Paul,

Have you read any of the BOINC papers
(http://boinc.berkeley.edu/trac/wiki/BoincPapers) lately?

There have been more than a few research papers on how BOINC works, what
failure rates one can expect, and how that compares to other distributed
computing models.

I recall reading a paper on how the host reputation system and adaptive
replication system increased overall project efficiency without
incurring much if any increase in error rates.  Problem is I can't seem
to find it at the moment.  Time for me to go to bed I suppose.

In any case, these things are constantly being looked at.  BOINC is
still a research project.

----- Rom

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Paul D. Buck
Sent: Thursday, October 01, 2009 12:19 AM
To: [email protected]
Cc: BOINC Developers Mailing List
Subject: Re: [boinc_dev] [boinc_alpha] Card Gflops in BOINC 6.10


On Sep 30, 2009, at 5:36 AM, [email protected] wrote:

> I don't care if it is generating credit.  It is not doing the real  
> science
> that we are supposed to be doing.  It is just a gold plated  
> benchmark.  It
> cannot guarantee that the next result is valid, it is only a better  
> credit
> estimator.  It is a waste of time to program, it is a waste of time  
> to run.

The problem is that what you don't want to see that real science is  
about careful attention to detail and the effort to eliminate error.   
When error CANNOT be eliminated we compensate.  But we also, to the  
extent possible, measure our level of error.  All of that, according  
to you is waste ... it isn't.  It is what science is about ... careful  
measurement, looking for and the elimination of error, and accurate  
measurement.

Sorry for the comparison because I don't know how else to get it  
across ... you are like the kid that wants it RIGHT NOW ... and  
anything that gets in the way of that ...

The real science we are supposed to be doing should include lots of  
stuff that we keep dropping by the wayside because too many like you  
are only interested in doing heaps of results ... not heaps of  
accurate results ... which would only be a slightly smaller heap ...  
just heaps ... the devil take the details and the accuracy as long as  
we can pile up the results ...

Its Ok John ... you are going to win the argument because my  
disability is on the rise and I will not be able to continue the  
argument ... besides, doing things the right way is always hard ...  
slapdash work is always faster ... the part that makes me sad is that  
I seem to be the only one that thinks that science involves rigor and  
accuracy.


And I still find your fetish for efficiency hysterically funny when  
you tolerate a level of waste in the operation of the base client that  
is actually unjustifiable ... and funnier still ... you pointed to as  
a problem in an earlier post ...
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.

Reply via email to