I've kept an eye on Darktable for a while. I do high volume photography for high school sports so while Lightroom Classic (LrC) is the final editing tool before exporting the images that go to the clients a more critical app for me is Photo Mechanic (PM) to cull down from 6,000 images to the 300 keepers from an event (a process that only takes about ten minutes on PM). The last time I tried to run Photo Mechanic on Linux (using Wine) I was able to get it to load but it crashed hard every time I tried to actually use it. If there's a capable competitor to Photo Mechanic for Linux I might take a serious look again at switching. Then again, the speed from the Apple Silicon CPU+RAM+GPU processors means that even if I could get comparable hardware for Linux it wouldn't be any less expensive than a new MacBook Pro M5 Max. Definitely keeping an eye on this space but it's not where I need it to be for my workflow.
> It is forked on Darktable 4.0, and is compatible with editing histories produced with Darktable 4.0 and earlier. It is not compatible with Darktable 4.2 and later and will not be, since 4.2 introduces irresponsible choices that will be the burden of those who commited them to maintain, and 4.4 will be even worse.
As a 25 year Linux user (for work and at home), I've been experimenting with FreeBSD in the last year or so and I've found its simplicity refreshing. Maybe I'm swimming against the current, but I'm sure there are dozens of us!
Same here, using Linux since the beginning (1993) but slowly migrating machines to FreeBSD (some to OpenBSD) as Linux slowly becomes ever more like windows which is exactly the opposite of what I want.
As an example, I had a headless devuan instance that took an extra ~50s to ssh to today, repeatably. Checked to see if it was DNS -- wasn't DNS. Checked to see if it was misconfigured ssh -- wasn't misconfigured ssh. Eventually it turned out to be that /etc/pam.md/common-session was misconfigured by installation default to have 'pam_elogind.so' set up, which was trying to do some cockamamie dbus bullshit to communicate with some linux desktop nonsense that wasn't even installed, and there were a bunch of extremely poorly configured timeouts (!) (!!!) which eventually caused ssh to hang.
Each of these components was obviously written by some deeply incompetent junior developer at IBM working on a jira ticket as part of the continuing effort to slather enough janky nonsense on top of linux that it might maybe behave one day enough like windows 95 to be usable as a desktop environment by normal people. And then the default was set by some deeply incompetent environment package maintainer and accepted by some deeply incompetent debian committee.
This lumbering corporate enshittification of what at the core used to be a simple and comprehensible system is why things like freebsd and alpine (and, to a certain extent until today, devuan) are a breath of fresh air to use. When the system is not being actively undermined by a bunch of new grads with jira tickets and no understanding of the entirety of the system, it's amazing what you can get done.
Your 1 user issue is not proof of anything that follows, and most of the comment looks more like getting some frustration out in an aggressive and unpleasant manner.
Just to be clear, you’re not entirely wrong. Linux is not about being simple and comprehensible anymore because that only serves a small subset of users. As the userbase grew, the needs and demands grew with it and so did the solution. But your problem isn’t a sign of “becoming shit” more than finding some hardware not fully supported by *BSD or middling performance optimizations justify some rant about its developers.
Just for the sake of it I whipped up a Devuan VM, SSH worked as expected.
It is becoming shit. There is no need for ssh to integrate with logind and even less reason to interact with dbus or any desktop components. Unix philosophy was there for a reason.
It’s becoming too mainstream is what they mean. Any other not immediately obvious parallel to Windows would have warranted an explanation from the get go.
Not too mainstream, and i still use Linux in a ton of places. Mostly freebsd is simpler and more consistent; having the whole operating system in a single place means that the documentation is better and I don't have to worry if some setting is in /etc or in some database somewhere or in one config language vs another.
I still really like Linux, but I think it's healthy to have multiple viable open source operating systems.
Linux on servers has been very mainstream for last like 20 years.
Big distros like Ubuntu become ever more complicated though, they run a ton of stuff by default, like Windows. OpenBSD is, first of all, very small in comparison. Then, it's obscure enough that most online attackers do not target it.
There might be a correlation with mainstreaming, but that's not the root cause. I'm happy to use a mainstream OS if it is awesome.
A full honest answer is too long for a HN comment, could write a long article about it but won't do that.
I need an OS which is fully comprehensible and fully configurable to my exact taste and fully diagnosable at all levels. So basically the exact opposite of windows which is a massive blob of obscurity. Mac OS has been great for a couple decades, but the enshittification accelerates. As a tiny example, dtruss doesn't even work on newer Mac OS. That kind of nonsense is unacceptable.
An OS is supposed to be a platform & toolkit to enable me to use and diagnose the computer any way I need. Linux is still that, sure, but enshittification is creeping in. Wayland/systemd/gnome/etc is all movement in the wrong direction.
Try e.g. Void Linux, it's built by BSD-inclined people and is also refreshingly simpler than most big distros. (It's my daily driver for the last 8 years.)
But if you want tighter integration of userland and kernel, I suppose what RedHat is doing is closer to that. They pursue a non-Unix way though: an immutable distro, ideally only running systemd and podman in the userland, put everything into containers.
As someone who very reluctantly fits that description, yes but no. I don't love FreeBSD any less or want to use it any less than anytime in the last 20 years, nor do I like Linux any more than at any time in the last 20 years (less, if anything). But NixOS is the "killer app" that has me on Linux instead of FreeBSD. If either FreeBSD gets a really good NixOS-like setup, or Linux-based NixOS becomes unmaintained, I'm happily going back to FreeBSD again, it has always been excellent and stable to me.
No. As enshittification encroaches onto Linux the user base is moving the other way. To their benefit, I might add. With ZFS on root in FreeBSD it's a no-brainer.
Linux is feeling more and more like a bunch of random tools thrown together as opposed to a complete OS designed to work as a whole.
ZFS works fine on root on Linux. I use it on multiple machines.
> Linux is feeling more and more like a bunch of random tools thrown together as opposed to a complete OS designed to work as a whole.
This has quite literally always been the case and seems intentional. Linux is the kernel only, others supply the userland of their choice (aka a Linux distribution).
But seriously, if one counts macOS and iOS as FreeBSD users, there are more than ever. Of course that means counting Android and Steam as Linux OSes, in which case Linux users still greatly outnumber FreeBSD users.
FreeBSD's license means it shows up in a lot of unexpected places -- the last two Sony Playstations run a FreeBSD derived OS for instance. It's around, more then you think, but it's very quiet...
One of the reasons it’s very quiet is that you can only do what the company that provided it to you allows you to do with it. You can’t reinstall a fork of PlayStation OS, for instance. Sony won’t provide you the sources and the changes they made. If it was a GPLv3 OS, they would provide everything you need to build your own PlayStation OS.
Would that have been a bad thing ? Between a proprietary OS on which we cannot do anything and an opensource-based OS on which we cannot do anything .. I'd argue that making the company paying the bill would be more healthy
1) People writing BSD licensed software want people to use it, and don't really care about any reciprocal contribution (or else they would write copyleft-licensed software)
2) "the company" never pays the bill. It is always passed to their customers in one way or another.
If we think a bit more, that may help the customers: companies like Sony would have to choice between either implement their own OS (increased cost, increase end product prices) or use opensource software
The topic would become a competitive (dis)advantage, not just a random topic for legal people
My take is that BSD-like licences are bad for humankind, but as a short-sighted person with no impact whatsover, it does not matter
> don't really care about any reciprocal contribution (or else they would write copyleft-licensed software)
Also don’t care about their users having the same rights they enjoyed getting software written by someone else (for free!) and making some changes before their customers could pay to use it.
And if not for the rise of AI it's possible that WASM as a machine-level compilation target for all languages might have happened. As much as Gary predicted he didn't see AI coming.
I believe the idea is that you don't care what language is being used if you aren't going to look at it anyway. Given that premise, the AI can write JavaScript instead of something you need to compile separately.
WASM is hard enough to use that I would say the number of people using it just because they don't like JS/TS is fairly small.
At least that's always been my impression. If you're willing to go through the effort to use it, it's probably because you need the performance gains, and AI hasn't changed that.
Not really: ASM.js became WASM. What killed the possibility of WASM being The One Way to run everything is AI... the one wildcard that Gard Bernhardt didn't predict.
Not op, but I only know it as the 32-bit single-threaded binary replacement for JavaScript that can't interact with anything naturally yet people get all excited about when they manage to turn their LLVM slop output into it.
Maybe we've been reading different HN submissions.
I re-watch that presentation two or three times a year because it's a great example of how to give a presentation, how to structure your slide deck to complement your presentation, and a surprisingly educational tour of the permission rings architecture of operating systems.
And at some point we're going to have a period or war and our psychological attachments to old programming paradigms will be released so that we can move on to a more advanced way of doing things (but that won't stop your bank from running YavaScript for at least another 85 years).
Why though? That could be true of any economic condition. Imagine if any time there was a race or competition in which one group is winning you had to just say "for now".
Michael Phelps is winning the race! ... for now
China is winning the EV race ... for now
It doesn't seem to add value to me, aside from being an opportunity to, as is the time-honored tradition of the haters, to sow doubt and create negative energy to anything related to American success.
Of course US has a huge head start, but if AI keeps growing, what matters is how the market's gonna look like years from now.
Most of my clients using AI in the business workflows (in products) use Chinese LLMs, because after benchmarking for a specific use case you nearly always end up finding that you pay half or a tenth.
That's not a new phenomenon. I've adapted Gemini Flash 2.5 years and years ago when people were dissing it as "crap", yet it was the best budget and quality fit for the task I had at hand back then (translating and summarizing tons of documents). It was both faster and around 100 times cheaper than the best GPT 4 model available.
Needless to say, medium-sized Chinese models are far better than those LLMs and a perfect fit for countless applications.
Even with all the Qwens and Kimis and GLMs etc, the latest Gemini Flash models are still an insane value! I recently settled on 3.1 Flash Lite after testing basically everything on offer on OpenRouter, and it was not a close call - cheaper, faster, and better (for translation and visual understanding tasks).
