|Perl: the Markov chain saw|
Actually, more of the problem now as opposed to then was that to participate in such forums, there was a certainly technological barrier that one had to succum; when I first started in 89 or 90, (getting out my cane... :-) ), you had to either be at a high-end computing school or a research lab/site to have internet access. Thus, to have internet access assumed that you were already some of the more intelligent people (I'm not trying to slight anyone here, just stating observations). In addition, the only clients were available were text-based, thus, one had to know how to find and read the documentation for those clients. Furthermore, particularly in the case of USENET, in reading the documentation, you were told to lurk some time before posting, and other tips that would help to maintain high S/N.
Then the endless September came, and with the introduction of the Web, AOL, ISPs, and the like, everyone and their dog was able to get on the net. While the web with it's mostly graphical interface was great, the text tools for accessing USENEt were arcane and strange to those that dealt little with computers on a regular basis. Some folks saw this and developed on of the major 'downfalls' of the internet; graphical email and usenet clients. Now one could get to usenet without reading any user's manaul, and more importantly, without understanding the netiquette that had prevailed prior to that point. And thus that September was born.
Now we are slowly seeing the S/N ratio return back to it's previous values, though ever so slowly; there's been a lot of abuses of usenet (private companies starting general propigation of newgrps outside of it's servers, lax news server operators, increases in crossover postings, and most annoying, binary groups) along with new technical problems (lack of space forcing the need for fast turnaround, throttling/speed capping from the ngs) but there's also a lot of (new) people realizing Usenet isn't 'all that', and are abandonding it. The S/N will never get back to where it once was, but it's getting much easier to read through established groups than it was, say, 3 years ago.
Now with web communities, similar things are happening; look at /. - before user accounts and the like, it was a very high S/N system. Then they made it 'easier' for everyone to use, and suddently you had trolls, first posts, more than enough Natalie Portman and Grits for a lifetime, and other abuses. In this case, the established users cried for help from the /. system, and they slowly added controls that helped to reduce the problems; moderation, meta-moderation, the so-called Lameness filter, and others. Slash code 2.0 adds even more, and while the S/N isn't where it was, it, just like Usenet, is slowly getting back there.
Is PM undergoing a similar change? It's hard to say. We've definitely hit the point where the S/N drops because of the ease to use the system without understanding, and IMO it's slowly rising, but we aren't quite at the same recovery point as usenet or /. may be at. Can things be done in the backend to improve it faster? Maybe, but I think that many of the issues that are pointed out in this thread are ones that will disappear on their own (the votebot runners get tired of that, or in the worse case, as unfortunately with jc, the ones that are targets leave, making the votebot useless). Exactly when that might be, I don't know, but it's definitely food for thought as to what might happen down the road.
In reply to Re: Signal to Noise Ratio (wasRe: Ruminations of an ex-PM monk)