GNU Hurd Hardware Support Remains In Very Rough Shape For 2018

[kpedersen]

Not directed at Michael but ... Why would anybody care about GNU Hurd?

You are on an alternative operating system hardware forums... You seem to be in the wrong place. Perhaps you should try here: https://answers.microsoft.com

I obviously joke... I wouldn't suggest anybody try to get useful information from there

[schmidtbag]

What's meant by "fanatics that give Linux a bad rep"?

I assume this is in reference to Stallmanites and neckbeards. Though they're entitled to their opinions, they're basically the extremists of the world of open-source computer software and hardware (like PETA is to vegans, NRA is to gun owners, ISIS is to Muslims, and so on). These people are often pushy with their thoughts and they demand a lot from companies, and as a result, they push away a lot of potential users or developers. It's difficult to work for FOSS when the first thing you hear about your work is why something else (whether it's directly related to your work or not) isn't yet open-sourced.

[Vistaus]

How does any project stay in pre-alpha for 28 years

I know, right? This is the complete opposite of Chrome/FF.

[Vistaus]

IOW it will never get further than research project/plaything, just like Hurd.

But at least it'll be much nicer to play with and it'll support more hardware.

[TheBlackCat]

I know, right? This is the complete opposite of Chrome/FF.

Obligatory xkcd (check the Alt text):

[Nelson]

Why not? A FOSS microkernel developed by the FSF. It has a ton of problems, but at the very least it's interesting.

A real multiserver microkernel OS is an interesting project. HURD uses Mach though, it wasn't developed by FSF and historically that has probably been part of the problem. They've attempted to use L4, there was talk of building their own (viengoos) They've picked perhaps the most ambitious design, it's not clear that all the problems are solvable.

They might be better served if they had a fairly modern system running that didn't satisfy all the desires. As it stands there it's really hard to get involved, there isn't enough stuff running to let you hack on the more simple bits. Maybe running an entire linux as a server like l4linux might lower the bar enough that people could get involved.

[azdaha]

I assume this is in reference to Stallmanites and neckbeards. Though they're entitled to their opinions, they're basically the extremists of the world of open-source computer software and hardware (like PETA is to vegans, NRA is to gun owners, ISIS is to Muslims, and so on). These people are often pushy with their thoughts and they demand a lot from companies, and as a result, they push away a lot of potential users or developers. It's difficult to work for FOSS when the first thing you hear about your work is why something else (whether it's directly related to your work or not) isn't yet open-sourced.

Thank you. I appreciate the lengthy response. FWIW, I would argue that it's possible to agree with Stallman on many things and at the same time accept reality and work towards that ideal that he and others like him profess. Once again, as I've stated here a couple of times, IMHO the more time passes the more "sane", resonant, or relevant ideas and statements from Stallman(ites) become. Whether it's a warning of privacy erosion through ubiquitos technology geared towards nothing else but profit, or even the idea of OpenSource software/development. BTW, I am aware of the naming and history of the aforementioned; nonetheless, I'm using that as an argument to show that, if anything, the world has moved closer toward Stallman rather than the opposite trend. It, obviously, has a lot to do with less restrictive licenses as well, which foster an environment that allows OpenSource and Proprietary to coexist. Still, at the risk of being labeled as a "fanatic that gives Linux a bad rep", I thought it's worth pointing out and maybe even provide a little more context and perspective.

[schmidtbag]

FWIW, I would argue that it's possible to agree with Stallman on many things and at the same time accept reality and work towards that ideal that he and others like him profess. Once again, as I've stated here a couple of times, IMHO the more time passes the more "sane", resonant, or relevant ideas and statements from Stallman(ites) become. Whether it's a warning of privacy erosion through ubiquitos technology geared towards nothing else but profit, or even the idea of OpenSource software/development. BTW, I am aware of the naming and history of the aforementioned; nonetheless, I'm using that as an argument to show that, if anything, the world has moved closer toward Stallman rather than the opposite trend. It, obviously, has a lot to do with less restrictive licenses as well, which foster an environment that allows OpenSource and Proprietary to coexist. Still, at the risk of being labeled as a "fanatic that gives Linux a bad rep", I thought it's worth pointing out and maybe even provide a little more context and perspective.

I think most of us feel the same way. But some people take what he says a little too far and have no concept of how following his philosophy isn't sustainable, and is inefficient in terms of evolution. Case in point: Hurd. Despite being older with a more concrete goal in mind, there's a reason why Hurd never succeeded whereas Linux did: Linux was more friendly toward proprietary developments. Linus himself stated that he didn't care if a company profited off of Linux. He didn't seem to care about whether or not the hardware or firmware of the platforms he worked on were open source, and he was fine with people submitting code for commercial reasons. All he cared about was as long as his kernel remained open source. But, that's not enough for Stallmanites - it has to be everything.

It's important to remember that we would not be where we are today if it wasn't for proprietary technology, closed-source efforts, and commercial involvement; we would be very far behind. I'm not suggesting these things are still needed, because that is debatable. But, it was a necessity in order to get things kickstarted. Back in the 80s and 90s, the Internet either wasn't a thing or wasn't a practical tool for community development. There was a demand for certain hardware and software, and there's no way the average person was going to devote their free time on multi-thousand dollar hardware to make these things. Because of how expensive the hardware was, professional attention to detail was crucial. So, companies had to hire trained developers to get the job done. The computer industry became profitable, and as a result, competition thrived. Everyone was trying to one-up each other, and doing so wouldn't be possible if everyone had access to all the information from all competitors. Sure, you could argue "it'd be more productive if all minds could work together on improving a design" but that's not how life works. Who is going to fund testing it? How to you manage so many contributors? Who do you call when things go wrong? How do you set deadlines? These are all faults of open-source. Note how in the Linux world, the answers to these questions involve companies like Red Hat, which is non-free. Open source works fine for things that are very segmented (such as a kernel or a desktop environment) but it doesn't work well on big projects that aren't so modular (which is why we don't have any good photo editors, for example).

[azdaha]

(which is why we don't have any good photo editors, for example).

Excellent post all around. Although, I would argue that GIMP is great as a photo editor, and perhaps change that statement to which is why we don't have any good video editors instead; correct me if I'm missing something, though.

[schmidtbag]

Excellent post all around. Although, I would argue that GIMP is great as a photo editor, and perhaps change that statement to which is why we don't have any good video editors instead; correct me if I'm missing something, though.

I actually think it's the opposite. Linux has a few pretty decent video editors, with KDEnlive, Blender, Lightworks, and Cinelerra being a few examples. GIMP is an adequate photo editor, borderline good, but nothing more. GIMP is really starting to show its age, with its outdated APIs/ABIs, lack of hardware acceleration, and missing features found in a lot of commercial applications. Considering how few people are left working on GIMP, it's fantastic. But, it needs an overhaul.

That being said, I'm not a professional photo editor, so GIMP has been plenty sufficient for my needs. But, pretty much every single artist and graphic designer I've talked to who has given GIMP a shot was not satisfied.