mersenneforum.org  

Go Back   mersenneforum.org > Extra Stuff > Science & Technology

Reply
 
Thread Tools
Old 2006-10-09, 21:49   #1
Xyzzy
 
Xyzzy's Avatar
 
"Mike"
Aug 2002

2×13×307 Posts
Default Lurkers...

http://www.useit.com/alertbox/partic...nequality.html
Xyzzy is offline   Reply With Quote
Old 2006-10-09, 22:13   #2
axn
 
axn's Avatar
 
Jun 2003

485610 Posts
Default

Quote:
Originally Posted by TFA
How to Overcome Participation Inequality
You can't.
Bummer

So what is our distribution?
axn is offline   Reply With Quote
Old 2006-10-10, 01:27   #3
Mini-Geek
Account Deleted
 
Mini-Geek's Avatar
 
"Tim Sorbera"
Aug 2006
San Antonio, TX USA

10AB16 Posts
Default

Quote:
Originally Posted by axn1 View Post
So what is our distribution?
With curtisc having done about 100,000 P90 LL years, and the top 9 having done over 10,000 years each, I'd say it's pretty extreme.

Last fiddled with by Mini-Geek on 2006-10-10 at 01:27
Mini-Geek is offline   Reply With Quote
Old 2006-10-10, 02:36   #4
Phil MjX
 
Phil MjX's Avatar
 
Sep 2004

5×37 Posts
Default

It is the same on this (and other) forums : 1304 users, how many over 1000 posts ?? Have a look at posts counts !
Is it a problem that a few persons contribute a lot in a field like distributed computing (and related forum)? I don't know, surely not in that case...

Last fiddled with by Phil MjX on 2006-10-10 at 02:41
Phil MjX is offline   Reply With Quote
Old 2006-10-10, 07:08   #5
jinydu
 
jinydu's Avatar
 
Dec 2003
Hopefully Near M48

2×3×293 Posts
Default

Quote:
Originally Posted by axn1 View Post
So what is our distribution?
It is somewhat hard to say; http://mersenne.org/ips/stats.html hasn't been updated in many months. But as of February this year, PrimeNet had completed over 1.32 million CPU years. This is could be a significant understatement of the total work done by GIMPS from its inception to today.

However, I would think that for a distributed computing project, an unequal distribution is not such a bad thing. Does it really matter where the contributions come from?

Last fiddled with by jinydu on 2006-10-10 at 07:08
jinydu is offline   Reply With Quote
Old 2006-10-10, 09:33   #6
axn
 
axn's Avatar
 
Jun 2003

23·607 Posts
Default

I meant the forum usage -- naturally, "lurkers" isn't a word you associate with GIMPS participation
axn is offline   Reply With Quote
Old 2006-10-10, 11:57   #7
Phil MjX
 
Phil MjX's Avatar
 
Sep 2004

2718 Posts
Default

Quote:
Originally Posted by axn1 View Post
I meant the forum usage -- naturally, "lurkers" isn't a word you associate with GIMPS participation
For forums "lurkers" aren't a problem IMHO since forums can be source of informations...In my gmail box, I do recieve every day posts from math-related yahoo groups, I read them and it's all...what's wrong with it?

The article says it could introduce bias in case of stats, polls, comments but it isn't a problem for mersenneforum (I doubt that average internet surfer will even have an opinion about the poll : shall we find a Mp by doublecheck like SoB does ? )...

regards...
Phil MjX is offline   Reply With Quote
Old 2006-10-10, 18:40   #8
ewmayer
2ω=0
 
ewmayer's Avatar
 
Sep 2002
Rep├║blica de California

265278 Posts
Default

The cited article does have one nice piece of "interesting" statistical conjecture:
Quote:
Blogs have even worse participation inequality than is evident in the 90-9-1 rule that characterizes most online communities. With blogs, the rule is more like 95-5-0.1.
I suppose the blog stats are an example of "giving 100.1%." ;)

Personally, I'm most interested in the breakdown of the 1% (or 0.1%, or whatever) of users who contribute the bulk of the postings on typical message boards, among the 2 major categories one tends to see:

1) People who actually have something interesting to say (even if not in every post);

2) Spammers/Trolls/Idiots-with-a-fast-internet-connection-and-too-much-time-on-their-hands.

In particular, is the ratio of (1) to (2) sensitively dependent on the precise nature, subject matter and administrative policies of a forum, and especially in the latter instance (admin policies), to what extent can a small amount of moderation act to discourage the trolls? The idea here is, one wants an effective way to discourage (2) without imposing too much administrative overhead (e.g. having all posts subject to review by a moderator before allowing them to be posted) or making regular users (both the good and the merely harmless) feel overly constrained.

Some lessons learned from this particular forum here:

* Decide whether a multi-track approach is warranted, i.e. a division of subfora into general ones open for anyone to post and more-specialized ones where only registered users may post (and/or which might be actively moderated).

* Decide whether a deliberate delay between consecutive posts by any user is warranted. Here it used to be 5 minutes, but even that proved too ripe for abuse by determined spammers, so it's now 15 minutes.

* Figure out which types of features/topics tend to attract one group or the other. Again using hard-won experience around here as an example, certain types of Lounge thread topics (e.g. "word association") appear to act as a kind of flypaper for spammers, but with the unexpected downside that once such a person gets in the habit of posting dozens or hundreds of times a day to a particular forum, if they find their "participation" is not limited to the silly threads, you may find them beginning to post with a similar content-free thoughtless-spam style to the more serious topics as well, which is at best extremely annoying and at worst highly disruptive and tending to drive the folks (1) (the ones you want to keep) away. I believe the technical term for this is "the inmates running the asylum."
ewmayer is offline   Reply With Quote
Reply

Thread Tools


All times are UTC. The time now is 05:52.

Tue Mar 2 05:52:14 UTC 2021 up 89 days, 2:03, 0 users, load averages: 1.94, 2.18, 2.41

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.