[RC5] new release candidate available.

Bruce Wilson bwilson at distributed.net
Tue Nov 26 01:09:39 EST 2002


This can also be explained by the fact that RC5-64 ran so much longer,
it had time for a lot more one-up installs to get "lost".  Even with the
servers denying RC5-64 requests, we're still getting randoms from lots
of embedded systems that don't know any better.

We have also encouraged people to make better use of teams and
perproxies to get the same detail of reporting that formerly required
multiple email accounts.  Remember that a "participant" is measured by a
single email address, regardless of how many computers are contributing
under that account.  1000 participants with one computer each count just
as much as 10 participants with 100 computers each (given the same
CPU's, of course).

I haven't looked myself, but historically we have seen significant
weekend dips and Monday spikes.  The reasons for this are as diverse as
our participants.

I haven't been in any of our IRC channels in over a week, but the wide
selection of "release candidate" platform support is enough that I can
tell it's not long before a full release.  Barring new problems, of
course.

On the OGR front, I've heard rumblings of some corrections to both the
client and our reporting mechanism.  I have no idea of a timeline, nor
of how/whether it will affect RC5-72 release.  All I can say is someone
is considering the issue.

We're as frustrated as you are that our current OGR release can't be
counted on 100%.  We still stand behind the statement that all OGR work
submitted to date can be useful for confirming work done by a
third-generation OGR client.  The second pass must be completed by a new
client, but the first pass can be done by any version, as long as it
agrees with the known-good core.

Yes, it would have been nice to have RC5-72 working closer to the time
RC5-64 was finished.  We started working on coding as soon as we were
sure of our results (I.E. before the announcement).  It would also have
been great if we could have poured as much effort into fixing OGR at
that time.  For good or bad, the completion of RC5-64 has given us the
impetus to get rolling on coding clients again.  I'm optimistic that the
surge of activity will continue after we release RC5-72, moving into the
OGR's and other possible projects.

On a personal note, distributed computing is an exciting project, and
one with mind-boggling possibilities, but it will never be a higher
priority than my work, my family, my health, my personal time.  Opposing
viewpoints are welcome, and even discussion of other projects and what
they do better or worse do not bother me.  Accusing me of neglect
because I have not posted a weekly update is a mockery.

There are times the accusations on this list make me mad enough to pitch
the whole thing.  Thank goodness for the few sane respondents, who speak
up to voice their support.

We're all in this together - some more than others.  There's no sense
getting your panties in a bunch simply because our timeline doesn't
match your expectations.  If the wait is intolerable, I suggest you find
a project that better fits your expectations.  If the problems in OGR
cause you too many doubts, find something to your liking, even if
temporarily.  But don't think that yelling in here is going to make us
excited to make you happy.

As the saying goes, lead, follow or get out of the way.

__
Bruce Wilson <bwilson at distributed.net>
PGP KeyID: 5430B995, http://www.toomuchblue.com/ 

"I want to move to Theory. Everything works in Theory."
    --John Cash, id Software


| -----Original Message-----
| From: owner-rc5 at lists.distributed.net 
| [mailto:owner-rc5 at lists.distributed.net] On Behalf Of Jason Hartzell
| Sent: Monday, November 25, 2002 12:12
| To: 'rc5 at lists.distributed.net'
| Subject: RE: [RC5] new release candidate available.
| 
| 
| Yesterday's count is what's available. Come back tomorrow. Do 
| you think
| 18,000 participants were shut off over the weekend? Even 
| assuming 1/3 of the
| participants did not return stat units yesterday still gives 
| you less than
| 10,000 active participants. So that doesn't change or 
| discredit anything I
| said.
| 
| Thanks.
| 
| 
| -----Original Message-----
| From: Victor A. Wagner, Jr. [mailto:vawjr at rudbek.com]
| Sent: Monday, November 25, 2002 9:14 AM
| To: rc5 at lists.distributed.net
| Subject: RE: [RC5] new release candidate available.
| 
| 
| Picking a Sunday for claiming current participants is folly.
| At Monday 2002/11/25 09:30, you wrote:
| >Some good points Mr. Marsh, but my view just isn't the same.
| >
| >Yesterday, 6,170 people participated in OGR-25 and 7 in 
| OGR-24. On the
| final
| >day of RC5-64, 25,739 people participated. Do you really 
| think that the
| >release of RC5-72 will bring back 20,000 daily participants? 
| Of course it
| >won't. Do you think, if RC5-64 and RC5-72 had been running 
| concurrently,
| >that a portion of those would never have quit? How many 
| participants were
| >lost just because they had OGR=0 in their ini file and they 
| aren't going to
| >bother going back to change anything.
| >
| >Besides, I don't agree with the notion that the 'whole 
| point' was to prove
| >how quickly RC5-64 could be completed. If it were, the project was a
| >complete failure. The purpose was to prove that it COULD BE 
| DONE. Period.
| >With that, it was not a failure. And starting the RC5-72 
| project sometime
| >after RC5-64 got to 75% or 80% of the keyspace was exhausted 
| would have
| made
| >a lot more sense than this. Remember, we are two months past 
| RC5-64 now...
| >
| >
| >-----Original Message-----
| >From: Timothy Marsh [mailto:Timothy.Marsh at usm.edu]
| >Sent: Sunday, November 24, 2002 6:43 AM
| >To: rc5 at lists.distributed.net
| >Subject: Re: [RC5] new release candidate available.
| >
| >
| >I know the purpose of dnet is not to break encryption, but rather to
| advance
| >distributed computing.  But the purpose of the RC5 project 
| is to crack the
| >code in as little time as possible, which has been seen as a way of
| >supporting distributed computing.  The whole point is to 
| show how quickly
| it
| >can be done.  As someone pointed out a while back on the 
| efficiency of
| >solving these types of problems, there are two important 
| things to keep in
| >mind when you are trying to finish a project in a short 
| amount of time (not
| >necessarily earliest).  First you should not get too 
| distracted, focus your
| >work as much as possible on one project.  For a variety of 
| reasons, dnet
| has
| >also provided OGR, but I really believe that adding two RC5 
| contests at the
| >same time would be counterproductive.  Second, wait until 
| computing power
| is
| >powerful enough to accomplish the task in a reasonable 
| amount of time.  You
| >may be thinking of this time as wasting a few months of 
| processing time,
| but
| >if RC5 72 had been released a couple years ago, I imagine 
| the work done on
| >it then would have taken a quarter of the time to do now, 
| but the work
| taken
| >away from the 64 project would have been very significant.  Much more
| >significant than the work that is being "wasted" right now 
| because people
| >don't like OGR.
| >
| >Timothy
| >
| >----- Original Message -----
| >From: "Jason Hartzell" <jhartzell at arcataassoc.com>
| >To: <rc5 at lists.distributed.net>
| >Sent: Friday, November 22, 2002 5:58 PM
| >Subject: RE: [RC5] new release candidate available.
| >
| >
| > > I was talking to a friend (and former d.net participant) 
| about this the
| > > other day. My thought was that RC5 project #1 and #2 
| should be running
| > > concurrently. It didn't really matter if RC5 project #1 took a bit
| longer
| > > just because processor time was split between the two. 
| Remember, the
| >purpose
| > > of d.net is NOT to break encryption.
| >
| > >>>snip
| >
| >--
| >To unsubscribe, send 'unsubscribe rc5' to 
| majordomo at lists.distributed.net
| >rc5-digest subscribers replace rc5 with rc5-digest
| >--
| >To unsubscribe, send 'unsubscribe rc5' to 
| majordomo at lists.distributed.net
| >rc5-digest subscribers replace rc5 with rc5-digest
| >
| 
| Victor A. Wagner Jr.      http://rudbek.com
| The five most dangerous words in the English language:
|                "There oughta be a law"
| 
| --
| To unsubscribe, send 'unsubscribe rc5' to 
| majordomo at lists.distributed.net
| rc5-digest subscribers replace rc5 with rc5-digest
| --
| To unsubscribe, send 'unsubscribe rc5' to 
| majordomo at lists.distributed.net
| rc5-digest subscribers replace rc5 with rc5-digest
| 
| 


--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list