You start with "I salute you. I’m a fan" and then call a lot of people "fundamentalist", then "You have made a choice based on an ideology", "blind to the realities of the world you live in".
Then "But like most utopian views, it’s naive", "It’s a pipe dream", and finally "Like all fundamentalists, they simply ignore some inconvenient truths".
Then you ramble through justifications of your business model (and quite a few logical fallacies).
For being a fan, you express your support in a strange way.
Now excuse me while I go back to writing software as a gift to whoever wants to use it.
I think one of the "better" approaches if you cannot survive with a "support/customization" model (which is certainly not for everyone) is to open-source elements of your software offering that are not business critical but still somewhat useful for other parties to live in the open. This way you do not take away your livelihood, but you still contribute back in some amount to the tech world. That's like being a good citizen.
Totally agree with your "good citizen" argument though. Something we should consider.
IIUC you mean this in the context of desktop environments (and not including things like Android, server OSes, or ChromeOS). The problem with your argument is that those of us who use Linux as our main desktop environment don't want just another platform: the entire point of its existence is its openness. The point is and has always been that you can download it, install it, and run it without having to jump through arbitrary hoops, agree to restrictive licensing, pay anyone, etc.
Watering that down for proprietary software vendors goes against the whole point.
Maybe we would get more, possibly better software. Steam has certainly shown that for games. However, the entire reason that Steam for Linux even exists is that folks at Valve didn't like what Microsoft was doing with Windows 8, specifically the Windows Store and its games for sale, competing directly with Valve. They created Steam Machines and Steam for Linux as a contingency in case Microsoft tried to use their position to prevent another, competing marketplace from existing.
They can do that because of the license, and specifically because of the license no one can later revoke Valve's ability to use the OS, even if Linus or GNU decide to go into the games marketplace business. That's why us purists insist on it.
World keeps turning.
Do your thing. See how that plays out for you.
As a proprietary app, we have encountered quite a bit of resistance from Linux users, usually berating us for not being open source. I think that's short sighted. We need more pragmatists.
And here lies the main point - they believe that many users will simply reject their proprietary software on principle. If the community was seen to embrace proprietary, which is what I'm calling for in the article, they might just support Linux.
Also worth noting that the largest contributor to the Linux kernel is... Microsoft.
Yes, because we've grown tired of companies abusing their relationships with customers: abandoning products and leaving them useless, charging more and more for the same or less levels of support, removing features, suing users for repairing their own property... The list goes on. Sure, similar issues can happen with FOSS, but at least you have some recourse if the creators disappear.
> Also worth noting that the largest contributor to the Linux kernel is... Microsoft.
Yes, but only now. And only because they did what they could to kill Linux, and they still lost. That former behavior is the kind of thing we like to never be an issue in the first place.
Which aren't properly any FOSS angels, when we examine their product portfolio.
Actually IBM is the king of patent submissions.
One does wonder whether this situation is intentionally exacerbated by the fundamentalists. Case in point, there are people against Flatpak because it makes it easier for developers to bypass repos and their package maintainers.
Hello there. An not so much against it, as not "for it".
> it easier for developers to bypass repos and their package maintainers.
No, that's not the problem at all. The biggest problem is duplication of libraries, most of which will not see security updates nowhere near as promptly as the system version would, integration with the DE and the problems there, fundamentally replicating the Windows/macOS model, while many see native package management as superior for the reasons I mentioned.
Flatpak isn't even good enough, in my opinion, since it adds complexity to an otherwise simple scheme in an effort to accomplish some of the deduplication you and other fans of package managers want. Ditto its repo model.
Yes I do. I was talking about selling packaged food, but still, restaurants make a good example also. They will always tell you if one of their recipe contains a particular ingredient (of which you might be allergic), and whether there are vegan options. Many restaurants even write the complete list of ingredients on each item in the menu!
Even then, except for very complex sauces, it is quite easy to see all the ingredients once you have the dish in front of you. This is impossible with proprietary software. Just grepping at a binary, you won't get too far.
https://github.com/Microsoft/Detours/wiki/Using-Detours
At one point I wrote a little piece of software as a proof of concept that used detours to redirect any file I/O in IE that was deemed "unsafe" to a special in-memory file system.
The bottom line is that you don't need the source code in order tell if an application is "phoning home" or if it makes suspicious API calls. In fact, it is often easier to just simply monitor how the software interacts with the host system to determine if it is performing (possibly) malicious actions. IOW, if you're concerned about a certain piece of software, then auditing the source code isn't going to be as good of a solution as just sandboxing the application so that it's impossible for the application to do something bad.
I authored the article. People ask us about our software all the time. We answer their questions. We have a post on our views on privacy, how we collect and use data etc. We give people the option to opt-out of data collection. Our email list is strictly opt-in. It seems that you are fine trusting the people that provide your food, but not your software.
Microsoft uses Linux on something like 60% of their own cloud servers, presumably because they find it more profitable to do so than to use the software they wrote themselves.
One thing you have possibly missed is that the people you think are fundamentalists rather than pragmatists, are for the most part being fundamentalist about pragmatism.
They are not telling you to open source based on some abstract philosophy, but because they think that the practical effects of doing so are better.
And given that open source as an economic system of production is currently out-competing liberal capitalism at its own game, on its own terms, on its home turf, I suspect that they may have a point.
Do what you want, but it may be a mistake to dismiss the open source community as not being pragmatic when they critique you.
I am dismissing those that reject proprietary software on principle.
I also made it clear that we would open source our software if we could find a business model that would work for us.
That dismissal presumably includes both Stallman and the Debian project. I would not dismiss them all that lightly.
>I also made it clear that we would open source our software if we could find a business model that would work for us.
You are making a nice email client. The service of rock solid business email management would presumably be the obvious model to start with. There are more people willing to spend money on that than there are people willing to buy a better email client.
If they dismiss it then they don't need it. You can't force anyone to buy things they don't need, at least in free society that is.
"But it doesn't run games or Excel on my desktop!" Ah yes you're right. Guess Linux is niche and needs spreading after all.
Linux runs the world. I highly doubt it needs more "spreading".
Example: I have an older laptop with a 5400RPM hdd that I've since bought a SSD to image and put a new OS on. Currently it runs Windows 7. Hate Windows 10 after being a lifelong Microsoftie (DreamSpark worked as it turns out), been learning Linux at work, so let's try it at home. I work on computers for a living, I can handle the learning curve, right? I will be reinstalling Windows 7 so I can AskWoody Group B it until it is out of support, then move it to Windows 7 POS Ready to get security updates until the last possible second. From there? I guess a mac because I refuse to use Windows 10 on personal devices aside from maybe a LTSB version. Different rant though, back to Linux on desktop:
I _want_ to run NeonOS or Debian KDE on that laptop, but cannot as the tools I use are not available on those platforms. The "fundamentalists" as portrayed in this article would say that I shouldn't have chosen to use the tools I chose. That doesn't help my problem though and won't do anything to get the 'unwashed masses' using the better software en mass. Here better software = FOSS since we are being a hypothetical fundamentalist a la Stallman.
If these Stallmen could "hold their nose" long enough to build Linux such that it could run most rando Windows programs without huge fuss and individual WINE config tweaks, the long term benefit would be that they would be positioning themselves to overtake Windows in usage on the desktop as well as the server side.
Knock-on effects are strong:
1. More users, more developers to create software
2. More developers, more improvements made to system codebases, package maintenance, dev tooling
3. More Linux, more Linux compatibility out of the box
4. More Linux on desktops, more people who've never been tainted by MS DreamSpark (like me) and due to baby duck syndrome prop up MS Windows, Office, VS and the Microsoft way
5. MS might even open source Windows itself (a core product) if it is no longer profitable. Making the job of dealing with Windows compatibility A) easier and B) nearly redundant
So the FOSS people win the war, at least on the desktop Linux vs Windows front.
If you include BSD as well as Linux, you get the Apple computers and phones.
If there is a choice use FOSS app A or proprietary app B I will use A if it does what I need, as a developer I adapted the open source apps to my needs where with proprietary apps I could never done it.
I also question him being a fan, it seems to me like he's a fan of building on it, that's about it.
If I need to do some short time job and there are no FOSS tools I usually buy the license for proprietary software do the work and then uninstall it. Since I can't even see if there are some obvious security vulnerabilities so why risk having it installed.
About the "poorer choice", by using not sufficiently featurefull FOSS tools you get the chance to improve those and everyone benefits.
In an open world you can't sell product, you sell skill, time, resources. If you try to be something like JustEat, Booking, Expedia, ... you are simply dead, but that's not a problem except at an hypothetical "model change", that's a good things for the society that prize real innovation ad work instead of marketing ad use (legitimate or abuse) of a commercial position.
It's not a matter of fundamentalism but a different social model.
I honestly don't find anything controversial in that post. It just lists some points which are between "obvious" and "common sense". There is, however, a typo in your header! now that is interesting ('and' instead of 'an')
A lot of people often confuse FOSS users and users of proprietary software. The difference is that FOSS users (for example users who do install linux, even on a Mac) typically feel more comfortable with the 'idea' that they or anyone else can look under the hook and tinker, if it comes to it and bothers them sufficiently. This is the primary reason why proprietary software is unappealing to them. Users of proprietary software don't value the freedom to look under the hood and tinker as much. Personally speaking, I won't judge people choosing one or the other.
However, arguments such as OSS makes "life a lot easier for a bad actor to pick it apart." or "..if you only have a handful of contributors, mistakes may not be caught." or "there is nothing to stop someone from forking it and selling it for less /offering it for free." is just disingenuous.
> bad actor
Unless your software is crazy popular and there is cred to be earned in the black hat community, people who go around looking for exploits in OSS typically would not be bad actors. In fact, a proprietary tool that run /on top of/ an OSS platform would be a much more appealing target for bad actors than an open source one -- because hey, finding a way to use 'paid-for' software for free is far more appealing than finding a way to reverse engineer some random binary custom format used to store addressbooks.
> mistakes may not be caught ...
total(mistakes caught by in house team) <= total(mistakes caught by in house team + 1 contributor) why is this hard to understand ?
> nothing to stop someone from forking it ...for free
Firstly - Licensing. Learn about it. Secondly, search for forks that are more popular than the upstream. I'm sure you might find a few (although I am also sure you won't find many). Understand why they are more popular (hint: will have nothing to do with code).
And finally, (most importantly) if you still feel that access to code is the only thing that will stop competitors from copying you and then beating you on price -- well, think hard about your future business strategy.