|Think about Loose Coupling|
Re^5: Breaking Out of the Perl Echo Chamber: A Call to Actionby AZed (Monk)
|on Sep 19, 2008 at 09:10 UTC||Need Help??|
I don’t see how you can credibly make this slippery slope argument given the fragmentation that already exists. I read use.perl but hardly Perl Monks, these days. I picked up p5p a year ago, which I was never reading before. I have never posted on comp.lang.perl.misc. Presumably at least one of these places qualifies as a good place to get information in your opinion. If so, how is your position self-consistent?
Because out of the places you listed there, only Perlmonks qualifies as a good place to find answers to a personal question. Use.perl appears to be news site/blogs, not community forum. P5P I hadn't even heard of until you mentioned it, and that's because it's a mailing list ostensibly devoted to further development of the language itself, which makes it a) a little difficult to find things in if you aren't tracking it continuously, b) somewhat inaccessible and difficult to find, c) not really oriented towards helping people with general Perl questions, and d) somewhat of a nuisance to maintain. Even now that I know about it, I'm unlikely to pick it up because I have more than enough junk cluttering my mail storage already. Usenet is... quiet, and has been for a long time. I got half of a functioning answer there recently, from one person, and it took longer to get it than from Perlmonks, even though I asked there a day later. I'm unlikely to bother with comp.lang.perl.misc again.
If you're looking for an active web-based question-and-answer forum for Perl questions, Perlmonks appears to be it.
Now, considering how many questions we get on Perl Monks from people who say “I know this is not Perl-related, but Perl Monks is the best place for this kind of thing that I’ve ever seen, so…”, how likely do you think are people new to Perl to ask Perl questions wherever they happen to already be regulars?
I wouldn't expect all that many, unless you know a lot of other places where they might be regulars that would be the best place for Perl answers that they had ever seen. One of the major complaints, as I was seeing it, was that the answers to Perl questions outside of Perlmonks tended to be rubbish.
What possible advantage does knowledge hoarding offer?
Call it "knowledge concentration" instead, and you'll see that it offers a very powerful one: someone with a question doesn't have to ask in half a dozen places before finding one with a good answer. I wasted a considerable amount of time repeating myself and waiting for answers that weren't forthcoming before getting to Perlmonks. Fragmentation means wasting even more time asking questions where the right people aren't watching that day and searching through archives that don't contain the answer. This is not unique to Perl. I don't come to Perlmonks for answers to PHP questions.
If you want Stack Overflow to be the new place to go to for Perl information, you don't want a scattering of people at Perlmonks over there: you want them all over there, and you want Stack Overflow to be hosting a mirror of all of the informational nodes currently hosted at Perlmonks. Otherwise, what you're doing is telling all new users that if they want a Perl answer to log into Stack Overflow, search, log into Perlmonks, search, and then log into Stack Overflow and post their question and then log into Perlmonks and copy and paste that question — or simply picking one blindly and praying that the person with the answer is at the same site that day. This is not a good model for information retrieval.
Incidentally, I wouldn't be greatly put out if Perlmonks did migrate its entire "Seekers of Perl Wisdom" section over to Stack Overflow, except that generalist sites generate a lot more false positives on archive searches than specialist ones. That's a problem that could theoretically be solved by a better search engine, however. What I'm not greatly enthused about is making the knowledge pool shallower. Limited answering time makes location-based question-and-answer competition a zero-sum game. You cannot improve the amount of time spent on questions at Stack Overflow except by taking time away from something else. The same is true for Perlmonks and c.l.p.m., or any other source of dilution.
The Perl community only sporadically pays attention to the hot issues of the rest of the programming world and almost never tries to sell its issues to a wider audience as worthy of attention, and the rest of the world happily ignores the Perl community.
I'm having a hard time parsing this. I'm using Perl because it's a programming language that currently fits my needs (and the emphasis here is on currently). You seem to see it as some kind of political movement. If you think the world is ignoring Perl, when organizations as large as Amazon build themselves on it, I shudder to contemplate what you think the world thinks of Ruby, Lisp, or even C++. I'm also not sure what 'issues' you think will be solved by having questions occasionally answered in a general questions site — are you expecting the people asking the questions to then turn around and contribute code that changes the direction of Perl development? I think you're being overly optimistic if you are even expecting them to contribute modules to CPAN. That's the kind of thing you do once you have committed yourself to a language for a particular project.
I'm also not quite sure what you mean by 'hot issues of the rest of the programming world'. Perl programmers are programmers, and are part of the rest of the programming world because they are usually being asked to solve problems with programs. If there's a hot issue in the general programming world, i.e. a problem that is generally difficult to solve with programs, then if some programmers decide that Perl is a good tool for solving that problem, it becomes an issue that the Perl community is paying attention to — by definition! And if it turns out that Perl is not a good choice of tool for that problem, then you don't want the Perl community paying attention to it, unless the fit is only poor or marginal because of a correctable flaw in design, in which case it probably doesn't matter if Perlmonks in particular pays attention to it, because this isn't where the core language development happens. (Unless I'm misinformed about who is building what, anyway — if Perlmonks discussions really do turn into core Perl features on a regular basis, I withdraw the statement, but I suspect this is in the category of things better shunted off to that P5P you mentioned.)
As for the rest of the echo chamber business, I've left code behind me in Pascal, C, C++, PHP, and now Perl. I've been at Perlmonks less than a month. If my opinion represents an echo chamber, the term needs a little redefining to be useful. Whether or not you believe me, however, you may wish to reconsider the usefulness of showing how little you respect the people you're asking to help you.