gaazoh

If anything, Ladybird is an independent implementation of the web standards, and the devs have identified and helped solving quite a few bugs and and ambiguities in the standards, which benefits everyone, from browser devs including the big guns to web developpers and users.

Very cool, I implemented this once in python, it's a fun exercise, and knowledge that is gradually disappearing (modern phones with 12-key physical keyboards usually don't even have a T9 implementation, and when they do it doesn't perform well, even higher-tech KaiOS phones).

Although I appreciate the effort, I see a couple of issues with this implementation:

* The demo doesn't seem to work properly, the first thing I tried to type was "hello world", but it didn't recognize "hello" and I got "43556" instead.

* The word list is generated generated C code, which makes it hard to use other dictionaries (languages) or to add words during use (you can't add all place and people names to the list, but people are going to want to reference a handful of them many times). Loading from and appending to a plain text word list would make more sense, and maybe additionally use a custom binary format for the trie structure for fast loading into memory once a word list is imported on first use (hardware that would benefit from T9 might not be fast enough for conversion to be "instant")

* Non-latin script support would be nice. Although I have no knowledge whether Greek or Cyrillic languages used a T9 mechanism, it would be a minor change to define. Korean 12-key typing is also very cool, but I don't know whether that counts as T9.

I tried using a Nokia 6300 4G for a while a few years ago, because I don't use phones for much else than phone, SMS and a bit of Wikipedia.

I had a terrible experience, nothing worked correctly, basic, fundamental apps such as phone, contact, alarms, had lag spikes of over a second for any action and randomly crashed, it had no CJK support (kind of a big deal for me, and like 20% of the world)...

Some apps worked remarkably well, but having Youtube work better than messages is kind of pointless given the sVGA screen size. And the app ecosystem was basically dead. I tried developing a bit for it (apps are a simple html+css+js package), but at the time publishing to their app store required integration of their advertising solution, which absolutely sucked (in 6 month of use, I was never served any other ad than "hot MILFs in my area"), which meant that neither commercial nor OSS apps would target the platfom. IIRC side-loading sort-of worked depending on phones, which means it doesn't work for the average user.

Oh and sometimes the Kai Store app would send advertisement through push notifications.

I loved KaiOS on paper, but using it was a chore. A 2010-era "dumb" phone could do most of what a KaiOS phone can, but better, except for connecting to 3g/4g networks, and older networks are deprecated in many regions, which means they're not an option anymore.

Maybe they got better over time, but what a huge disappointment that was.

OEMs have been doing basically this for years with their phones for decades at this point, pushing customized builds of Android with every phone they make, this has been successful to close the gap Apple created when they released the iPhone.

I guess a hurdle smartphones didn't have as they were breaking into a new market is compatibility; outside of the tech world, virtually all of corporate and personal environment is dependent on Windows and Windows-only software. Steam has shown it can work with SteamOS and Proton, making gaming on Linux a reality for a wide audience. What's missing is a major OEM to build a high-spec laptop with a custom Linux build to optimize performance and battery life, with a decent Windows compatibility layer and that would provide software companies an incentive to sell native Linux versions and support. Is Samsung really going to keep their laptop line depend on Windows, and leave it on the side-line as they will never be able to really optimize battery life and performance and compare to the MacBooks?

Mostly DDG, but that's beside the point. Kagi seems to be marketed at the general public, for whom FAANG companies control the narrative. Even though they are obviously bad actors wrt privacy and UX dark patterns, they claim otherwise, that they value privacy and strive for the best user experience, and having a startup just claim that they do better, but offers no hard guarantee and require a payed sign-up to actually try it out with a pricing incomprehensible to most of the world shows that progress can be made. From afar, it looks like an interesting and good product, but I'm just not going to bite the bullet just yet.

Kagi has popped up a couple of times here recently and looks interesting, but there are a few things keeping me from actually trying it out

* I don't trust the product's claims. Sure, privacy and user-centered results sound cool, but literally every company on the internet claims to cater to the user and value their privacy. Kagi can apparently afford to be more specific than usual, but how binding is that? I don't know, I'm not a lawyer and definitely not versed in US/California law, and given all the obviously exaggerated claims in this domain by all kind of actors, I can't give it much credit. I guess Kagi has to pay for the whole industry's decades of malpractices in this regard and that sucks, but I guess you could do better if you opened more about your

* I don't trust the product's ability to stay around. Startups come and go, and I'm not subscribing to a paid service and switching workflow without a reasonably solid belief that I won't have to do it again in a near future. Your new pricing policy actually helps quit a bit in this regard, the other bit requires you to actually stand the test of time, so just keep on doing your best I guess.

* Pricing has is shown excluding taxes. I'm not going to figure out the US tax system just to know how much I actually to shell out, and I'm not paying if I don't know how much. In Europe, VAT is around 20%, so it's a pretty significant figure, that would be 60 bucks a year for the Ultimate plan. I don't have the slightest idea if that's the order of magnitude expected in California. Have your lawyer or accountant figure it out, because I sure as hell am not. Allowing me to pay in euros would also be a quite large hurdle removed, for similar reasons: exchange rates fluctuate, banking operation costs fluctuate, and even if I can work it out more easily than US taxes, I'm not going to do because this should be your job, and whatever figure I work out will be obsolete by the next time I'm billed.

Not only that, but underground infrastructure and surrounding buildings put a high constraint on pavement design by putting a hard limit the total thickness of the pavement: can't build too deep or you'll disrupt other infrastructure, can't build too high or the road will be higher than surrounding building entrances or sidewalks.

Interstate construction don't have such limits are typically half a meter or more, not counting foundation earthworks, which can easily double that figure. In cities where telecom networks are 60cm deep and gas and electric networks 80cm deep, you just don't have the luxury of designing a meter-thick pavement that will have a decent IRI for decades to come.

Google has a near monopoly on web search, and they are very aggressively doing everything they can to keep it that way, both by tightly integrating it in their products (browser, browser engine, and OSs, mainly) and funding competitors (Safari and Firefox main revenue source are contracts with Google, under the condition of making Google the default search engine).

Sure, you're free to use DDG and find a workflow, but people are less likely to remember it, especially as they overwhelmingly use Google.

Acceleration and deceleration aren't relevant here, the limit would be passenger comfort either way.

Curved tracks do mean limits on top speed, as the centrifugal force needs to be kept below what would be uncomfortable for passengers. It also can cause some issues for building stations, as curved stations require a wider gap between the train and the platform. But neither is a blocking issue, all train tracks in the world do have curve, and curved stations aren't unusual in public transport systems.

The grandparent probably means disk, not ring-shaped when mentioning circles anyway. Pick any 2 point at random into a 33km² circle, the average distance will be 4km, and worst case scenario 6km. Do the same thing in a 33km², 200m-wide "line", the average distance will be 85km, with a worst case of 170km. A circular city doesn't need nearly as much raw speed for it's public transport to be more efficient than transport in a nonsensical linear city could ever be, even if you throw in ridiculously fast trains and sprinkle magical AI thinking.

It's not a coincidence that all major cities are roughly circular even though they are built around roughly linear features (navigable rivers and/or coast line). It's just what naturally works.

This is Switzerland, where land is not as plentiful as in Texas, not by a long shot.

Plus, the support structure is free in this case, it's definitely cheaper to lay solar panels on those dishes than tearing them down just to lay panels on the floor. Also, having solar panels laying flat on ground level sounds like skimming on infrastructure to save money short term but having to pay higher maintenance long term: you have to keep the panels relatively clean for them to work, having them slanted and elevated a bit means that less dirt get on top of them and rain washes most of it regularly.

Still requires hosting AFAICT. Would you know of a native Windows software alternative, perhaps? I could probably manage to host Photo Prism or Memories, but there's no way my parents would take the time to learn and set up a server, yet they have decades of pictures to sort through.

Of course it didn't, it wasn't designed to be either the fastest nor the best. Just OK and simple. Yet in some cases it's not completely overtaken by competition, and I think that's cool.

I don't believe QOI will ever have any sort of real-world practical use, but that's quite OK and I love it for it has made me and plenty of others look into binary file formats and compression and demystify it, and look further into it. I wrote a fully functional streaming codec for QOI, and it has taught me many things, and started me on other projects, either working with more complex file formats or thinking about how to improve upon QOI. I would probably never have gotten to this point if I tried the same thing starting with any other format, as they are at least an order of magnitude more complex, even for the simple ones.

The inclusion of QOI in the lossless benchmarks made me smile. It's a basically irrelevant format, that isn't supported by default by any general-public software, that aims to be just OK, not even good, yet it has a spot on one of these charts (non-photographic encoding). Neat.