[RC5] Removing Stats?

Andrew MacKenzie andy at edespot.com
Fri Sep 27 13:12:15 EDT 2002


> > People, Distributed.net stats don't give you longer life, make you a
> > The contest is over, the winner is the one with the key.  Does the fact
> > that you may be 6034th in stats vs. 6033th make *any* difference?
> 
> Not to you, maybe not to me neither, but to a lot of people it actually
> does. Do you frown upon that? It's your right to. But if too many people
> openly frown upon that, these people will (and some have) find another
> project where the stats have more status.
>
> You may feel you "don't need those people". But a distributed computer
> project doesn't primarily need people, it needs processor cycles.
> Regardless (within certain limits) of the owners reasons for participation.
> Loosing processor cycles for a disagreement in what is important will only
> hurt the project.
> 
> IMHO, not fully understanding that is the weakest part of the RC5 project
> management sofort. And I believe that with another approach vs less 'geeky'
> participants, many more would participate in the RC5 and related projects.
> Participation==providing processor cycles.
This point I can see.  It's not very politic of Dnet et al to just delete
the stats since the key was found.  


> > If people are so concerned about their stats, why don't they run
> perproxies
> > and maintain their own stats like I do?
> 
> Please, correct me if I have misunderstood how you use the perproxie,
> but...
> 
> The real 'stats guys' (and gals) prefers to *compare* their stats with
> other people, not with their own result x time in the past. And remember,
> there are tons of processor cycles out there owned by people with none or
> limited knowledge of computing. (For instance, installing the dnetc client
> on a Win-32 PC doesn't require a whole lot of knowledge.) These people will
> not build their own routines for maintaining dnetc stats. Missing out on
> these peoples processor cycles because they don't feel they have the right
> or enough feed-back is unlucky. Again, IMHO.
I can understand the inability to setup a system for keeping track of stats
(though there are some programs that will scan logs and present stats based
upon them.  But don't blame the volunteers who setup the system for lack of
stats.

I'm, admittedly, a stats nut!  That is why I wrote a series of perl scripts
to analyze my perproxy stats (for my team "PlaidHat") and generate stats
accurate up to 5 minutes time (rather than wait for midnight).  Perhaps
I'll make it more 'user-friendly' and release it to the public for use on
other teams (the code isn't that clean, it would need some fixing).  


-- 
// Andrew MacKenzie  |  http://www.edespot.com
// Clothes make the man.  Naked people have little or no influence on society.
//                 -- Mark Twain
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
Url : http://lists.distributed.net/pipermail/rc5/attachments/20020927/d92e5cd1/attachment-0001.bin


More information about the rc5 mailing list