Open Source Is Eternal
linuxjournal.com
linuxjournal.com
I can’t help quibbling with the lead:
> Open source has won the present, but what about the future?
Open source has only “won” for programmers. With rare, beautiful exception, open stops just shy of end users, at the point where they have to pull out their credit cards. And GDPR notwithstanding, it’s still impractical for most end users to get most data about them in any practical, continuous fashion.
Great time to be a software producer. Not such a great time to be a software consumer. Unless you’re willing to trade nearly everything off for convenience and more freemium.
> Open source has only “won” for programmers
Both statements are true.
Open Source was always about developer benefit. The movement around end-user benefit is called Free Software. The one promoted by Stallman and the FSF, the one practiced by people releasing software under GPL family of licenses. This movement has definitely not won.
This is completely false. Corporations and Venture Capitalists are the ones who capture most of the value from open source projects. Independent developers got screwed over big time. We are the biggest losers. To the point that we probably would have been better off without open source. I've met popular, very skilled open source developers who can't even make a living as developers.
The top 10 open source developers in the world only earn a high/normal salary (e.g. Evan You of VueJS earns about $200K from Patreon; probably less than what he got at Google as a regular employee). He probably also makes money from conferences and talks; it's not bad, but if you look at the top musicians in the world, the 200th most popular musician alive probably earns millions per year.
The 200th most popular open source developer alive is probably a homeless person.
(capitalists ended up selling hanging platforms instead, the ropes come with the platform for free)
Nearly everyone builds their software on top of open source libraries, so the beneficiaries or "free riders" if you prefer include both lots of other programmers and the end users who download their software.
But, that's the choice you make when you choose to work for free and give away your stuff. Isn't it a bit weird that we expect anything else? And yet occasionally it works, you can sometimes capture a small fraction of the value through Patreon or whatever.
If you want a less precarious income, there are jobs working on open source software. The 200th most popular open source developer probably works at Google.
Choice is a relative thing. A lot of developers go into open source as a last resort because they believe that it's the only way that they can get access to users. Corporations have a near monopoly over users.
Sometimes the real issue is that the market is very competitive and user expectations are high, which is the opposite of a monopoly, though it can still make it very tough to break in.
https://blog.licensezero.com/2018/01/22/largesse-oblige.html
Same goes with anyone who used open source programming/scripting languages. The likes of PHP and Python are open source, as are Ruby, JavaScript, C/C++ and even now Java.
Hell, it's probable web development wouldn't exist with anywhere near as much of popularity if most basic components of it weren't open source.
Does that mean some people can't make much like they used to? Sure. But it also means a lot more people can make a living at all.
And given the salaries of many developers/programmers, it's a lot better as a situation than the one journalism is in right now. Or the ones involving creative fields like art, music, media, game development, etc, where the top 0.01% make money and the rest kinda struggle.
Which is a roundabout way of saying that Free Software is "a" movement around end-user benefit, with a very specific perspective about what that means and how to accomplish it that may or may not make sense to a lot of people, but it is not "the" movement around end-user benefit. I'm certain that the FSF and RMS see themselves as "the" movement, but that's actually their #1 problem in a nutshell IMO.
The other two freedoms, to run it for any purpose and redistribute it aren’t in any way unique to Libre software. They’re important sure, but are really just free beer benefits.
Users can benefit from access to better software, but only as an incidental or indirect benefit if developers happen to release it or if they pay developers to do so. In the latter case now you’re in a commercial relationship and arguably a closed model can give users more rights and control they might want, for example exclusivity.
If you look beyond licensing issues, Unix scripting, Emacs, Smalltalk, BASIC interpreters, and HyperCard were other early attempts at helping users become programmers, though spreadsheets turned out to be far more successful.
But if you want to help everybody, even the most nontechnical, even spreadsheets are sort of beside the point and not really what they want. Instead, making things easier to use (as Apple does) is closer to the mark.
A counter to this is that in the process of "making things easier to use," companies like Apple have in fact taken the "computing" out of the medium. Today Apple provides tools that allow very snappy and intuitive imitations of things that already existed: music, movies, etc., along with some basic tools to organize these things.
Spreadsheets and Hypercard are examples of media that are unique to computing -- they allow users to do things they literally cannot without having a computing device. This is not true for many of the things consumers use personal computers for today: as a phone, as a messaging system, as a radio, etc, etc. Computing today is almost myopically focused on making more convenient and efficient copies of things that already exist.
It might be true that Hypercard and spreadsheets are "not really what people want." But what if tools like those are exactly what people need?
I have come across innumerable situations where regular office people need to go just one step further than their shrinkwrapped desktop applications allow them to. At that point they have two options: either find some software that also does the very specific thing you need to do, or learn an entire general programming language and all the low level (and immediately irrelevant) information that goes with it. There is no longer anything viable in between those two extremes. As technologists we have really failed these people.
It seems like "imitations of things that already existed" is a flexible category that includes practically everything if you squint hard enough? Spreadsheets existed before computers. People did calculations by hand. You can do a lot of business things on paper.
"Things that already existed" also tend to be serving universal human needs that already existed. If you really needed it done and had the money, you could hire a human to do it. If the history of some human desire or need starts after computers, it's probably not all that essential, or maybe was caused by computers (like anti-virus software).
Second aspect: while freedoms 1 and 3 are developer-specific, they assume and secure an open marketplace. They're about the same thing that "Right to Repair" movement fights for in the hardware space - the ability for me, as a lay user, to ask a professional friend/neighbour, or a local shop, to pretty please fix/modify my software/hardware. The ability for that friend/local professional to perform that work legally. Where Open Source only allows this to happen, FSF-style Free Software tries to force it to happen, by legally ensuring the benefits I ordered and someone made for me could be used by the entire community.
It's a hard sell for businesses earning money by making software, but from the user's perspective it's important on a longer-term - particularly if said user believes in competitive markets, as competition in a market doesn't really work unless it happens over a commodity.
You're assuming that users cannot themselves be developers. Why can't they? Writing software is a skill and can be taught.
The fact that they could theoretically in principle access those benefits and therefore everything is in reality ok, is the sort of magical thinking that has put the Libre Software movement out on the fringes of relevance.
If I refuse to exercise a capability, that doesn't mean that I don't have it.
Refusing to tell a computer what to to is an odd choice, because what else is a computer for?
It's complicated.
I think you're right as far as the effect of "open source", and probably also perception of "open source". And perception is of course nine tenths of reality.
But if you ask some of the folks who were around early at OSI, and some who are around now, like Simon Phipps and Bruce Perens, I think they'll tell you "open source" is just a rebranding of free software. Having totally eclipsed FSF for their intended audience, OSI are now reaching back to that idea more and more explicitly, as by making "software freedom" a (vague) part of their license review criteria:
http://lists.opensource.org/pipermail/license-discuss_lists....
Very recently updated: https://opensource.org/approval
The barrier has always been fluid. Earlier on, when Bruce was feeling edged out of OSI:
https://lists.debian.org/debian-devel/1999/02/msg01641.html
So there's a cycle in the history. How much that history really matters, and how much is just palace intrigue or inside baseball, stands wide open to interpretation.