Right now is it not worthy to build a business selling desktop software to Linux, unless we are talking about enterprises, movie studios and a few other niche markets that migrated from UNIX workstations to Linux desktops.
If you want to make honest money with software then you need to solve a problem I have and help me do more work with less effort, for such a software most people will pay.
If you want to game, I recommend dedicating a separate Windows machine to the task.
As a result, games are getting nicely filtered /proc.
The gist is to prepare the directory structure, icon, .desktop file, flatpak metadata, appstream info, add dependencies not provided by flatpak runtime (for games, libGLU for example), put everything into ostree repo (flatpak build-export ...), and then either install from that, or export the content into .flatpak file (flatpak build-bundle ...).
The last step can take a while, .flatpak files are lzma compressed and the compression during export uses only one cpu core. For 20GB game, that isn't exactly quick, to be diplomatic.
As a fallback if some new game appears and I really want to play it I can get it on my kids console.
I'm happy to pay for lots of things. But I'm not interested in paying for crappy software. I wouldn't, for instance, pay for Microsoft Word.
It amazes me how little people value other people's work on this site.
An office document format is basically source code.
That's why the openness of a graphic format doesn't matter, but clear specifications on how to interpret an office document make an enormous difference.
I would seriously fork money over if MS would ever build Linux version.
I have never lost a document in LibreOffice, Vim/gvim, AbiWord, WordPerfect (long time ago), or OOO.
I lost a ton of documents, sometimes to the cost of many thousands of dollars, in Word. (Disclaimer: haven't used word since like 2010 or so)
People get burned, and thus jaded, by lots of different, sometimes opposing or contradictory, experiences.
As a lawyer, I deal with other people's Word products all day every day. What Word features am I missing out on? I am genuinely curious and not trying to be snarky in any way.
All those pesky macros that would allow your computer to be infected
Not seen Word in a while I guess.
I cringe every time I have to deal with a Word file.
People do say this though. People have personal preferences for the software they have to use all day, and most of the time people don't agree on what that software is. And most of those decisions are driven by the type of work people do. I don't expect a lawyer or a business analyst to use command line interfaces, and I don't think they expect software engineers to code in Word...
And the other people, who were discussing the command line, ignore them.
The point is that you can't really criticize something when your criticism seems to boil down to "I don't understand this".
And who said anything about writing code in Word? What are you on about?!
For the last year Word has actually supported LaTeX for math equations, see https://blogs.msdn.microsoft.com/murrays/2017/07/30/latex-ma...
But aside from that, Word is great for short and medium length documents, especially ones that need group editing. Put a Word file up on Sharepoint, and now there is user friendly change tracking and commenting.
I've written multiple technical specs in Word, it serves as a great way to collaborate, so long as everyone has bought into the ecosystem.
Well that's the problem isn't it, every potential collaborator who can't or won't buy it will be shunned.
The same argument can be made for any piece of software. People have to buy in, even free software requires taking time to learning it vs whatever existing workflow is in use (if any).
OP points out LaTeX, but for non-technical and non-academic people, this is a non-starter. Word has an extremely low barrier to entry so it makes tons of sense for the general public and K-12 education markets.
There's a reason it's so popular and why LibreOffice and GSuite have emulated the bulk of it's feature set.
I know another example that goes in opposite direction, of Humble Bundles, where Linux had consistently higher average payment than Windows, but that may have changed in recent years.
> Again, we've done the research. The profits aren't there -- very few Linux users are willing to pay for commercial software. And the cost of entry is still high because of the fragmented Linux landscape. The Linux world has to change before commercial software will have reason to invest in Linux ports. And we haven't seen much real change in the Linux market in several years.
https://forums.adobe.com/thread/487814
> At the height of the Dot-Com era, dozens of startups tried selling proprietary applications for Linux -- and not one survived more than a couple of years. Even Adobe, after releasing a popular beta of FrameMaker, withdrew it from circulation after a poll suggested that users were unwilling to pay for applications.
http://www.linux-magazine.com/Online/Blogs/Off-the-Beat-Bruc...
It wasn't perfect software by any means, but perfect is the enemy of good anyways.
But, lot of software I need to use doesn't need to be that high level and for what I need plenty of high quality open source options exist. I have no problem paying for software, whatever os I use, if it's the software I need. But if an open source alternative with comparable or sometimes better features exists why wouldn't I use it?
And games honestly, There are plenty I would spend money on if they supported linux and i've purchased ones that have interested me.
I used open source software on Windows, and I carried that habit over when I switched to Linux. I have only ever paid for games on Linux, because everything I have needed has a satisfactory FOSS alternative.
Hell, I bought a macbook a couple years ago, and I have never bought a piece of software for that either.
It's not because of the OS.
My gut sense is that open-source/paid-support model is an ideal, which also provides an incentive for the publisher to produce as good of software as possible. I don't know if Red Hat or Canonical are profitable doing that, but it would be nice if that were the way that everything could work (maybe, I haven't really thought that much about it and am open to rebuttal).
I am averse to paying up front for something that I don't really know whether it'll be what I want and won't just be a pile of shit.
I don't buy many phone apps either, for that reason.
Having said that, I haven't really observed a difference in Linux users willingness to pay for large functional applications. I don't think there are objections to paying for IntelliJ or Photoshop grade applications in general.
Ubuntu seems to be trying to make a paid appstore/proprietary appstore using Snap, which could be good, could be bad. We'll see.
I think the way the software economy works today has a lot of malincentives. I don't think simply transplanting that same economy to linux would solve any of the problems.
Selling Free Software doesn't tend to work very well, since every customer is free to resell it and undercut the developer. Still, I do spend a lot of money on Free Software since I have a bunch of automatic donations to affiliated organisations (FSF, WikiMedia and Open Rights Group in particular).
(I don't use proprietary software due to ethical concerns; see, for example, this HN submission)
Not everything is a web app these days and many people still prefer to own their software instead of being a recurring revenue source for the same functionality month after month.
I worked at MS for a long time. I can say that sort of stuff hasn't happened for quite awhile, and people involved in it happening were fired, some of them rather publicly.
Now days, expense reports are gone over very carefully. Everyone at Microsoft attends yearly ethics training, which covers all manner of topics, including what is and is not acceptable to do to win sales.
Realistically, when it comes to cloud based productivity systems, the choices are Microsoft or Google.
> then the top executives with no technical knowledge are happy to sign multi-year contracts that lock them into those vendors
Any half decent company is going to have at least a CTO in charge of these decisions. The CTO should have multiple people working for them that are capable of making engineering decisions.
If a company sends a bunch of non-technical people to a technical presentation to make a technical decision, that is the company's fault.
It must be a toxic environment, working with people who needed to be continuously trained on how to be ethical.
Good on them for trying to do something about it, however.
2016 isn't that long ago, and that's just one that was badly-judged enough to have made the news:
https://www.eurogamer.net/articles/2016-03-18-microsoft-gdc-...
The overall feeling inside MS at the time was "wow, that was stupid, hasn't that team being paying attention to how the industry has been changing?"
An even more formal policy change was then sent out, reminding people of dumb stuff they aren't supposed to do.
https://en.wikipedia.org/wiki/Microsoft_licensing_corruption...
It was nothing in-your-face like strippers, but still pretty immoral.