open source as unpleasant subculture

[tim_maroney]

A /. story led me to the archives of the Linux kernel mailing list. In this I am reminded of one personal reason I have never been attracted to the Free Software or Open Source ideology. It requires a major lifestyle commitment which involves the company of unpleasant people, as a quick browse through those archives will demonstrate.

Comments

[isomeme]

I've definitely run into the same dynamic on many open source projects. At a previous job, we were using an open project as a component in our product, and needed to get a code modification integrated into the open project in order to do what we needed without forking the codebase. Now, this was a very beneficial change to the code, which subsequently has been used by many people; but we still had to pass through a "hazing" period unrelated to the merits of our submission before we were judged worthy of having our code committed to CVS.

Not every open project is like this, though; there are huge cultural variations from community to community. For example, I'm currently working with Apache Cocoon, which is sparsely documented and fiendishly complex, but is supported by a developer/user community unusually willing to provide straight, pertinent answers to well-phrased technical questions on the project mailing lists.

[tim_maroney]

Apache does seem way above the mean for open source software. many of the side projects are quite impressive, and I'm glad to hear at least some of them place value on collegiality.

Thanks for the tip on Cocoon. The separation of functions to reflect different core competencies is particularly intriguing -- whether this particular model works out or not, it reflects much more concern for the authoring process than most tools.

I was sufficiently interested to start installing it, but I found the 13-page install instructions daunting, and that doesn't include the other tools I would have to install like Tomcat and Ant. Even the relatively good open source projects still have the Frankenstein problem, or more specifically, painful installation and administration costs, which I would say result from deferring integration costs onto the user. As a programmer and designer I want to concentrate on development tasks, not on system-integration tasks.

[isomeme]

Yes, you're definitely right that the Apache projects in general are the cream of the crop in terms of both attention to architectural and authoring issues and in how they treat newcomers.

If you already have a servlet container running (Tomcat, for example, but anything 2.2-compliant should work), then installing the basic Cocoon package is as easy as building or obtaining its warfile and deploying it. Most of those 13 pages are about setting up the tools needed to build (Ant) and host (Tomcat) Cocoon, rather than Cocoon itself.

That being said, getting beyond the default install and actually doing your own stuff with Cocoon is a bit daunting, hence my forays onto cocoon-dev and various wikis in recent weeks. The "Frankenstein" image definitely applies here. I'm hoping the payoff in separation of concerns in our View layer will pay for my investment of time in climbing this learning curve, but alas there's no way to be sure a priori.

[tim_maroney]

Tomcat installation on OS X from source was giving me major headaches (kind of surprising since I've successful installed it before on Solaris). I backed off for a while, then did a Google search for better instructions and found this, which made it much easier. A successful install is pleasing, of course, but at the same time, this free software experience always leaves me scratching my head and asking, is there any reason just installing something needs to be this hard?

The question itself is politicized and subject to the status dynamics of the open source subculture, which is to say, asking it is punished in a variety of ways. Gee, I didn't have any trouble, wonder what's wrong with you is the most common one; a more tolerant approach simply complains in an exasperated way why this person is objecting to the same kind of thing we all have to do every day. A cynical approach welcomes the difficulty as a guarantor of job security and lashes out at the threat to the established order

[ex_stonemir]

This is an excellent analysis of the situation

[isomeme]

I can understand the risk to a developer of trusting GPL to protect their work against commercial exploitation, but what is the risk to a company which is willing to work within the fairly well-known boundaries of legitimate commercial use of GPL software?

[ex_stonemir]

Being sued by someone who claims that the company is not exposing everything it's obligated to, largely. As I mentioned, the language of the agreement is both unclear and untested, particularly insofar as it applies to extensions which are intended to be proprietary to the basic material.

You say "well-known boundaries", but well-known by whom? Many conversations with the general counsel to the Open Source Foundation failed to elucidate what was and what wasn't required of us, and--as noted--not a lick of this has ever seen the light of day in a court of law.

We're not interested in becoming a test case.

[isomeme]

Interesting point. Of course, the vast majority of software licenses and contracts are never tested in court; GPL just happens to be (a) especially well known and (b) ideologically driven and couched in unusual language. Those latter factors may make it a more attractive or likely target, but it should not be forgotten that nearly all licenses and contracts remain untested in the courts without ill effect.

[phygelus]

instead, you post to slashdot.

[tim_maroney]

Fortunately, kibitizing is easy!

Still, I've been gratified that by keeping my concerns civil and reasonable, I've been able to help make more space on /. for positions more complex than the formerly de rigeur "d00d, you need to read The Cathedral and the Bazaar again." I'm not the only poster who's helped to legitimize discussions of problems with open source and free software by any means, but I do have some feeling of accomplishment.

I've also tried to work the issue from the contribution side, but that's where I acquired the experiences I summarize above. I could do it but it would require more lifestyle commitment than I am willing to give.

[phygelus]

Oh, kibitzing can be lots of fun, and sure, contrary to folk wisdom, sometimes one can even accomplish things in the process.

I mostly intended a snide comment on the nature of the slashdot "community". I nearly wrote "Instead, you post to slashdot and join occult orders"...