Hacker News Reader

Comment by Chris_Newton | Hacker News Reader

I wonder how much demonstrable interest it would really take to convince the management at Serif that a real Linux port of Affinity is a worthwhile exercise. Every time the software comes up in discussions, it feels like half the comments are just enthusiastic people hoping for a Linux-friendly version. I understand the hesitation, and surely Serif have had a conversation internally and maybe done some market research. Still, if the cost guesstimate for doing it that I saw someone apparently from Serif mentioning in a previous discussion was even close to realistic, I can’t help wondering if they could easily raise that much just in pre-orders if they ran a Kickstarter-style “Will it work?” campaign, and probably a lot of goodwill from the tech community as a bonus.

View on HN

The Linux community is a very vocal minority when in comes to the internet. While it would be nice to have available on all platforms, the reality is the majority of people are happily working on the other two major platforms and don't care.

And Linux[0] is absolutely the worst "platform" you could ever have to port commercial software to. There is a ludicrous level of fragmentation where even individual distributions break compatibility with themselves every few years, and there is no widely accepted standard for binary application distribution[1]. That's without even getting into things like driver issues that are arguably not Linux's fault. As game developers discovered, Linux will often account for a vastly disproportionate amount of your bug reports for its share of installs.

I sympathise with people who want to be able to use their favorite software on a FOSS OS, I'm right there with you, but that will always be an uphill battle on Linux because the culture is deeply anti-platform.

[0] "Linux" here is taken as shorthand for "The GNU/Linux Desktop". The kernel itself is a fine platform.

[1] unless you lump all the different package repos, formats, and managers together. I shouldn't have to explain why this is a problem here.

> There is a ludicrous level of fragmentation where even individual distributions break compatibility with themselves every few years

Even worse, there is a vocal group of users who believe that this fragmentation is a good thing, and that software developers should target multiple distributions (or be shamed for it), as opposed to doing the sane thing and just targeting Ubuntu (or some common abstraction layer).

The vocal group you speak of overlaps significantly with the group that wants the price of all software to be voluntary donations. At some point, their desires become parasitical.

While undoubtedly true, it’s also true that Linux is nowadays the default OS for VFX, for example, and this means that a lot of commercial 3d software targets it (for example, Autodesk Maya or the Substance suite). There is a standard “platform” of reference that allows companies to have a fixed target for their software: https://vfxplatform.com

I think targeting it would be probably the best solution for a company willing to porting to Linux. Alternatively, Flatpaks/Appimages provide a very nice and user friendly solution to this problem.

>While undoubtedly true, it’s also true that Linux is nowadays the default OS for VFX

That's mainly because it's cheap and VFX is like a "factory production line" style job. But VFX is also very much a niche compared to vector graphics and DTP (and even more so, compared to bitmap graphics).

In other words, if Affinity's concern is that the Linux market for their stuff if small, then the intersection of "the Linux market for their staff" that "has the standard platform for VFX" is much much smaller.

Except there is a widely accepted standard for binary application distribution: AppImage (https://appimage.org/). It's a fire-and-forget solution: You download the AppImage and run it. No need to install anything.

AppImage was created for precisely this scenario: When you want to distribute a proprietary application without worrying about packaging it for different Linux distributions.

And for the driver issues: This is being brought up now and then, but comparing the amount of issues I've had with my AMD graphics card on Linux (essentially none) [1] with the amount of anecdotes of Windows driver issues I regularly see in certain forums, I'd say driver issues are a myth by now (to a certain extent of course, please don't bring up how you had to install a firmware package for your Broadcom wireless chipset...).

[1] I'm not saying there aren't ever any driver bugs now and then, but just a food for thought: My graphics card works perfectly out of the box, without any configuration or driver installation at all, on most recent (read: not ancient) Linux versions. On Windows I'd have to look for whatever piece of ad-laden "graphics management suite" I need and then install it as well as regularly update it.

> Except there is a widely accepted standard for binary application distribution: AppImage (https://appimage.org/)

Oh, if only AppImage was widely accepted. Very few developers distribute via AppImage and basically no DEs understand what they are. Not that AppImages need them to, but it'd be nice.

> AppImage was created for precisely this scenario: When you want to distribute a proprietary application without worrying about packaging it for different Linux distributions.

Yes, and it does about as well as could be expected. Which is to say, it kinda mostly works for most distributions that have some level of sanity.

> And for the driver issues: This is being brought up now and then, but comparing the amount of issues I've had with my AMD graphics card on Linux (essentially none) [1] with the amount of anecdotes of Windows driver issues I regularly see in certain forums, I'd say driver issues are a myth by now [...]

Consider that there are an order of magnitude more Windows Desktop users than Linux Desktop users. And I know plenty of people experiencing frequent driver regressions who'd disagree with you.

> On Windows I'd have to look for whatever piece of ad-laden "graphics management suite" I need and then install it as well as regularly update it.

Incorrect. Windows quite often has a driver available via Windows Update, sans overengineered bullshit control panel. It is usually a bit behind the latest-greatest though.

> Oh, if only AppImage was widely accepted. Very few developers distribute via AppImage and basically no DEs understand what they are. Not that AppImages need them to, but it'd be nice.

To clarify: Not many FOSS developers distribute via AppImage, or not many proprietary software developers distribute via AppImage? Since the former is to be expected, but I'm curious about your experiences with the latter. To be honest, I haven't seen that many AppImages in the wild either, but I'm also not using a lot of proprietary software on my system at the moment.

> Incorrect. Windows quite often has a driver available via Windows Update, sans overengineered bullshit control panel. It is usually a bit behind the latest-greatest though.

That's good to know. Do you happen to have a document or some other resource explaining the Windows Update graphics driver distribution system? IIUC Windows a) configures some sort of "Windows Basic Display Adapter" on first boot, before b) installing basic IHV-specific drivers via Windows Update. But I understand the IHV drivers installed by Windows Update support all graphics card features, only without the control panel? I always thought there is no alternative to the separately installed IHV graphics driver, especially if you need OpenGL/Vulkan/CUDA/etc. support. But if Windows manages to provide that functionality automatically through Windows Update, that sounds like a good alternative.

They're pushing very hard to have the actual standard manufacturer drivers available via Windows Update, and shipping the control panel as an optional download via the Windows Store.

I had to do a Windows 10 reinstall recently after I broke something severe in the networking stack (entirely on me, I was playing with something I knew I shouldn't have been playing with) and I had the NV 2060 drivers back on my machine pretty much immediately as part of the update installation chain. I just had to go back into the Windows Store to redownload the control panel since it's a hybrid video laptop and I wanted more control over settings.

I wrote up an edit for my comment but then barely missed the edit deadline, so I'll have to do it this way: I realized my comments on driver issues are a little snarky, sorry for that. Please ignore that part of my comment.

Using more neutral language, my main point is that Linux drivers nowadays are better than their historically bad reputation would suggest, and that they often offer a better or at least comparable out-of-the-box experience compared to Windows drivers. An important caveat is that free software-oriented distributions often make it somewhat more difficult to enable on-free firmware or drivers.

These discussions remind me of the people asking for a small iPhone. Apple responded but it failed to find enough customers.

Linux has 1% of the desktop market and even within that small number the majority are used to getting their software for free.

I'd say the chance of a Linux version is effectively zero.

Yep, those folks engaged in "performative commenting" upvoting and agreeing that they would all buy small phones if only Apple would make them. Then Apple made the small phones... and crickets.

It takes time. With luck I shouldn’t need to buy a new phone for at least another year or two.

Er, the original 4" SE sold really well.

https://tcrn.ch/3jxVB6p

As does the current budget phone. The thing is people were regularly complaining that they couldn't get a high end device in a small package. Yet when Apple delivered that it didn't sell in large enough volumes.

https://siliconangle.com/2021/04/14/report-apple-drop-iphone...

So they sold really well, but not in large enough volumes. The only way both can be true is if it was at different times. I can believe that, maybe there was a significant market, but then after a few years it got saturated.

There is a distinction between a budget phone and a small phone. People were calling for a high end device in a small case.

And the 1st Gen SE provided that, it had the same internals as the then-current flagship phones.

Note that Apple's small form factor phones have very often been released out of sync with the main launches in the Autumn. Just because there won't be a new one this autumn doesn't mean the current version will be dropped, and doesn't mean a new small form factor phone won't come along next year, possibly early next year.

There is an SE in the current range and a mini. They are clearly targeted at different audiences.

I'll not that all the complaints I read were during the period when the SE (or equivalent) was available. People were clearly calling for something else yet when it was offered they didn't take it up.

> I wonder how much demonstrable interest it would really take to convince the management at Serif that a real Linux port of Affinity is a worthwhile exercise.

Probably more than is demonstrable. It will never be a money maker. It might still happen for other reasons.

Reply on news.ycombinator.com