The Battle Between Purists and Pragmatists

Glyn Moody has a thoughtful piece taking a long look at the never-ending battle between pragmatists and purists in free and open software. "While debates rage around whether Mono is good or bad for free software, and about 'fauxpen source' and 'Faux FLOSS Fundamentalists,' people are overlooking the fact that these are just the latest in a series of such arguments about whether the end justifies the means. There was the same discussion when KDE was launched using the Qt toolkit, which was proprietary at the time, and when GNOME was set up as a completely free alternative. But could it be that this battle between the 'purists' and the 'pragmatists' is actually good for free software — a sign that people care passionately about this stuff — and a major reason for its success?"

Comments

Purists are just pragmatists who...

[Roxton]

Purists are just pragmatists who believe that moral imperatives are an adequate tool for achieving effective collective bargaining.

When the bargain fails to materialize, the purists blame a defective culture. And the pragmatists just roll their eyes.

Purist and pragmatist

[JohnFluxx]

The purist seeks to change the world to fit him, whereas the pragmatist changes himself to fit the world.

Ergo all progress relies on the purists. :-)

Re:Purist and pragmatist

[dkf]

The purist seeks to change the world to fit him, whereas the pragmatist changes himself to fit the world.

Ergo all progress relies on the purists. :-)

While the purist is sounding off about some moral crusade for cuter kittens or something, the pragmatist will have finished what they're doing and be in the bar with a beer. The purists see this as proof that they are right. The pragmatists see this as proof that they've got a beer.

Re:Purist and pragmatist

[dgatwood]

The purist seeks to change the world to fit him, whereas the pragmatist changes himself to fit the world.

Precisely.

Ergo all progress relies on the purists. :-)

No, you have it backwards. All true change depends on the pragmatist. While the purist is seeking a way to fix everything that's wrong (because it's all or nothing), the pragmatist is adapting himself/herself enough to actually solve as many problems as is practical, one at a time.

As Nietzsche put it (I think), before you can change the world, you must first change yourself. As long as you're on the outside looking in, you cannot effectively cause change. All you can do is spew rhetoric. Only when you come to accept that you can't save the world can you begin to save individuals within it, and in so doing, actually make the world better.

Re:Purist and pragmatist

[gbarules2999]

The pragmatists see this as proof that they've got a beer.

But it's not free beer, now is it?

Problem with pragmatism

[DaleGlass]

It can backfire.

For instance, take the whole mess with BitKeeper: The pragmatic option was to use a product with really obnoxious licensing terms, because it was good and worked at the time. Then one day Larry McVoy got really annoyed with Andrew Tridgell, and decided to refuse to even sell licenses to people associated with the OSDL, including Linus Torvalds.

That's the problem, while it works everything seems fine, but when the rug is suddenly pulled from under you, it suddenly creates a lot of complications that get in the way of getting useful things done. I think there's quite a lot of value in making sure that you'll be able to use tomorrow something you're using today.

Re:Problem with pragmatism

[dkf]

For instance, take the whole mess with BitKeeper: The pragmatic option was to use a product with really obnoxious licensing terms, because it was good and worked at the time. Then one day Larry McVoy got really annoyed with Andrew Tridgell, and decided to refuse to even sell licenses to people associated with the OSDL, including Linus Torvalds.

But it was pragmatists that fixed it. Indeed, purists would have kept Linux using a tool like CVS or SVN because going to a distributed versioning system would have let them to giving up their principles. It was the experience with BK that enabled the creation of git.

I suppose that this just illustrates a deeper truth: the world needs a mix of both purists and pragmatists. It's called "creative tension".

Re:Problem with pragmatism

[DaleGlass]

See, I see it differently.

Linus tried the pragmatic way. It worked for a while. Then it blew up in his face.

So he ended up having to do what the purist way would have required (writing a new SCM if none of the available ones were suitable), except that since events unfolded quite suddenly, there was no time for a smooth transition, and something had to be hacked up fast.

Git is certainly interesting, but I doubt half the people who use it really understand how it works. Maybe if it was started in less a dire situation it could have been more user friendly.

Re:Problem with pragmatism

[Chris Burke]

For instance, take the whole mess with BitKeeper: The pragmatic option was to use a product with really obnoxious licensing terms, because it was good and worked at the time. Then one day Larry McVoy got really annoyed with Andrew Tridgell, and decided to refuse to even sell licenses to people associated with the OSDL, including Linus Torvalds.

No, all that did is show that Linus Torvalds made an error in judgment that is common among the pragmatic "use the best tool for the job" crowd: Failure to consider the license as an aspect of the tool that affects its usefulness as much as the software itself.

I personally consider myself to be solidly in the "pragmatist" camp, and I argued against using BitKeeper not because I thought Linux development should be "pure" and only use OSS, but because I saw the BitKeeper license as a ticking time bomb that made the tool unsuitable for its purpose. It made some sense if you only thought short term, but I think that's foolish for such a long-term project. Then the bomb blew up faster than I even imagined, and in hindsight we can see it was in fact not the best move.

The problem with pragmatism, then, is that it involves reasoning about the future, trade-offs, risk evaluation, and so on and thus people can be and often are wrong about what constitutes the "pragmatic" choice. Identifying the "pure" choice is comparatively simple. "Is it free software or not?" Pick the free one and you did it right. You may choose to follow such a principle in part because you believe it leads to better practical outcomes too, but if it turns out not to be in some instance you were still "pure" which is what you were trying to be.

So the "purists" were right in this case because the pure choice ended up being the practical choice, but it was quite possible for a pragmatist to arrive at the same conclusion.

There is no such thing as "pragmatism"

[Bruce Perens]

People who label themselves as "pragmatic" simply aren't willing or able to consider their own interests on a longer timeline. A lot of them tell me that they finally realized that RMS was right about something, but it took them years, including a bad experience that was their own fault, to realize.

False dichotomy

[orzetto]

There are periodically arguments of ideological integrity vs. pragmatism in all areas. I usually react by asking "which foot do you use to walk?" or "when you climb a mountain, to you look at the path to the summit or to your feet?". Both ideology and pragmatism are required. If you use only ideology, you will not get anything practical done; if you use only pragmatism, you get something done, but it may well be in the wrong direction.

Re:False dichotomy

[CopaceticOpus]

I can't tell if this a purist view of pragmatism, or a pragmatic view of purism.

Life is not infinite, so I go with the pragmatists

[harlows_monkeys]

A few months ago, some usenet crackpot posted his latest mathematical research. Among the usual nonsense and ravings about a world-wide conspiracy of academic mathematicians, possibly under the control of aliens (the space kind), to suppress his work, there were some points of mathematical interest--some potentially neat patterns and relationships in how he was wrong.

I spent a very enjoyable few weeks investigating these, using Mathematica to aid in this. I was able to find things using Mathematica that I would not have found otherwise--even using the best current free mathematical software, and those taught me a lot, both directly, and from the books I then consulted.

The most pure purists, such as RMS, take the position that I should not have done that mathematical investigation, because I could not do it without using non-free software. I'm supposed to wait until I can do it with free software, and maybe contribute to developing said free software if I want to speed things up.

If life were infinite, I would consider that. Life is not infinite, so I will go ahead and use the tools that let me get done the things I want to get done during this short life. I see no difference between, say, riding in a vehicle like a boat or plane where I cannot inspect and study the engine and using a piece of software where I cannot see the code. For the boat, all I care about is that it accomplishes the task I need--getting me safely to my destination. Same for software.

Re:Ha!

[Anonymous Coward]

Define "define".

#define define

Can we move on?

Without the purists ...

[Thomas Larsen]

As a 'purist' in this sense myself, I feel somewhat justified in claiming that without the purists, there would never have been such a pragmatist movement as there is now. Let's face it: in all likelihood, if I were a pragmatist, I'd be using proprietary software tools to write programs and share information--because that was, and perhaps still is, the easy option. Purists, on the other hand, reject compromise when that compromise will eventually result in their freedoms being restricted.

Actually, free software is the pragmatic option: it guarantees that, in the future, I'll be able to code using free, compatible tools. Software that compromises on freedom will eventually fall into the trap of convenient, non-free, proprietary software that will eventually restrict my freedom in the future to write and share and change programs in an upstanding, moral way.

My two cents.

Re:Good for both!

[Bogtha]

I think that pressure from Gnome and the fundamentalists helped make Qt change their license to the LGPL.

I can't agree with that. TrollTech were doing nicely dual licensing Qt. They were essentially making proprietary software companies fund Free Software development. That was their business model. LGPLing it would have lost them plenty of money, and for what? The respect of GNOME fundamentalists? Their customers were already choosing Qt over GTK and the GNOME libraries, throwing away their business model might have helped their "market share" in the Free Desktop "market", but it would have lost them loads of money.

Qt was LGPLed right after TrollTech were bought by Nokia. The licensing fees for Qt aren't significant to Nokia, they wanted a high-quality toolkit and the developers that built it. So it wasn't necessary to keep it GPLed. Then it makes sense to LGPL it to get more users/developers/market share/whatever. But before that point, the "pressure" from GNOME was insignificant compared with the pressure of bills to pay.