). When my TomTom broke I tried a bunch of apps, it seemed like none of them was designed by somebody who ever drove a car, Google Maps was the absolute worst. I could go on a rant here.
Also, navigation apps could be so much better, I'd love to have an open source, hackable thing on my dashboard. The bottleneck for really nice features seems to be data access (live traffic info, live public transit status; I've never seen it, but I'd love a map/navi that would utilize live weather info ("Heads up: you're about to drive into a thunderstorm" type alerts or a wind indicator on a navi screen would be great))
I did a slow blink at this - for context, maps.me spawned Organic Maps, which spawned comaps. All use data from OpenStreetMap and can show maps offline.
Really good points, thanks! Yes, weather info would be great. Public transit is in beta at OsmAnd and I tested it successfully in Prague a few weeks ago. Apart from that, I really like Öffi, but they are fighting an uphill battle as more and more public transport companies close their APIs and force their shoddy apps on everybody.
I tried OsmAnd a few years ago and it was complete garbage. Happy to report it is much faster now and has gained many useful features.
That said, it's a loooong way off from the usability and performance of Google Maps. This is not something regular consumers are going to consider using. Which is a shame, because OSM data is really good.
GMaps does have an offline mode :)
Of course you need to download the map before.
What worries me more is that you don't get GPS coordinates from GMaps if you
didn't download the area you are in.
Granted, but the offline mode in Google Maps is really primitive in comparison. I can download a complete map of Germany or even Europe and it only takes a few GBs of space, which is nothing in many of today's smartphones. And no matter where you place me, it would just be there. There is also a world overview map with coarse roads and large cities.
I used it years ago. It is a nice app to work with OSM data.
It seems it does not show/use traffic information. Beyond that, Google Maps offers all the additional information about shops and restaurants (opening times, menus, etc).
You can download Google live traffic data as an overlay, but it's not (yet) included in the routing. It may sound like a cope, but it gives me an incentive not to turn my brain off when navigated.
Regarding opening hours, Google of course is king, but at least here in Germany and larger european cities, this data is pretty up to date, too, though nothing beats businesses proactively updating their times for Google.
You won’t have that because Google has put the data sharing app in all android phones and AFAIK you can’t opt out.
It’s part of the reason Apple and Google are able to do live traffic reporting. Android phones in cars do data reporting, without the consent of the driver and on aggregate is the most accurate “live view” system we have.
I don't understand this at all. Even as someone who doesn't use AI at all and isn't directly impacted by it, I feel like we should burn the entire "AI industry" to the ground because of all the negative externalities it has. And here's someone who's business is directly being destroyed because of AI and is "optimistic" about it?!
we should burn the entire "AI industry" to the ground because of all the negative externalities it has. And here's someone who's business is directly being destroyed because of AI and is "optimistic" about it?!
If you are a saddle maker that makes saddles for horses and your business goes away, you can still be excited about cars and adjust to new realities. Those things are not necessarily in conflict.
The major difference is that tailwind is far from obsolete unlike horses in your simile. If tailwind development goes away it hurts all users. AI is not a replacement for tailwind.
Tailwind is not, but Tailwind was not commercialized. Tailwind UI was the cash cow and that is obsolete now that you can have an agent build your own components quickly. The entire business model in this form broke away.
To be fair it was a business model that shouldn't exist in the first place, AI just made it more obvious. I place it on the same shelf as other types of jobs that only exist because of some external deficiency not related to the technology they serve, like train fare collectors and pedestrian crossing guards. It's the human element that is lacking in all three of these, either human skill issues for CSS, human dishonesty for trains or human inattention for roads. Once that is solved by some method the need for those business models disappear.
I feel like train fare collectors and crossing guards serve more purpose than just compensating for human dishonesty or inattention.
A fare collector can also help people who have some problem with a ticket. A crossing guard can help people across the street, regardless of how reckless or careful drivers may be.
I also just ... cringe at the idea that everything should just be ruthlessly automated and efficient and there is no place for humans to do jobs that may not be "essential" for some value of "essential" but where a human touch is a good thing. I insist, for example, on using cashiers rather than self checkout.
I'm old. I remember when our town still had a gas station that had an attendant. Yeah, I can pump my own gas. But it sure was nice for my mom when she was wrangling toddlers, or no doubt for older people, etc. Absolutely necessary? No. A fulfilling job? I dunno - ISTM some folks enjoyed the work. I'm not sure we're better off for the automation.
I also just ... cringe at the idea that everything should just be ruthlessly automated and efficient and there is no place for humans to do jobs that may not be "essential" for some value of "essential" but where a human touch is a good thing. I insist, for example, on using cashiers rather than self checkout.
If the human adds value, the human should have a place. But there are plenty of cases where humans add negative value compared to the automation. The self checkout? Depends on the situation, sometimes the human is helpful, sometimes it is not or negative. Ordering at McDonalds? Quite frankly the touch screen version is such an improvement over the human version because the screen is patient and the human is not.
I think there is so much more nuance between machine vs human and there is no one opinion fits all situations here.
If the human adds value, the human should have a place. But there are plenty of cases where humans add negative value compared to the automation. The self checkout? Depends on the situation, sometimes the human is helpful, sometimes it is not or negative.
I think "adds value" is too vague here. Adding value in a job can mean:
The labor provided by the human generates even greater profit for the business so them having the job adds value to the company.
Someone has access to a job that pays them where they might otherwise struggle to find work.
Customers or other people who interact with the employee have an enriched better experience because there is a human doing that job.
2 and 3 are really important for people and society, but businesses choosing to replace a job with AI only give a shit about 1.
I think "adds value" is too vague here. Adding value in a job can mean
I'm only talking about value added in the consumer interaction. As a consumer, I don't care whether the company makes money. What matters is whether an interaction with a human, compared to an interaction with a machine, adds value for me in that moment. And in some cases, the human adds negative value.
But we also shouldn't be surprised when businesses don't focus on that because they have no incentive to and if they do, they can be outcompeted by other businesses who don't.
I'll admit, I prefer the touchscreen ordering. But I'm not sure it's an improvement for everybody. It's certainly not an improvement for people with visual disabilities (at least my recollection of the touchscreen stuff is that it's not a11y-friendly, I could be wrong here).
"I think there is so much more nuance between machine vs human and there is no one opinion fits all situations here."
Well, we agree on that for sure. I'm taking the position that, because people must have jobs to exist in society as it stands right now, that's priority one. Nobody who is willing to work a full-time job should be unable to find work that pays well enough for the person to sustain themselves and (if necessary) their dependents at a non-poverty level. That's my starting position.
at least my recollection of the touchscreen stuff is that it's not a11y-friendly, I could be wrong here
I don't have disabilities so I cannot judge from first hand, but I know that it has a huge accessibility button that I pressed out of interest and it offers a zoom mode and a reachability mode. The latter I actually use for my kids sometimes.
Nobody who is willing to work a full-time job should be unable to find work that pays well enough for the person to sustain themselves and (if necessary) their dependents at a non-poverty level. That's my starting position.
I agree with this on the surface, at least in theory. But it can’t hold for 100% of people. There will always be individuals who are willing to work but whose skills, or what they can offer, don’t find a buyer on the other side. It’s still a market, and no one can be forced to employ someone. As a result, there will always be a segment for whom finding work is hard. In periods of uncertainty like the one we're in now with AI changing things, that segment naturally grows. I'm sure that over time, things will rebalance. IMO the idea that no job can ever be lost doesn't seem right and it might not even have to be the highest priority.
I started writing this today right after leaving from a Taco Bell where I had just witnessed a fight nearly break out with a customer and some staff because of how behind they were.
This was with:
Multiple ordering kiosks
DoorDash, etc. orders coming in
Backed up drive-thru
Visually, to me, fully-staffed kitchen cranking orders out as fast as they could
The problem I think isn't automation really, it's that the wrong things are being automated.
It seems like the fast food industry is only optimizing for the Producer (placing orders) and never touching the Consumer (making orders) part of the fast food industry's version of the Producer-Consumer problem.
There is literally nothing managing back pressure and it is super obvious. "fast food" in 2026 is only fast in the sense that large amounts of orders get to the kitchen very quickly. If you replaced fast food workers right now, it still wouldn't solve anything unless you could plop a giant food factory on top of an existing fast food place or something.
It's part of the reason why dine-in places in 2026 are so much better speed wise. The back pressure management is built into the experience, lol.
If you take out the kiosks and the DoorDash, much to many people's dismay, we'd probably see a vast improvement in how quickly folks can deliver product.
I noticed this at Starbucks in San Francisco. There is no visual clue of what is going on because of mobile orders. I ordered in store a cappuccino and i had to wait for 30 minutes because of how many people were apparently ahead of me.
I'm not sure though if that is a problem at McDonalds. I did not notice it there much, in part because they also make it very clear how long the queue in front of you is visually on the screen. So if there is a lot happening, you can bounce as a customer.
I think there is so much more nuance between machine vs human and there is no one opinion fits all situations here.
Hard agree and I would love to see more (friendly) discussion about this!
But in my experience the current wave that people around me are riding is productivity/result over empathy without much nuance. Buying art generated by a model instead of from people is one of the first examples that come to mind.
because the screen is patient and the human is not.
From a consumer perspective I can understand this, but depending on where you are, instead of paying/improving quality of life for people in service industry...they got replaced with screens? I know that's not what you're saying, but I'm having a hard time understanding the goal here with "humans should add value" w.r.t. McDonalds.
I don't think I'm advocating for automation per se, or even efficiency. Just pointing out that that some jobs are inherently unproductive and merely exist because some other element of society is broken.
In Brazil we have a bus ticket "cashier", sort of an overpowered fare collector. You enter the bus and they sit on a special seat with a turnstile, you pay them and go through the turnstile. If you don't pay they're there to physically stop you. They exist because without them a lot of people would not pay for the bus.
That this job even exists is a deficiency in the Brazilian society both in terms of income inequality as well as dishonesty. In all countries in Europe, Asia or Oceania I've been they don't exist: the driver does that job. In Luxembourg the bus is free so there's not even that issue.
This is an example of a "useless" job. I don't advocate for their disappearance because here at least it'd make a lot of people unemployed but it's a very bad usage of a human brain. That person could be doing better things with their time if society was less broken in that particular way.
"That person could be doing better things with their time if society was less broken in that particular way."
Could they? This assumes there's an available job for them to transition to -- and I'm not convinced that's the case. I don't know how public transit works in Brazil, I'll admit, but if the entity that allocates funds for work eliminates that job -- is it going to create a job elsewhere that the people can move to?
I see a common thread in tech circles where people will voice opinions along the line of "just get a better job" or "that job shouldn't exist because it's a dead-end job", etc. Basically, there seems to be some underlying assumptions that 1) there's an endless supply of jobs, and 2) everybody should always be "upskilling" and trying to hustle into more challenging work, etc. But a lot of people really only care about jobs because they must have one to exist. They don't want to be constantly retraining for new things. They just want to pay their bills and be left alone.
So endlessly automating things so that these "useless" jobs don't exist is, IMO, not a net good for society.
Now - if a job/need exists and you can't find people willing to do it, that's a good sign that it should be automated, eliminated, or otherwise changed so that people are willing to do it. In my previous example of gas station attendants -- if that job was eliminated because nobody wanted to do the work, then that seems like a net benefit for society.
I don't think I'm advocating for automation per se, or even efficiency. Just pointing out that that some jobs are inherently unproductive and merely exist because some other element of society is broken.
I don't really understand how, by this logic, some jobs are productive, and some are not. All jobs are there to "fix" deficiencies with the way things are. Some jobs mold nature into different shapes. Some jobs attempt to fix broken bodies. Some jobs correct for human nature.
Once that is solved by some method the need for those business models disappear.
Telerik has the same business model Tailwind does, and has done so for decades. The business strategy is valid because customers don't want to hire someone to do the work themselves and would rather rely on a consultancy. Are you suggesting consultancies are invalid businesses?
Tailwind UI Plus provides ready-made components and templates, it's not a consultancy.
By the way "invalid" is your own take on this. I didn't suggest they are invalid business models, just unnecessary ones that a slight change in external factors would obviate. They provide a value to current society due to these human inefficiencies but their value is not intrinsic.
I didn't suggest they are invalid business models, just unnecessary ones
I think you're just arguing semantics when your phrasing was "shouldn't exist in the first place", i.e., not valid.
provides ready-made components and templates
Sure, but those components have something that makes them special over the OSS offering that people wanted to pay for and were asked for by some customer in the first place.
My point is the "custom widget" market exists, has existed for a long time, and AI killing this business model and being built on stolen code kind of makes this another drop in the bucket against its use.
Sure, but those components have something that makes them special over the OSS offering that people wanted to pay for and were asked for by some customer in the first place.
I'm no expert in this business and I don't know what their consumer base looks like, but I would not be terribly surprised if the purchasers of those components were a very specific crowd. The adoption of tailwind-ui within my circles has been exactly zero. Yet they make a lot of money, so they must have some pretty good usage numbers but they are probably from some specific crowds that were looking for that precise solution.
Among my circles on the other hand shadcn took off quickly and can be found in many applications now. Obviously shadcn is also free, but the value proposition also seems better for react developers than tailwind UI was.
So "people wanted to pay for" is valid, but "some people wanted to pay for" and those people are now increasingly looking elsewhere.
those people are now increasingly looking elsewhere.
I don't think that's the takeaway here. The knowledge of the offering's existence has basically fallen off entirely because the AI isn't showing it to them in the first place unlike if a human was looking at the docs instead.
The knowledge of the offering's existence has basically fallen off entirely because the AI isn't showing it to them in the first place
That is an assumption which might or might not be true. My suspicion is that it's wrong. The AI is building people entire components so the need does not exist. If i ask my coding agent where to get high quality tailwind components the first link it gives me is tailwind UI. If someone were to still ask that question, they would be pointed right at tailwind. But the work the agent does even without having access to it, is sufficient and might even be better.
The traffic to the docs is down is a fact. That someone were to reach for tailwind UI if only the agent would advertise tailwind UI is the assumption that is not validated. If that was the case, I'm sure they could make the experiment with amp which offer ads in the free coding agent and see if that helps conversions.
Doom on a pregnancy test does not belong there, because Doom never ran on a pregnancy test. They crushed Doom gameplay video down to low resolution for an OLED display they mounted inside a pregnancy test, wired to a raspi or whatever off-screen. It's just a stupid hoax.
It was a fun thing to do, Foone explained what they'd done, it was not meant to fool anyone into thinking that doom ran on the pregnancy test hardware itself.
to be honest, I'm not certain - my doctor thinks it's an immune condition of some kind. I'm allergic to most American staples - eggs, dairy, most forms of gluten, etc. I'll have to give old speckled hen a try!!
I suspect you've probably stumbled on Burning Brothers, and Sociable Ciderwerks; for anyone who hasn't, there are actually good GF beer/beer-adjacent options these days! ☺
I'm a little sad that CPU and RAM are soldered on. They claim they were unable to find a modular solution without sacrificing performance. Interested into seeing some reviews soon.
I'll always choose the bandwidth and performance over socketed components. Truth is, many people want socketed components, but then never do anything with it. In fact, the vast majority of PC owners follow this. It's a small portion of enthusiasts who might upgrade.
If you think about it a bit, it really doesn't make as much sense as it could seem to make at first. Who, exactly, upgrades RAM? Broadly, I know two categories of people: laptop owners and people who lack enough funds to build a workstation to their liking at once (so they end up buying less RAM than they wanted, with the intent to buy the rest in future).
Upgrading RAM in a laptop is somewhat of a self-inflicted pain: many laptops are sold with woefully inadequate RAM, and sticking more RAM into an already owned laptop is one of the very few ways to delay the obsolescence of a given hardware platform. This is not a problem here.
Framework is not a cheap vendor, not even a cost-effective one. This automatically self-selects for people who do have enough funds to purchase a workstation to their preferred spec. So, again, not a problem here.
I had ram failure that would cause weird errors on my T480s. Luckily it was the external ram that failed and not the soldered one or else my laptop wouldve been e-waste. Now 5 years in, it’s still running extremely smoothly.
I have just bought more ram for my work machine which is about 1 year old. I switched from python to android development which is much more memory hungry due to emulators and large IDEs. So I'm upgrading from 32 to 64 gigs.
I have also upgraded the CPU once in the past. AMD had very good compatibility on am4. I only needed to upgrade the bios.
RAM requirements and expectations have historically increased very rapidly. It wasn’t too long ago that 16 GB was considered questionably excessive, and is now considered the minimum to run modern games (or run several Electron apps simultaneously). A basic developer workstation probably starts at 32 GB these days, with option for expansion to 256 GB or more. I would not have much use for 256 GB of RAM today but in 5-10 years I probably will.
This Lobsters cohort is going to naturally be an exception to this.
But if we take a step back and think about the average consumer, it’s just not happening. I’d say >99% of MacBooks, for example, are never upgraded.
And of course Strix Halo is in the early adopter part of the adoption curve now. So I recognize the mismatch between the current consumer, who are technical enthusiasts. But as it commoditizes, that’ll shift.
For now, I as an early adopter/enthusiast, definitely want performance over upgrade ability. But I acknowledge others may not see it the same way.
Yeah. This is due to technical constrains in the specifics of the AMD CPU/GPU/chipset.
In the LTT video[1] someone from the company said they "Asked AMD" about replaceable memory and AMD came back and said due to "signal integrity [...] over the 256-bit bus" it isn't possible.
(Check out the clip. Starts at 7:25).
Some way to expand the system with something faster than an normal SSD? Hell, I suppose slower DRAM to act as a ram disk would be good enough, if the chip set could drive both. PCIe connected ram disk maybe?
Some way to expand the system with something faster than an normal SSD? Hell, I suppose slower DRAM to act as a ram disk would be good enough
You're missing the point entirely.
The problem that's being solved by Framework is not "expanding the system with something faster than an SSD". It's "expanding the system with something faster than DDR5".
My experience with trying to keep machines from become obsolete suggests that the ability to add any extra RAM - or something nearly equivalent - is perhaps more important than its speed. Admittedly though I am an old git and maybe my use cases are out of date.
Personally a machine with soldered down very fast RAM with an option to add a "faster than SSD" ram disk for use as swap later on seems like it would remain useful for longer than a machine without that option. But my machines tend to get used as conventional servers or build farms, maybe other uses cases that I'm not thinking about would not benefit as much.
The entire subthread begins with your assertion that something faster than SSD (but slower than DRAM) "would be the answer here", where "here" is a demand for a type of extremely fast DRAM that uses a 256-bit wide bus (i.e., several times more than regular DDRx SDRAM).
If I'm "missing your point", then it means you have changed your point mid-conversation, which is not a valid way of debating anything.
We are going to be increasingly inclined to not accept new proposed features in the library core.
Alex/Carson -- you've been implicitly pushing the idea that "complete" is a worthy goal, and this finally makes it explicit. Software driven by a cohesive philosophy always shines, and htmx is no exception. I'm very appreciative of what you are doing, along with the rest of the htmx contributors.
I don't know if it was ever made formal policy, but I seem to remember that at one point a lead maintainer of RSpec opined that version 3 was basically the final form of the project. Upgrades have been pleasantly unexciting for about a decade now.
It's not that they never add features. It's that, at least since sqlite 3 (so for the past 20 years):
I never need to upgrade sqlite unless I'm feeling pain from a specific bug or want a specific new feature.
Code that I wrote against sqlite in 2004/2005 works just as well against sqlite now as it did against sqlite then.
I have no worries about reading a database now that I wrote in 2005. And I'm confident that I could write against 2025 sqlite and, if I'm careful about what features I use, read it with code that I wrote in 2005.
I think sqlite is very much an example of software driven by the cohesive philosophy that @jjude asked about. As @adriano said, it's not necessarily feature complete, but features are added very carefully and deliberately. There aren't many things I'm as confident using as I am in sqlite. It makes me happy that htmx (another thing I like a lot) aspires to that, but it's got to keep going a long time to prove it's in the same league. (I suspect it will.)
I'm not sure whether sqlite has a cohesive philosophy, but note that @jjude's question, as I understand it, is about software with a cohesive philosophy; not necessarily software with feature completeness as its philosophy.
If I were to guess what the sqlite authors' philosophy might be, it's that the world needs a high quality SQL implementation that remains in the public domain.
Thus highlighting the issues with "feature-complete for stability's sake".
Things change. It is the one constant. If the software is static in a sense, then rolling a new major version or forking with this kind of "fix" is both reasonable and necessary for the long term needs.
I also have rosy memories of the last good Python in my mind (2.5), but realistically, it was always accreting features at pretty fast rate. It's just a lot of people got stuck on 2.7 for a long time and didn't observe it.
You have to go back further than that IMO. Pre-2.0 was when you could argue that Python was more or less adhering to its own Zen. To me the addition of list comprehensions marks the time when that ship has set sail.
Good content, but typography is "killing" me. Specifically the choice of Roslindale as the main text typeface. Please consider using something that breathes a little more since even tweaking line-height is not enough.
Croat here, in practice we write them as separate characters dž instead of dž, etc. This makes correct aphabetical sorting like d -> dž -> đ -> e difficult to implement (and usually we don't bother).
This is the right decision and it has nothing to do with "US law" as some of the lwn people seem to be talking about. Russia is a dictatorship with sophisticated state-powered cyberwarfare capabilities. Regardless of whether a Russian-based maintainer has malicious intent towards the Linux kernel, it's beyond delusional to think that the Russian government isn't aware of their status as kernel developers or would hesitate to force them to abuse their position if it was of strategic value to the Russian leadership. Frankly it's a kindness to remove them from that sort of position and remove that risk to their personal safety.
It may or may not have been the right decision, but it was definitely the wrong way to go about it. At the very least there should have been an announcement and a reason provided. And thanks for their service so far. Not this cloak and dagger crap.
Indeed this was quite the inhumane way to let maintainers with hundreds of contributions go, this reply on the ML phrases it pretty well:
There is the form and there is the content – about the content one
cannot do much, when the state he or his organization resides in gives
an order.
But about the form one can indeed do much. No "Thank you!", no "I hope
we can work together again once the world has become sane(r)"... srsly,
what the hell.
Edit: There is another reply now with more details on which maintainers were removed, i.e. people whose employer is subject to an OFAC sanctions program - with a link to a list of specific companies.
I hope we can work together again once the world has become sane(r)
This would be a completely inappropriate response because it mischaracterizes the situation at hand: if the maintainers want to continue working on Linux, they only have to quit their jobs at companies producing weapons and parts used to kill Ukrainian children. It has nothing to do with the world being (in)sane, and everything to do with sanctions levied against companies complicit in mass murder.
Yes, the decision is reasonable whether or not it is right, but the communication and framing is terrible. "Sorry, but we're forced to remove you due to US law and/or executive orders. Thanks for your past contributions" would have been the better approach.
This is true of quite a few governments, including those you think are friendly, and it is a huge blind spot to believe otherwise. Dictatorship doesn’t have anything to do with it, it isn’t as though these decisions are made right at the top.
Do you have the same reaction to contributions from US-based companies that have military contracts? While the US isn't a dictatorship, the security and foreign policy apparatuses are very distant from democratic feedback.
Regardless of whether a Russian-based maintainer has malicious intent towards the Linux kernel, it’s beyond delusional to think that the Russian government isn’t aware of their status as kernel developers or would hesitate to force them to abuse their position if it was of strategic value to the Russian leadership.
It's hard to single out Russia for this in a post-Snowden world. Not to mention that if maintainers can be forced to do something nefarious, then they can do the same thing of their own will or for their own benefit.
Frankly it’s a kindness to remove them from that sort of position and remove that risk to their personal safety.
The Wikimedia Foundation has taken similar action by removing Wikipedia administrators from e.g. Iran as a protective measure (sorry, don't have links offhand), but even if that's the reason, the Linux actions seem to have a major lack of compassion for the people affected.
It wasn't xenophobia. The maintainers who were removed all worked for companies on a list of companies that US organizations and/or EU organizations are prohibited from "trading" with.
The message could have (and should have) been wrapped in a kinder envelope, but the rationale for the action was beyond the control of Linus & co.
Here’s what Linus has said, and it’s more than just “sanction.”
Moreover, we have to remove any maintainers who come from the following countries or regions, as they are listed in Countries of Particular Concern and are subject to impending sanctions:
Burma, People’s Republic of China, Cuba, Eritrea, Iran, the Democratic People’s Republic of Korea, Nicaragua, Pakistan, Russia, Saudi Arabia, Tajikistan, and Turkmenistan.
Algeria, Azerbaijan, the Central African Republic, Comoros, and Vietnam.
For People’s Republic of China, there are about 500 entities that are on the U.S. OFAC SDN / non-SDN lists, especially HUAWEI, which is one of the most active employers from versions 5.16 through 6.1, according to statistics. This is unacceptable, and we must take immediate action to address it, with the same reason
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system. Telling people you are laughing at them doesn't help anything.]
The same could be said of US contributors to Linux, even moreso considering the existence of National security letters. The US is also a far more powerful dictatorship than the Russian Federation, and is currently aiding at least two genocides.
The Linux Foundation should consider moving its seat to a country with more Free Software friendly legislation, like Iceland.
In other words, refusing to comply with international sanctions. This is in fact an incredibly high bar to clear for Iceland. It would require the country to dissociate itself from the Nordic Council, the EEA, and NATO.
a kernel dev quoted in the Phoronix article wrote:
Again, we're really sorry it's come to this, but all of the Linux infrastructure and a lot of its maintainers are in the US and we can't ignore the requirements of US law. We are hoping that this action alone will be sufficient to satisfy the US Treasury department in charge of sanctions and we won't also have to remove any existing patches.
that made me think it was due to US (not international) sanctions and that the demand was made by a US body without international jurisdiction. what am I missing?
Without a citation of which sanction they're referencing it's really hard to say. I assumed this sanction regime was one shared by the US and the EU, and that Iceland would follow as a member of NATO and the EEA. If it is specific to the US, like their continued boneheaded sanctions against Cuba, than basing the Linux foundation in another country would prevent this specific instance (a number of email addresses removed from a largely ceremonial text file in an open source project) from happening again.
Note however that Icelandic law might impose other restrictions on the foundation's work. The status of taxation as a non-profit is probably different.
even if it has to do with international sanctions, their interpretation and enforcement seems to have been particular to the US. it reeks of "national security" with all the jackbootery that comes with it.
A bit of an odd and pedantic remark, but their website (and ayllu itself) defaults to an arrow cursor rather than an I-beam, which strikes me as kind of odd. It gives off a texture that doesn't quite feel right for browser content.
I like the performance orientedness though. It loads really fast. I clicked the little preview on the main site and was surprised I got sent to an actual instance rather than just an enlarged screenshot. Another good example showing you can make fast websites if you want to.
By default Ayllu will traverse the last 100 commit messages while most forges only return somewhere between 5-20. The server also runs on a tiny Intel Atom host circa 2012 in the Netherlands which probably contributes to some of the slowness.
I appreciate the feedback here! I agree the UI on Ayllu could definitely use some work in terms of accessibility and general design, it's not an area I'm particularly strong in but I'm trying to make incremental improvements on it. Any patches in this regard would be greatly appreciated!
As an aside-aside, I really liked that post on digital textures; gets at something that I think many of us will have felt but not known how to put words to, or even where the boundaries of the concept lie.
I never needed cond in Erlang, and I rarely use it in Elixir. For that snippet I'd likely end up doing this (not sure if I got the tuple pattern matching syntax correct for Gleam):
case {your_hand == their hand, your_hand == winning_hand} {
// if
{True, _} -> Draw
// else if
{False, True} -> Win
// else
{False, _} -> Lose
}
The only downside is both expressions are always evaluated, whereas in that gleam snippet only one expression would be evaluated at a time. But these equality expressions are so trivial I doubt it impacts performance.
That's equivalent, but I think the parent commenter's point was that writing it your way was less ergonomic for them. Which I can agree with. You could do something gross like
case your_hand == their hand {
// if
True -> Draw
// else if
False if your_hand == winning_hand -> Win
// else
False -> Lose
}
to avoid casing on Nil, but it feels kludgier to me.
At work we have an email account set up for this. We replace the user's email with
deleted+<uuid>@<company>. Any emails sent end up in one of our inboxes.There's a fun talk with the same message. Very theatrical. Tell me a story by Saša Jurić. https://youtu.be/GOrKfCs-mr0
My brain dislikes justified paragraphs. Can't put my finger on why. I just find them less pleasant to read. Even if the word spacing is mostly uniform.
I believe the ragged-right whitespace pattern offers visual landmarks that make it easier to keep track of the line you're reading. No such cues in justified paragraphs.
Do you dislike them only on the web, or in print, too? It might be worth looking at some books published between about 1880 and 1950 to see what justified paragraphs are really supposed to look like, if you haven't already.
I created this stylesheet which hides everything but the videos and makes videos smaller so I can fit more on the screen, and I've been very happy with the results.
if you use ublock (which lobster readers probably already have), you can use this. no need to install another extention to apply custom css.
I made a similar stylesheet with some other modifications to space videos further apart and make things less busy.
A gui application and they have zero screenshots on their github homepage. 😆
Well to be fair, the first link in the readme leads to screenshots.
I appreciate that they just have a link to screenshots right at the top, instead of including them inline in the README.
Three reasons:
Also, it's a nice informative README. No notes from me. If you want the screenshots, they make them very easy to find.
I seriously feel like such links should be banned… it is so low-effort (not on the part of the OP, but on the project).
It's a link to their git and the readme, the home page itself has screenshots (https://git-cola.github.io/index.html). Maybe that would be a better link.
Sorry, I've thought that direct source code link will be better than redirection page with little info and images, and didn't realised there were no screenshots included in the
READMEfile.Personally, I don't like images embedded in README file, that's because I'm expecting to provide only the necessary and important instructions on how to use and build the project. No docs, code snippets, images, and all that.
I'm with you. Their README is not a low-effort README; I find it quite informative, and I'm glad you linked it instead of their marketing site. They included a link to screenshots right at the top of the README, and anyone who couldn't be bothered to click that but calls the README "low-effort" puzzles me.
Last time somebody shared here a link to WASM-focused language. You land on the marketing page with two lines of text, “Getting Started” prompting info about being not written yet, broken GitHub button below, and the only way to get into its page was another GH link in the navbar that actually worked…
Well, I still would like to see a navigation app based on OpenStreetMap, which is as good as Google Maps. Here We Go is considered the European alternative, but it is not as reliable and lacks some convenient integrations. I don't see how Open Source could help here.
The biggest take is that it is impossible to have fine-grained traffic data like Google without having billions of spyware enabled smartphone on the road. You just cannot do it.
2. I would like to stop comparing solutions on a pure technical basis. "I would like to quite Google Thing but the open source solution I tried doesn’t have the shiny color I like"
There’s so many problems with that kind of answers.
Firstly, it means you value "shiny coloured featured" more than political independence and ethical value. That’s a huge problem. So, basically, you don’t want to change! You would only migrate to a perfect clone (and, guess what? even for a perfect clone, you would not do it because "the clone doesn’t add value)
Secondly: You are not even trying. When you try alternative such as Comaps, you discover that there are many many ways in which Google Maps sucks: ever tried planning bikepacking trips or walking treks with Google Maps? Ever used Google Maps in a forest? It sucks. Every solutions shines in a given context. Saying "but Google Maps is superior" is simply false. The fact that most people, including those in tech, cannot see it is a huge problem.
Thirdly: The whole point is to advocate for institutions to invest into Open Source. We don’t care that it is not perfect yet. Nobody cares that you still use Google Maps on a personal level. That’s not the point. The point is : how can we make alternatives that first allows to survive without monopolies. If those alternative get better than monopolies, well, that would be the cherry on the cake. But that’s not the goal!
Thank you for arguing clearly exactly what has been in my head for the longest time! You first and second points under "2." really drive home the underlying problems. Open source (or any European alternatives) are having to fight an uphill battle. The deck is stacked against them.
I also think it will stay that way, probably forever, because of how easy it is to build a product and become Really Big in the US: few pesky consumer laws to comply with, a single unified culture and language with many states to grow into and solidify the product, before expanding into the European market. And of course, a big pool of VC investor money to get yourself bootstrapped. All of these mean the competition from closed source US products is fierce. We cannot compete with them on sheer features, and where we can (as you point out with Comaps etc), we cannot compete on marketing. Because this also boils down to simple brand loyalty and awareness.
All of the above are also big reasons why I advocate for the EU to attack the problem on multiple fronts. Education on the impact of software choices being a very important pillar, and requiring by law that government institutions use open source software being another.
Otherwise, this pressure to stick with the status quo is too strong to overcome, even with the very best software alternatives. And even if the alternatives come, people will just jump ship the moment the next product with a "shiny new color" comes along.
One solution to this is to pass legislation that mandates Google/others share live traffic data via a public API at minimal or no cost. This could apply to other similarly useful kinds of data as well.
In Czechia, phone operators share anonymized phone locations with public transportation companies and urban planners so that those can figure out how to make the trips more efficient.
I think this is a very important part that I fail to completely reason about.
In any area where there is profit to be made, a commercial company will have a huge incentive to make a shiny and even truly very user-friendly product. They have money to spend wisely on a good user experience.
Most open source projects cannot compete- they simply won't have comparable resources.
Plus, they will invest in marketing.
I think you are right that many times, the open source will be superior in some senses. For example, just by virtue of not being oriented towards a profit, it can be better for the user.
But, esp. in software that benefits from network effects, you're not in a good position because simply most people do not care. And why would they? I only have bandwidth to care about a finite number of things- and the things I care about mostly gravitate around things I know. I do not expect most people to care about which instant messaging platform to use!
And honestly, for example, in instant messaging, the popular choice comes and goes- we've been in a long stretch of WhatsApp dominance in many places, but the leader has nearly always been a proprietary IM platform. So WhatsApp will die and be replaced with something else, people will move their chats to the new shiny, and move on. While we the nerds appreciate that it would have been great if XMPP would have won in the early 2000s and had evolved through these days, it's hard to articulate to people who do not care about instant messaging about what basically they'll be using for a number of years and then change...
Maybe the answer is governments. I agree with what someone else said in the thread- just trying to be independent of the US does not seem like a good motivator for me. I think we just need to realize that some services are mandatory for modern life, and large companies should not be able to control them.
I mean, we could insist on standards for mass email providers. Even limit their bundling options, move to a similar model we have for mobile operators... even advance the standard to enable address e.g. portability.
And IM providers could be forced to agree on a set of live standards they will follow to enable interoperability and again, account portability.
We managed with phone numbers, so...
It's a policy choice. Either we allow private roads that can only be used with road owner rental vehicles or we force everyone to be compatible.
The federated XMPP/Jabber network has not won in terms of market share, but it sure has evolved and is still doing so. I am taking the train tomorrow to go the XMPP Summit 26, (some of) my family members and friends have been using Conversations and Monal for several years, so I would know. ;-)
Yes, it's moving forward, but their objectives don't really align with mine. I would like to set fire to WhatsApp and use any IM system that allows for third-party clients without encumbrances.
I decided to play again with XMPP very recently to see where it was at, esp. because I had played with Delta Chat and I was very impressed with that.
So I went with Conversations, that has the nicest onboarding I've ever seen for an XMPP client. (Slightly inferior to Delta Chat, but good enough.) But nowadays, even non-technical people expect to be able to join their IM to their computer.
Which should precisely be the forte of XMPP!
However, in WhatsApp (and Delta Chat, Telegram, Signal...) the mobile app guides you through this process, but I found no guide on Conversations to do this. Which means, if I recommend XMPP/Conversations to people I know, they'll likely won't be able to figure this out on their own.
(I filed a bug. Someone said this was not something that Conversations should do. It's still not closed, but it's been idle for over a month...)
I use Gajim on my Linuxes, so does my wife on her Windows, I used to use Beagle on my mac work laptop, I am not sure what you mean? You have to type a username@server and password, and that's it. There are also web clients like movim or converse. Maybe you mean the "flash this QR on your phone to login" flow? I believe Fast Authentication Streamlining Tokens could be used as a building block for such feature, but this is really outside my "expertise".
I was surprised to realize this, but among friends and family that have migrated -some- of their comms to XMPP, most do not care the slightest about using it outside their phone. I contributed some stuff to Gajim and would have been pleased to get some of my friends to use it, but the answer was 99% of the time: "oh no, I don't want messages on my computer, that's what the phone is for". The "nowadays" is what surprises me most about your sentence, I'm old enough to remember my non-technical friends using MSN Messenger on their 1024×768 CRT monitors, but "nowadays" they just want to use their phone.
Delta Chat and Telegram have proper non-mobile clients indeed, but for Signal and WhatsApp, the mobile client is the "main device" and the desktop/web client is a second-class citizen. Close the app on your phone and it will eventually stop working until you start the app again (and in the case of WhatsApp, you'll have to re-auth, using your phone).
Unlike any other alternatives you mentioned, there is no such thing as "their" objectives. The Conversations maintainer has his own objectives, sure, but XMPP has a democratic process for the protocol development. You don't even have to be involved in this process to just start building stuff that uses the protocol. It is a common good. If you want to get involved, it can just be participating in mailing lists and/or chat rooms. Relevant recent thread in the mailing list: Participation of civil society in the standardization process. Come back and let's "set fire to WhatsApp" together. :-)
I mean any process where someone follows the onboarding in Conversations, then taps on "add a device", and they get a client on a different device. I don't care the process as long as it's straightforward and it offers a "golden path" that works.
Then I can recommend people "get Conversations and follow the instructions to do a second device".
I'm all for there being choice, for a XEP standardizing the QRs, but there must be a clear simple process for people who want to do things without having to do research.
You might be right in that many, perhaps most, just care about having their IM on their phone, but I meet enough people who use a second device that for me it's one of the clearest obstacles for me to recommend XMPP. (In contrast, Delta Chat works and provides everything I need to recommend it.)
I know it's open source, and I like the idea of XMPP. But I need a mobile client to recommend that has a good onboarding that doesn't require research, including a second device. Conversations seems to be the closest (although the costs on the Play Store are a turn-off for me, although I understand it's likely needed), so that's why I filed the bug- because the more decent options available, the better. At the moment, Delta Chat ticks my boxes better- I hope some XMPP combo does too at some point.
Maybe we can learn from XMPP not winning the IM world at the time. And from email being more and more centralized. And for the web itself (which is now gatekeeped by two browsers).
https://ploum.net/2023-06-23-how-to-kill-decentralised-networks.html
I think there's another lesson to be learned. Beyond user experience and marketing, private companies trying to dominate a market will collaborate with existing open standards etc. exactly as long as it benefits them. But not a second longer.
So yes, Meta's Threads is supposed to federate with ActivityPub. No one should base any decision on that, because if (when) Threads achieves sufficient marketshare, ActivityPub will disappear.
I use ActivityPub and I refuse to engage with BlueSky or Threads, because it's a trap.
However, a lot of people on ActivityPub- even those who are escaping from proprietary networks and aware of their tactics, will still engage and feed BlueSky and Threads.
I don't disagree with you here, but I wonder if it answers the right question?
The question is not if the EU should invest more into political independence or its values. The question is how. Why should the EU rather invest into OSS than into proprietary EU companies?
Also, I would assume they hope for details. Like, should the EU invest more into the OSS social media or OSS operating systems or OSS smartphones or ...?
The most important question though, since this is a "call for evidence": What is the evidence that such investments will pay off (quickly and bigly)?
Your article is a well-written call for action. When I look at this call for evidence, specifically the section "Consultation strategy", I don't think it addresses their questions.
Coming back to the example of a "map app": It is probably not hard to find evidence how OSM has made mapping data cheaper and easily available without downsides if you get the data from Google. Would investing a few more millions there be a good investment? How would it pay off? Probably not in the navigation business because "it is impossible to have fine-grained traffic data like Google without having billions of spyware enabled smartphone on the road".
Maybe there is something else related to OSM? The data is already great and while maintenance is important, what could there be as a next big step? Thousands of municipalities in the EU deal with mapping data. Maybe OSM could help there and save billions of tax money within ten years? I don't know and neither does the EU commission.
OSM has severe limitations for building a modern mapping data model, ignoring the well-known data license and graffiti issues for the moment.
It is impossible to use human crowdsourcing to detect and process the number of changes per day that actually occur in cutting edge mapping data models. Even with fully automated pipelines processing petabytes of real-time telemetry per day, it is generally recognized that there are still large gaps in accuracy and freshness that limit usability for many applications. The large closed source mapping data models only use the OSM data model for parallel construction, cueing their systems to look for potential changes they can find in other data sources.
This is the crux of the matter. Maps are no longer slowly changing display artifacts created by humans. Today, they are real-time petabyte-to-exabyte scale data models that are being continuously updated from data flowing in at rates most software devs can barely imagine. The data informing the data model is often noisy, erroneous, corrupted, or manipulated so you need sophisticated and compute-intensive inference pipelines to clean this up. Put this way, it is obvious why Google is so much better at this than the traditional mapping companies.
There is a large opportunity here. Industry has an enormous desire for a competitive non-Google map infrastructure, so if the EU were to build one it would be immediately relevant globally. The large caveat is that it is impossible to build a state-of-the-art map infrastructure on current OSS. Creating it is a commitment to building a completely bespoke data infrastructure with a lot of computer science that is not in the literature. The reason this doesn't already exist isn't because many companies have not tried to build it, it is because it requires extreme technical sophistication to have a prayer of implementing it successfully.
The barrier to an MVP in mapping is very, very high and there is no way to get around that fact. In principle the EU could acquire the necessary data sources to build these data models, that isn't the real limitation.
My house showed up on OSM as under construction when the work started. When I moved in, OSM showed the building and had the unfinished half of the estate marked as under construction. It took Google Maps over a year to notice that the estate existed.
Some roads around here became one-way streets a while ago. Again, OSM had up-to-date data within a week. It took Google Maps and the major SatNav systems over a year to stop sending people down the roads.
One of the roads nearby was being used as a cut through so they put a barrier in the middle and turned it into two dead ends. OSM had that marked before the construction work finished. I helped some folks who were struggling to turn around towing a caravan a few months later. Why had they gone down a dead-end street? Because Google Maps had told them to.
So, while I agree that updating maps is hard, I have seen plenty of evidence that OSM is better at it than proprietary map data sets.
The quality and freshness of all base maps vary widely based on where you are. OSM is good in some areas but is much more uneven globally than the closed source base maps.
There are only about half a dozen global base maps, of which OSM is one. Every company that builds a base map regularly compares their coverage and quality against their competitors. All of the base maps, including Google, have regions where they perform well and areas where they perform poorly.
No current base map is a clear winner globally. Beyond requiring more consistent access to data globally, it requires technical infrastructure at a scale that OSM and most closed source base maps will never be able to build.
I guess it's locality-dependent. Waze (which I assume shares data with Google Maps) picked up newly constructed roundabouts around here 6 months or so before they were added to Open Street Map.
Ultimately someone has to sit down and edit OSM, and if the people who know how to edit it aren't evenly distributed then some places will lag behind.
The same thing is true for Google Maps, and it's much harder to scale a thing where a single company has to maintain the data thna it is a distributed environment.
In Cambridge, the University of Cambridge used to maintain their own maps. These were very detailed for university / college sites, but also needed to be kept up to date for surrounding roads because they were used to show people how to get to various university places. Once OSM reached a certain level of quality, they realised it was cheaper to contribute their map data to OSM than maintain their own, so they wrote an OSM renderer that matched their old style and switched over. They still have people updating the areas in and around university sites, but now other people mostly handle the rest of the city. Everyone wins. With Google, you can put in requests for them to add / fix things in the maps, but they can't easily do the things like footpaths inside colleges because they're private property. And you can't then use their data for all purposes, you need to have a legal agreement to use them for any commercial purpose and using them also comes with a load of GDPR obligations.
It's quite possible that the mapping of hiking and biking trails is better in OSM than Google for where I live. I think there's a certain disunion between people who commute regularly by car - thus providing input to Waze/Google - and those that hike and bike who might be more familiar with navigation aids that rely more on OSM.
Possibly. But when travelling I've found that there's a very sharp drop off in Google Maps quality when you get out of populous areas. My mother lives in rural France and Google maps doesn't have half the roads near here, whereas OSM has them with metadata about the kind of road, and also has a load of the walking paths and most of the buildings marked. In the middle of Paris, Google Maps data is quite good.
I'm ready to believe that, but the fact is OSM is very often more right than Google Maps here in Belgium, a country that should have more than enough data available to display a 120 km/h speed limit on highways instead of the random 90, 110 and even 130 (that exists nowhere in Belgium) limits that Google Maps often shows.
After years of taking my own street the right way, Google Maps still doesn't know that that street goes that way. And I do use Google Maps, and so do probably a sizeable amount of the cars that travel through that street every day. It's not a low-traffic road.
It's probably different in other places. But in Belgium and France at least, when I want to be sure the data on my map is correct, I use OpenStreetMap. The rest of the time I use Google Maps for traffic information.
This is true, and true of more than just maps.
I've worked at a few health care companies, for example, and payment in that world is generally based on a combination of factors like name, billing address, license ID number, etc. But when I last worked in the industry, the measured half-life of that data was 18 months.
Meaning, if you compiled a list of all the doctors, hospitals, etc. you have a payment relationship with, and their names, addresses, license numbers, etc., then 18 months later half of them would have changed in ways that invalidate your data set. Which is a huge source of unnecessary friction and anger, since to the patient "it's just Dr. Smith, the same Dr. Smith I've been going to for years!" but in that time Dr. Smith has, for payment purposes, operated under four or five different names, used half a dozen different billing addresses... and so, no, the payment systems can't just automatically verify that "it's still Dr. Smith".
I disagree. The reason it hasn't been built is because there is no straightforward way to pay for it.
This is the kind of unprofitable baseline infrastructural thing that is absolutely the bailiwick of government. Sure, the government won't do a great job, but it can provide a publicly accessible baseline that decreases the entry cost of other people building on it.
To be competitive, the base map needs to be consistent and global. Many governments, including those in Europe, actively resist the notion of giving their mapping data to an organization not based in their country. To the extent they are willing to pool their data in a common system, most want absolute veto power over any mapping edits that concerns their country in that system. The intransigence of governments on these issues has been a major roadblock despite many efforts to build a common base map.
It would be great to have a global base map operated as a public utility. Most companies and people would prefer that. The base maps that we have now are a reaction to governments being unwilling to participate in a public map, but most heavy users of them recognize that they are not really fit for modern use cases. Perhaps ironically, the US government is by far one of the friendliest to this idea and always has been.
The direction the industry has been headed is the creation of a new commercial mapping entity that is deemed sufficiently neutral by design that all the major tech companies (except Google) are okay with them controlling the One True Map. There are plenty of ways for this to be paid for, that isn't a material issue.
Thanks for the thoughtful reply.
As someone from the the US, the idea that the government would try to hoard and control map data is a bit weird to me since the USGS (and most government entities) has had that kind of thing available for general purchase for decades.
One difference is that in most countries the mapping data models were originally created for national defense. Those organizations often still view that data through a national security lens, or at the very least, control of that data for national security purposes is an input to the process.
In the US, these concerns have long been separated. The USGS is a purely civil organization, essentially a public service. For national defense, there is a large and entirely separate US government organization (NGA) that is responsible for creating mapping data models in the national security context.
I travel a lot in the UK and Europe and I use Organic Maps (OSM), google maps and CityMapper. OSM has always been at least as good for road maps and always better for walking and pedestrian paths (often much much better).
I use CityMapper and Google maps for turn-by-turn directions, and google maps for finding restaurants (OSM has pretty out of date info on restaurants and businesses, in my experience)
My article is a translation from an old article and I agree that it doesn’t really answer the call for evidence.
And the main reason is that I believe the very first thing we need to do is to shift our priorities. To change our very definition of success. As long as we play by US-monopolies rules, we will not be able to make anything but sub-par cheap clones. It simply cannot work that way. Even if it start to work, it will be easily subverted (see my history of the Nokia Meego debacle)
I use Organic Maps on my phone and, unlike Google Maps it:
It doesn't have live traffic data, but it also isn't connected to a massive surveillance system that it uses to build the traffic data so I consider that a win overall.
mapy.cz (now, mapy.com) is one of these I believe.
They are in extremely common use in Czechia, much more common than Google because Google Maps seems to have a difficult time mapping and pathfinding in Czechia.
I was recommended this because I wanted to cycle to work.
(I also learned that “cycling to work” means totally different things to Scandinavians and Czech people).
Scandinavian here, in what way are the terms different?
"Cykla till jobbet" means taking a bicycle to your workplace in Swedish.
I can't describe it so maybe it helps a bit with a visual difference (note: I used London and NL, but the difference is core to what I mean).
Cyclist (cycling to work) for Czech people: https://standfirst-thecriticmag-production.imgix.net/uploads/2025/07/feature-milibank-lycra-cyclist.jpg?fit=crop&crop=faces,entropy&q=40&auto=compress,format&w=956&h=1438
Cyclist (cycling to work) for Scandiavians: https://bicycledutch.wordpress.com/wp-content/uploads/2014/08/utrechtsummer2014-04.jpg
LOL in Stockholm every cyclist basically looks like pic nr 1.
If someone rode around with a kid who was not wearing a helmet and not securely strapped into a child seat on the rear, they'd probably be arrested.
Oh, I live in Malmö- people in the first picture basically don’t exist.
To be fair with you, the few dozen times I have been to Stockholm it didn’t strike me as a very friendly place to cycle, and I didn’t see many cyclists. But the ones I did see didn’t see to stand out as being especially spandex laden with clip-ins.
I don’t think I even saw a helmet, but; your experience likely trumps mine as I’ve only lived in Sweden for 11 years and most of that was spent in Malmö.
Stockholm is significantly more bike-hostile than probably any other city in Sweden. Maybe Göteborg beats it?
I would never bike without a helmet. I know of too many people who have been seriously injured or killed when biking.
I will not let this random dunk and affront to my city of residence to stand unchallenged. GBG cyclist here, both for leisure and work commuting. The road traffic is kind of random with lane changes and trams and all, but the proportion of a'holes and psychos at the wheel is not higher than elsewhere.
I figured there would be even more salt-of-the-earth Volvo car builders and owners there, than the Beemer and Merc drivers up here. Not saying they're any less assholish about cyclists, though.
time to jump on a train and revise your priors because man, you need it
My most trusted source on Göteborg is Veyron i Ottan, is that bad? 😉
That begs the question of whether or not helmets significantly reduce serious injuries and death on net. Obviously, all other things being equal it’s not worse to have a helmet on than to be bareheaded when struck on the head. But the helmet can not be effective at all in certain head strikes (e.g. when struck by a bullet or fast vehicle). And of course it is not effective if not hit on the head: helmets do nothing against hand, wrist and forearm injuries.
Helmets are not costless, either. Besides the actual cost, they make cycling less spontaneous and more inconvenient (one has to lug the helmet around after parking the bike). The number-one threat to cyclists is not themselves but drivers: studies have shown that cyclist safety increases with the number of cyclists on the road: by decreasing the number of cyclists, helmets reduce safety. Worse, other studies have shown that drives drive more closely to helmeted cyclists, increasing the risk of collisions. So helmets lead to drivers less familiar with cyclists, and more prone to driving dangerously close.
I believe it is true that helmets on net are bad for individuals, bad for cycling and bad for society. I don’t believe that they should be banned, but I definitely don’t believe that they should be encouraged or mandated.
For other readers, this isn't just one person's opinion. Several influential cycling campaign group hold similar positions, e.g. Cycling UK recommends that helmets should not be promoted.
I think using a helmet is a deterrent to cycling for a bunch of reasons:
Personally, I cycle without a helmet unless I am going on a long cycling holiday, then I sometimes wear one because I am more likely to be cycling in poor conditions and on steep roads.
I think data on hire-bike users pretty clearly suggests that helmets are not necessary for cycling in cities.
If you want more rationale for not mandating helmets, I recommend Ruwen Ogien's works, a French philosopher who strongly advocated against any laws meant to prevent self-inflicted harm or risks (while at the same time absolutely not being a "libertarian" in the US sense). I don't know how many of his books have been translated in English though.
Back to the subject, I do think it's more reasonable to wear a helmet, but I generally agree with Ogien's "minimal ethics" and don't think mandating them would count as "progress".
Your bicycling environment sounds much more challenging than mine, to be honest.
Yeah that second picture is from the Netherlands (it even says NL on it and in the URL). Perfectly normal here, even though there's heavy campaigning from the government to try and change things...
The point was to show the difference, because the dutch way and the british way are wildly different and its easier to find pictures of those.
Sorry for not taking first hand photos to illustrate the point.
That first pic is from London, and spandex wearing commuters are fairly rare - usually going long distances. Most other people cycle commuting are wearing a helmet and some outside clothes. Maybe clips and cycling gloves.
[Comment removed by author]
Appreciate the pictures but how did you mix up the countries so badly? LOL. The first one is from London. The second is from NL. But thanks for the pics.
You should probably read the bit of my comment where I explicitly mention that these are london and netherlands.
The point was to illustrate the difference, not to specifically show a Czech cyclist (they are somewhat rare) and a Copenhagen/Malmö cyclist (common, but they just look the same as the dutch photo and sourcing photos is more difficult).
Munich has both, I'd say - although pic1 would more be "full bike touring kit (rain gear or leggings+gloves+rainjacket/jersey) on a touring bike with panniers" instead of spandex. And I'm kinda equating p1 because it's specialized cycling kit, just not the performance aspect :P
(Me, I'm probably in between. If the weather is bad I'll use different pants than I'm wearing in the office and because I own rain gear and panniers, I'll use that if needed - but you'd 90% see me commuting in normal clothes and a backpack on a reasonable bike with fenders)
Not only for cycling, but also for hiking. I have never been disappointed by a hiking route suggested by Mapy.cz. It's much better than Komoot because mapy.cz can suggest a hiking route based on your current position and the distance you want to hike. However, the app tends to crash frequently and some time (years?) ago they started advertising premium features.
Have you tried OsmAnd (developed by OsmAnd BV in the Netherlands, fully open source)? I have been using it for years, and they also have a very nice funding model: If you download from F-Droid, you get all the features, but if you download from Google Play or the iOS App Store, some non-essential functionalities are in-app-purchases, e.g. unlimited offline map packs.
Yes, you can download the entire world as an offline map, if you wanted. Device storage has become so large, there is little reason not to download your entire home country or state. More than once have I found myself in the forest with no cell reception with friends, who then struggled to navigate on their always-online apps. In contrast, OsmAnd just worked for me. It has dozens of modes (walking, bike, car, airplane, etc.).
Many weaknesses have been resolved in recent years, including sometimes choppy map rendering and slow route calculation. Given the excellent OSM data, you get lane-level navigation hints, which Apple Maps and Google Maps often get wrong.
I found OsmAnd to be super clunky on iOS, and I didn't love the style in which the maps were rendered. I've heard it's a better experience on Android. If you say the choppy rendering has improved, maybe I'll try it again.
I personally like Organic Maps – also open source, Estonian. Also has offline maps.
Haven't used navigation very much in either of them, TBH. When on foot, I usually need public transit timetable, and when driving, I prefer a dedicated device over a phone app (and want some live traffic info). When my TomTom broke I tried a bunch of apps, it seemed like none of them was designed by somebody who ever drove a car, Google Maps was the absolute worst. I could go on a rant here.
Also, navigation apps could be so much better, I'd love to have an open source, hackable thing on my dashboard. The bottleneck for really nice features seems to be data access (live traffic info, live public transit status; I've never seen it, but I'd love a map/navi that would utilize live weather info ("Heads up: you're about to drive into a thunderstorm" type alerts or a wind indicator on a navi screen would be great))
There been some drama regarding organic maps. I switched to comaps which is yet another fork.
I did a slow blink at this - for context, maps.me spawned Organic Maps, which spawned comaps. All use data from OpenStreetMap and can show maps offline.
Really good points, thanks! Yes, weather info would be great. Public transit is in beta at OsmAnd and I tested it successfully in Prague a few weeks ago. Apart from that, I really like Öffi, but they are fighting an uphill battle as more and more public transport companies close their APIs and force their shoddy apps on everybody.
I tried OsmAnd a few years ago and it was complete garbage. Happy to report it is much faster now and has gained many useful features.
That said, it's a loooong way off from the usability and performance of Google Maps. This is not something regular consumers are going to consider using. Which is a shame, because OSM data is really good.
GMaps does have an offline mode :)
Of course you need to download the map before.
What worries me more is that you don't get GPS coordinates from GMaps if you didn't download the area you are in.
Granted, but the offline mode in Google Maps is really primitive in comparison. I can download a complete map of Germany or even Europe and it only takes a few GBs of space, which is nothing in many of today's smartphones. And no matter where you place me, it would just be there. There is also a world overview map with coarse roads and large cities.
I used it years ago. It is a nice app to work with OSM data.
It seems it does not show/use traffic information. Beyond that, Google Maps offers all the additional information about shops and restaurants (opening times, menus, etc).
You can download Google live traffic data as an overlay, but it's not (yet) included in the routing. It may sound like a cope, but it gives me an incentive not to turn my brain off when navigated.
Regarding opening hours, Google of course is king, but at least here in Germany and larger european cities, this data is pretty up to date, too, though nothing beats businesses proactively updating their times for Google.
You won’t have that because Google has put the data sharing app in all android phones and AFAIK you can’t opt out.
It’s part of the reason Apple and Google are able to do live traffic reporting. Android phones in cars do data reporting, without the consent of the driver and on aggregate is the most accurate “live view” system we have.
Worth noting from the podcast Adam remains “optimistic” about AI and is interested in using it, suggesting GitHub triage as a possible area. 🤷
I don't understand this at all. Even as someone who doesn't use AI at all and isn't directly impacted by it, I feel like we should burn the entire "AI industry" to the ground because of all the negative externalities it has. And here's someone who's business is directly being destroyed because of AI and is "optimistic" about it?!
If you are a saddle maker that makes saddles for horses and your business goes away, you can still be excited about cars and adjust to new realities. Those things are not necessarily in conflict.
The major difference is that tailwind is far from obsolete unlike horses in your simile. If tailwind development goes away it hurts all users. AI is not a replacement for tailwind.
Tailwind is not, but Tailwind was not commercialized. Tailwind UI was the cash cow and that is obsolete now that you can have an agent build your own components quickly. The entire business model in this form broke away.
To be fair it was a business model that shouldn't exist in the first place, AI just made it more obvious. I place it on the same shelf as other types of jobs that only exist because of some external deficiency not related to the technology they serve, like train fare collectors and pedestrian crossing guards. It's the human element that is lacking in all three of these, either human skill issues for CSS, human dishonesty for trains or human inattention for roads. Once that is solved by some method the need for those business models disappear.
I feel like train fare collectors and crossing guards serve more purpose than just compensating for human dishonesty or inattention.
A fare collector can also help people who have some problem with a ticket. A crossing guard can help people across the street, regardless of how reckless or careful drivers may be.
I also just ... cringe at the idea that everything should just be ruthlessly automated and efficient and there is no place for humans to do jobs that may not be "essential" for some value of "essential" but where a human touch is a good thing. I insist, for example, on using cashiers rather than self checkout.
I'm old. I remember when our town still had a gas station that had an attendant. Yeah, I can pump my own gas. But it sure was nice for my mom when she was wrangling toddlers, or no doubt for older people, etc. Absolutely necessary? No. A fulfilling job? I dunno - ISTM some folks enjoyed the work. I'm not sure we're better off for the automation.
If the human adds value, the human should have a place. But there are plenty of cases where humans add negative value compared to the automation. The self checkout? Depends on the situation, sometimes the human is helpful, sometimes it is not or negative. Ordering at McDonalds? Quite frankly the touch screen version is such an improvement over the human version because the screen is patient and the human is not.
I think there is so much more nuance between machine vs human and there is no one opinion fits all situations here.
I think "adds value" is too vague here. Adding value in a job can mean:
2 and 3 are really important for people and society, but businesses choosing to replace a job with AI only give a shit about 1.
I'm only talking about value added in the consumer interaction. As a consumer, I don't care whether the company makes money. What matters is whether an interaction with a human, compared to an interaction with a machine, adds value for me in that moment. And in some cases, the human adds negative value.
I agree that that's the right thing to focus on.
But we also shouldn't be surprised when businesses don't focus on that because they have no incentive to and if they do, they can be outcompeted by other businesses who don't.
I'll admit, I prefer the touchscreen ordering. But I'm not sure it's an improvement for everybody. It's certainly not an improvement for people with visual disabilities (at least my recollection of the touchscreen stuff is that it's not a11y-friendly, I could be wrong here).
"I think there is so much more nuance between machine vs human and there is no one opinion fits all situations here."
Well, we agree on that for sure. I'm taking the position that, because people must have jobs to exist in society as it stands right now, that's priority one. Nobody who is willing to work a full-time job should be unable to find work that pays well enough for the person to sustain themselves and (if necessary) their dependents at a non-poverty level. That's my starting position.
I don't have disabilities so I cannot judge from first hand, but I know that it has a huge accessibility button that I pressed out of interest and it offers a zoom mode and a reachability mode. The latter I actually use for my kids sometimes.
I agree with this on the surface, at least in theory. But it can’t hold for 100% of people. There will always be individuals who are willing to work but whose skills, or what they can offer, don’t find a buyer on the other side. It’s still a market, and no one can be forced to employ someone. As a result, there will always be a segment for whom finding work is hard. In periods of uncertainty like the one we're in now with AI changing things, that segment naturally grows. I'm sure that over time, things will rebalance. IMO the idea that no job can ever be lost doesn't seem right and it might not even have to be the highest priority.
I started writing this today right after leaving from a Taco Bell where I had just witnessed a fight nearly break out with a customer and some staff because of how behind they were.
This was with:
The problem I think isn't automation really, it's that the wrong things are being automated.
It seems like the fast food industry is only optimizing for the Producer (placing orders) and never touching the Consumer (making orders) part of the fast food industry's version of the Producer-Consumer problem.
There is literally nothing managing back pressure and it is super obvious. "fast food" in 2026 is only fast in the sense that large amounts of orders get to the kitchen very quickly. If you replaced fast food workers right now, it still wouldn't solve anything unless you could plop a giant food factory on top of an existing fast food place or something.
It's part of the reason why dine-in places in 2026 are so much better speed wise. The back pressure management is built into the experience, lol.
If you take out the kiosks and the DoorDash, much to many people's dismay, we'd probably see a vast improvement in how quickly folks can deliver product.
I noticed this at Starbucks in San Francisco. There is no visual clue of what is going on because of mobile orders. I ordered in store a cappuccino and i had to wait for 30 minutes because of how many people were apparently ahead of me.
I'm not sure though if that is a problem at McDonalds. I did not notice it there much, in part because they also make it very clear how long the queue in front of you is visually on the screen. So if there is a lot happening, you can bounce as a customer.
Hard agree and I would love to see more (friendly) discussion about this!
But in my experience the current wave that people around me are riding is productivity/result over empathy without much nuance. Buying art generated by a model instead of from people is one of the first examples that come to mind.
From a consumer perspective I can understand this, but depending on where you are, instead of paying/improving quality of life for people in service industry...they got replaced with screens? I know that's not what you're saying, but I'm having a hard time understanding the goal here with "humans should add value" w.r.t. McDonalds.
I don't think I'm advocating for automation per se, or even efficiency. Just pointing out that that some jobs are inherently unproductive and merely exist because some other element of society is broken.
In Brazil we have a bus ticket "cashier", sort of an overpowered fare collector. You enter the bus and they sit on a special seat with a turnstile, you pay them and go through the turnstile. If you don't pay they're there to physically stop you. They exist because without them a lot of people would not pay for the bus.
That this job even exists is a deficiency in the Brazilian society both in terms of income inequality as well as dishonesty. In all countries in Europe, Asia or Oceania I've been they don't exist: the driver does that job. In Luxembourg the bus is free so there's not even that issue.
This is an example of a "useless" job. I don't advocate for their disappearance because here at least it'd make a lot of people unemployed but it's a very bad usage of a human brain. That person could be doing better things with their time if society was less broken in that particular way.
"That person could be doing better things with their time if society was less broken in that particular way."
Could they? This assumes there's an available job for them to transition to -- and I'm not convinced that's the case. I don't know how public transit works in Brazil, I'll admit, but if the entity that allocates funds for work eliminates that job -- is it going to create a job elsewhere that the people can move to?
I see a common thread in tech circles where people will voice opinions along the line of "just get a better job" or "that job shouldn't exist because it's a dead-end job", etc. Basically, there seems to be some underlying assumptions that 1) there's an endless supply of jobs, and 2) everybody should always be "upskilling" and trying to hustle into more challenging work, etc. But a lot of people really only care about jobs because they must have one to exist. They don't want to be constantly retraining for new things. They just want to pay their bills and be left alone.
So endlessly automating things so that these "useless" jobs don't exist is, IMO, not a net good for society.
Now - if a job/need exists and you can't find people willing to do it, that's a good sign that it should be automated, eliminated, or otherwise changed so that people are willing to do it. In my previous example of gas station attendants -- if that job was eliminated because nobody wanted to do the work, then that seems like a net benefit for society.
I don't really understand how, by this logic, some jobs are productive, and some are not. All jobs are there to "fix" deficiencies with the way things are. Some jobs mold nature into different shapes. Some jobs attempt to fix broken bodies. Some jobs correct for human nature.
I come from Brazil and I haven’t seen a bus with a cashier for more than a decade. Curious to hear where they are still employed.
Pretty much all major cities still have them. Smaller cities usually don't but then again they mostly never did.
With more tech (RFID and NFC fares) there's talks of phasing them out, some places have but it's not widespread.
Telerik has the same business model Tailwind does, and has done so for decades. The business strategy is valid because customers don't want to hire someone to do the work themselves and would rather rely on a consultancy. Are you suggesting consultancies are invalid businesses?
Tailwind UI Plus provides ready-made components and templates, it's not a consultancy.
By the way "invalid" is your own take on this. I didn't suggest they are invalid business models, just unnecessary ones that a slight change in external factors would obviate. They provide a value to current society due to these human inefficiencies but their value is not intrinsic.
I think you're just arguing semantics when your phrasing was "shouldn't exist in the first place", i.e., not valid.
Sure, but those components have something that makes them special over the OSS offering that people wanted to pay for and were asked for by some customer in the first place.
My point is the "custom widget" market exists, has existed for a long time, and AI killing this business model and being built on stolen code kind of makes this another drop in the bucket against its use.
I'm no expert in this business and I don't know what their consumer base looks like, but I would not be terribly surprised if the purchasers of those components were a very specific crowd. The adoption of tailwind-ui within my circles has been exactly zero. Yet they make a lot of money, so they must have some pretty good usage numbers but they are probably from some specific crowds that were looking for that precise solution.
Among my circles on the other hand shadcn took off quickly and can be found in many applications now. Obviously shadcn is also free, but the value proposition also seems better for react developers than tailwind UI was.
So "people wanted to pay for" is valid, but "some people wanted to pay for" and those people are now increasingly looking elsewhere.
I don't think that's the takeaway here. The knowledge of the offering's existence has basically fallen off entirely because the AI isn't showing it to them in the first place unlike if a human was looking at the docs instead.
That is an assumption which might or might not be true. My suspicion is that it's wrong. The AI is building people entire components so the need does not exist. If i ask my coding agent where to get high quality tailwind components the first link it gives me is tailwind UI. If someone were to still ask that question, they would be pointed right at tailwind. But the work the agent does even without having access to it, is sufficient and might even be better.
It's not an assumption, it's a stated fact from their CEO.
The traffic to the docs is down is a fact. That someone were to reach for tailwind UI if only the agent would advertise tailwind UI is the assumption that is not validated. If that was the case, I'm sure they could make the experiment with amp which offer ads in the free coding agent and see if that helps conversions.
One might guess that it's more a case of...
optimistic about the potential of AI technologyoptimistic about hyping up some new idea by labelling it AI in some way in order to secure VC fundingDoom on a pregnancy test does not belong there, because Doom never ran on a pregnancy test. They crushed Doom gameplay video down to low resolution for an OLED display they mounted inside a pregnancy test, wired to a raspi or whatever off-screen. It's just a stupid hoax.
Qualifies about as much as "Doom on sound" or "Doom on Desmos". It's not a very high standard.
It was a fun thing to do, Foone explained what they'd done, it was not meant to fool anyone into thinking that doom ran on the pregnancy test hardware itself.
On Twitter (screenshot)
Tumblr
Reeks of hoax to me. Letting an unfunny joke go too far is also manufacturing a hoax
I'm allergic to barley :(
Wheat beer for you then!
i’m also allergic to wheat :(
Is gluten the problem, or something else? My coeliac friend likes a bottle of gluten-free Old Speckled Hen.
to be honest, I'm not certain - my doctor thinks it's an immune condition of some kind. I'm allergic to most American staples - eggs, dairy, most forms of gluten, etc. I'll have to give old speckled hen a try!!
I suspect you've probably stumbled on Burning Brothers, and Sociable Ciderwerks; for anyone who hasn't, there are actually good GF beer/beer-adjacent options these days! ☺
All the best,
Thanks, my mom is gluten-intolerant and has often lamented being unable to have a beer anymore.
Wheat beers must have barley as well
Why
Enzymatic power and lautering
I'm a little sad that CPU and RAM are soldered on. They claim they were unable to find a modular solution without sacrificing performance. Interested into seeing some reviews soon.
I'll always choose the bandwidth and performance over socketed components. Truth is, many people want socketed components, but then never do anything with it. In fact, the vast majority of PC owners follow this. It's a small portion of enthusiasts who might upgrade.
Still. Performance please.
A lot of people upgrade RAM. CPU, ehhh much less often; that usually entails upgrading the motherboard anyway.
If you think about it a bit, it really doesn't make as much sense as it could seem to make at first. Who, exactly, upgrades RAM? Broadly, I know two categories of people: laptop owners and people who lack enough funds to build a workstation to their liking at once (so they end up buying less RAM than they wanted, with the intent to buy the rest in future).
Upgrading RAM in a laptop is somewhat of a self-inflicted pain: many laptops are sold with woefully inadequate RAM, and sticking more RAM into an already owned laptop is one of the very few ways to delay the obsolescence of a given hardware platform. This is not a problem here.
Framework is not a cheap vendor, not even a cost-effective one. This automatically self-selects for people who do have enough funds to purchase a workstation to their preferred spec. So, again, not a problem here.
Anything I've missed?
I had ram failure that would cause weird errors on my T480s. Luckily it was the external ram that failed and not the soldered one or else my laptop wouldve been e-waste. Now 5 years in, it’s still running extremely smoothly.
I have just bought more ram for my work machine which is about 1 year old. I switched from python to android development which is much more memory hungry due to emulators and large IDEs. So I'm upgrading from 32 to 64 gigs.
I have also upgraded the CPU once in the past. AMD had very good compatibility on am4. I only needed to upgrade the bios.
RAM requirements and expectations have historically increased very rapidly. It wasn’t too long ago that 16 GB was considered questionably excessive, and is now considered the minimum to run modern games (or run several Electron apps simultaneously). A basic developer workstation probably starts at 32 GB these days, with option for expansion to 256 GB or more. I would not have much use for 256 GB of RAM today but in 5-10 years I probably will.
A lot more people don’t.
This Lobsters cohort is going to naturally be an exception to this.
But if we take a step back and think about the average consumer, it’s just not happening. I’d say >99% of MacBooks, for example, are never upgraded.
And of course Strix Halo is in the early adopter part of the adoption curve now. So I recognize the mismatch between the current consumer, who are technical enthusiasts. But as it commoditizes, that’ll shift.
For now, I as an early adopter/enthusiast, definitely want performance over upgrade ability. But I acknowledge others may not see it the same way.
Yeah. This is due to technical constrains in the specifics of the AMD CPU/GPU/chipset.
In the LTT video[1] someone from the company said they "Asked AMD" about replaceable memory and AMD came back and said due to "signal integrity [...] over the 256-bit bus" it isn't possible.
(Check out the clip. Starts at 7:25).
1: https://www.youtube.com/watch?v=-lErGZZgUbY&t=447
Edit: that "someone from the company" is Nirav Patel, the CEO of Framework
I'm half seriously thinking that a socket you could stick something Optane-like in for superfast swap / cache would be the answer here :)
How is Optane, which is slower than DRAM, the answer to something where even "regular" DDR5 RAM isn't fast enough?
Some way to expand the system with something faster than an normal SSD? Hell, I suppose slower DRAM to act as a ram disk would be good enough, if the chip set could drive both. PCIe connected ram disk maybe?
You're missing the point entirely.
The problem that's being solved by Framework is not "expanding the system with something faster than an SSD". It's "expanding the system with something faster than DDR5".
I think you're missing mine, actually.
My experience with trying to keep machines from become obsolete suggests that the ability to add any extra RAM - or something nearly equivalent - is perhaps more important than its speed. Admittedly though I am an old git and maybe my use cases are out of date.
Personally a machine with soldered down very fast RAM with an option to add a "faster than SSD" ram disk for use as swap later on seems like it would remain useful for longer than a machine without that option. But my machines tend to get used as conventional servers or build farms, maybe other uses cases that I'm not thinking about would not benefit as much.
The entire subthread begins with your assertion that something faster than SSD (but slower than DRAM) "would be the answer here", where "here" is a demand for a type of extremely fast DRAM that uses a 256-bit wide bus (i.e., several times more than regular DDRx SDRAM).
If I'm "missing your point", then it means you have changed your point mid-conversation, which is not a valid way of debating anything.
We might be able to get top-spec socketed RAM again once CAMM2 gets enough penetration, or this could just be the way things are in the future.
Alex/Carson -- you've been implicitly pushing the idea that "complete" is a worthy goal, and this finally makes it explicit. Software driven by a cohesive philosophy always shines, and htmx is no exception. I'm very appreciative of what you are doing, along with the rest of the htmx contributors.
any other examples of such software available in public?
sqlite?
I don't know if it was ever made formal policy, but I seem to remember that at one point a lead maintainer of RSpec opined that version 3 was basically the final form of the project. Upgrades have been pleasantly unexciting for about a decade now.
[Comment removed by author]
[Comment removed by author]
It's not that they never add features. It's that, at least since sqlite 3 (so for the past 20 years):
I think sqlite is very much an example of software driven by the cohesive philosophy that @jjude asked about. As @adriano said, it's not necessarily feature complete, but features are added very carefully and deliberately. There aren't many things I'm as confident using as I am in sqlite. It makes me happy that htmx (another thing I like a lot) aspires to that, but it's got to keep going a long time to prove it's in the same league. (I suspect it will.)
I'm not sure whether sqlite has a cohesive philosophy, but note that @jjude's question, as I understand it, is about software with a cohesive philosophy; not necessarily software with feature completeness as its philosophy.
If I were to guess what the sqlite authors' philosophy might be, it's that the world needs a high quality SQL implementation that remains in the public domain.
TeX - Knuth said many many times that it was feature-complete.
And yet the default configuration flames anyone daring to have a non-ASCII character in their own name.
Stability also means that defaults in many cases can't be changed, otherwise you could break existing users.
Thus highlighting the issues with "feature-complete for stability's sake".
Things change. It is the one constant. If the software is static in a sense, then rolling a new major version or forking with this kind of "fix" is both reasonable and necessary for the long term needs.
From the answers it seems the boring tech would be: go + htmx + sqlite.
Thanks.
Common Lisp, I think. If I were starting a fresh Web project I'd look to Common Lisp + htmx. Probably SBCL.
Emacs. You can often find a package written (and forgotten) easily over ten years ago and it’s highly likely it will just work.
Go. Python used to be like that too before it got huge.
I also have rosy memories of the last good Python in my mind (2.5), but realistically, it was always accreting features at pretty fast rate. It's just a lot of people got stuck on 2.7 for a long time and didn't observe it.
You have to go back further than that IMO. Pre-2.0 was when you could argue that Python was more or less adhering to its own Zen. To me the addition of list comprehensions marks the time when that ship has set sail.
I would say the "unix philosophy" is the most central one, guiding hundreds of terminal apps in the POSIX standard set and beyond.
Common Lisp, perhaps?
Hare, although it's not finished yet. https://harelang.org/blog/2023-11-08-100-year-language/
Didn't Paul Graham once say that about his lisp?
Good content, but typography is "killing" me. Specifically the choice of Roslindale as the main text typeface. Please consider using something that breathes a little more since even tweaking line-height is not enough.
My brain cannot process that, I just see a wall of text. If it wasn't for reader mode I would have bailed.
Croat here, in practice we write them as separate characters dž instead of dž, etc. This makes correct aphabetical sorting like d -> dž -> đ -> e difficult to implement (and usually we don't bother).
A curious aside: this browser (Firefox on Android) is rendering the digraph with tighter kerning than the separated characters too.
This is the right decision and it has nothing to do with "US law" as some of the lwn people seem to be talking about. Russia is a dictatorship with sophisticated state-powered cyberwarfare capabilities. Regardless of whether a Russian-based maintainer has malicious intent towards the Linux kernel, it's beyond delusional to think that the Russian government isn't aware of their status as kernel developers or would hesitate to force them to abuse their position if it was of strategic value to the Russian leadership. Frankly it's a kindness to remove them from that sort of position and remove that risk to their personal safety.
It may or may not have been the right decision, but it was definitely the wrong way to go about it. At the very least there should have been an announcement and a reason provided. And thanks for their service so far. Not this cloak and dagger crap.
Indeed this was quite the inhumane way to let maintainers with hundreds of contributions go, this reply on the ML phrases it pretty well:
Edit: There is another reply now with more details on which maintainers were removed, i.e. people whose employer is subject to an OFAC sanctions program - with a link to a list of specific companies.
This would be a completely inappropriate response because it mischaracterizes the situation at hand: if the maintainers want to continue working on Linux, they only have to quit their jobs at companies producing weapons and parts used to kill Ukrainian children. It has nothing to do with the world being (in)sane, and everything to do with sanctions levied against companies complicit in mass murder.
it has everything to do with sanity or lack thereof, when such a standard is applied so unevenly
Yes, the decision is reasonable whether or not it is right, but the communication and framing is terrible. "Sorry, but we're forced to remove you due to US law and/or executive orders. Thanks for your past contributions" would have been the better approach.
This is true of quite a few governments, including those you think are friendly, and it is a huge blind spot to believe otherwise. Dictatorship doesn’t have anything to do with it, it isn’t as though these decisions are made right at the top.
Dictator, you say? I chuckled. Linus is literally a "BDFL".
Maybe we'll eventually see an official BRICS fork of the Linux kernel? Pretty sure China has been working on it.
Do you have the same reaction to contributions from US-based companies that have military contracts? While the US isn't a dictatorship, the security and foreign policy apparatuses are very distant from democratic feedback.
[Comment removed by author]
much more distant than russia's in fact
It's hard to single out Russia for this in a post-Snowden world. Not to mention that if maintainers can be forced to do something nefarious, then they can do the same thing of their own will or for their own benefit.
Did you hear this from the affected parties?
The Wikimedia Foundation has taken similar action by removing Wikipedia administrators from e.g. Iran as a protective measure (sorry, don't have links offhand), but even if that's the reason, the Linux actions seem to have a major lack of compassion for the people affected.
[Comment removed by moderator pushcx: Don't try to make your point by bringing in every current extreme political event.]
It wasn't xenophobia. The maintainers who were removed all worked for companies on a list of companies that US organizations and/or EU organizations are prohibited from "trading" with.
The message could have (and should have) been wrapped in a kinder envelope, but the rationale for the action was beyond the control of Linus & co.
Thank you for the explanation, makes sense as is common and compatible with sanctions to other countries. I was replying to the comment above mostly.
[Comment removed by moderator pushcx: Adding Nazis into the discussion didn't improve it. Don't do this.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
[Comment removed by moderator pushcx: Pruning thread from thardin's political trolling.]
This was what Hangton Chen has to say about this:
Hi James,
Here’s what Linus has said, and it’s more than just “sanction.”
Moreover, we have to remove any maintainers who come from the following countries or regions, as they are listed in Countries of Particular Concern and are subject to impending sanctions:
Burma, People’s Republic of China, Cuba, Eritrea, Iran, the Democratic People’s Republic of Korea, Nicaragua, Pakistan, Russia, Saudi Arabia, Tajikistan, and Turkmenistan. Algeria, Azerbaijan, the Central African Republic, Comoros, and Vietnam. For People’s Republic of China, there are about 500 entities that are on the U.S. OFAC SDN / non-SDN lists, especially HUAWEI, which is one of the most active employers from versions 5.16 through 6.1, according to statistics. This is unacceptable, and we must take immediate action to address it, with the same reason
did you just deliberately ignore the fact that huawei is covered by special exemption in the sanctions?
[Comment removed by author]
[Comment removed by author]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system. Calling people delusional doesn't help anything.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system. Telling people you are laughing at them doesn't help anything.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
[Comment removed by moderator pushcx: Pruning an argument about defining Russia's political system.]
The same could be said of US contributors to Linux, even moreso considering the existence of National security letters. The US is also a far more powerful dictatorship than the Russian Federation, and is currently aiding at least two genocides.
The Linux Foundation should consider moving its seat to a country with more Free Software friendly legislation, like Iceland.
I'm Icelandic and regret I only have two eyebrows to raise at that.
it's an incredibly low bar that Iceland has to clear, as this story demonstrates
Please expand on how Iceland would act to be seen as a more FLOSS friendly place, as opposed to for example the United States.
not mandating the removal of maintainers
In other words, refusing to comply with international sanctions. This is in fact an incredibly high bar to clear for Iceland. It would require the country to dissociate itself from the Nordic Council, the EEA, and NATO.
a kernel dev quoted in the Phoronix article wrote:
that made me think it was due to US (not international) sanctions and that the demand was made by a US body without international jurisdiction. what am I missing?
Without a citation of which sanction they're referencing it's really hard to say. I assumed this sanction regime was one shared by the US and the EU, and that Iceland would follow as a member of NATO and the EEA. If it is specific to the US, like their continued boneheaded sanctions against Cuba, than basing the Linux foundation in another country would prevent this specific instance (a number of email addresses removed from a largely ceremonial text file in an open source project) from happening again.
Note however that Icelandic law might impose other restrictions on the foundation's work. The status of taxation as a non-profit is probably different.
even if it has to do with international sanctions, their interpretation and enforcement seems to have been particular to the US. it reeks of "national security" with all the jackbootery that comes with it.
[Comment removed by author]
TIL I share the birthday with lobsters. :)
Happy birthday to you as well!
You're... 12?
I hope not, they would be banned per the privacy policy.
Happy birthday to you, and to lobsters!
Alles Gute zum Geburtstag!
Tsoding is fun and informative, does coding for fun - best kind of coding.
A bit of an odd and pedantic remark, but their website (and ayllu itself) defaults to an arrow cursor rather than an I-beam, which strikes me as kind of odd. It gives off a texture that doesn't quite feel right for browser content.
I like the performance orientedness though. It loads really fast. I clicked the little preview on the main site and was surprised I got sent to an actual instance rather than just an enlarged screenshot. Another good example showing you can make fast websites if you want to.
Log is a bit slow, taking over 2s on my end. https://ayllu-forge.org/projects/ayllu/log Room for improvement.
By default Ayllu will traverse the last 100 commit messages while most forges only return somewhere between 5-20. The server also runs on a tiny Intel Atom host circa 2012 in the Netherlands which probably contributes to some of the slowness.
Then it actually sounds decent. :) Good work.
I appreciate the feedback here! I agree the UI on Ayllu could definitely use some work in terms of accessibility and general design, it's not an area I'm particularly strong in but I'm trying to make incremental improvements on it. Any patches in this regard would be greatly appreciated!
As an aside-aside, I really liked that post on digital textures; gets at something that I think many of us will have felt but not known how to put words to, or even where the boundaries of the concept lie.
The frontpage renders extremely slowly in firefox on my laptop. Mousover effects take around a second to render.
I found it strange that it does not have cond. Only case. So I used this hacky construct instead: https://git.sr.ht/~ihabunek/aoc2022/tree/master/item/src/aoc2022/day02.gleam#L100
I never needed
condin Erlang, and I rarely use it in Elixir. For that snippet I'd likely end up doing this (not sure if I got the tuple pattern matching syntax correct for Gleam):The only downside is both expressions are always evaluated, whereas in that gleam snippet only one expression would be evaluated at a time. But these equality expressions are so trivial I doubt it impacts performance.
Maybe?
(Have never written any Gleam before so this might be syntactically wrong.)
That's equivalent, but I think the parent commenter's point was that writing it your way was less ergonomic for them. Which I can agree with. You could do something gross like
to avoid casing on
Nil, but it feels kludgier to me.