Prof.Wizard

LOOKING FOR A NEW 'SR' METACOMPUTING PROJECT

Recommended Posts

D2OL is on the following:

Windows 98/2000/NT/ME/XP

Mac OS X

Solaris

Linux

FaD I think is only on:

Linux

Windows

So I guess we should go on D2OL.

Damn I have no clue what the write-up would say. PW...can you help on that? BTW, once (and if) we go to D2OL, please resume your responsibility as Team Captain. I'm sure you're a better one than I would be...

C

Share this post


Link to post
Share on other sites

Keith Davies the chief of the Find-a-Drug project can give us detail answers about aspects of the project. When I posted to F-a-D forum about Storage Review community trying to find the most interesting and useful project he wrote:

Posted: Mon Oct 13, 2003 1:20 pm Post subject:

"On occasions with other teams, I have visited their forums to answer specific questions as well as answer extended e-mail questionaires. I am quite happy to do so again although I do not make unsolicitated posts because this might be badly received as promoting a DC project. Perhaps the best approach is if Vladimir posted the essence of this post.

_________________

PS I'm the guy in charge of Find-a-Drug

(This means I take all the blame and give all the credit to others)"

So we need only make questions about details of project and reasons to go with him

Share this post


Link to post
Share on other sites

I was made aware of this thread through the fad forums.

My reason for posting, though, was because I see you are considering d2ol as well. I wanted to suggest another choice if you are seriously considering d2ol.

http://www.childhooddiseases.org/

The rothberg institute's tsc project.

Both this project and the d2ol project are overseen by the rothberg group now. They both are based on sengent d2ol code. Originally d2ol was a sengent set of projects and sengent collaborated with the rothberg institue to get the institute's tsc project off the ground. Now both are under the institute.

The reason I bring this up is because tsc may be a more appealing cause than the d2ol targets such as anthrax or sars.

I'll let their front page give some details::

" finding a cure for children suffering from Tuberous Sclerosis Complex (TSC)...

the work we do on TSC will also have a direct impact on cancer research, and we will be testing the drugs we find as weapons against cancer. In addition the work we do to create a system to attack orphan diseases will form a prototype for work on other orphan diseases.

TSC is a genetic disorder that leads to benign tumors in multiple organs, including the brain, kidneys, heart, eyes, lungs and other organs. During the first few years, the severity of TSC can range from mild skin abnormalities to, in severe cases, seizures, mental retardation or renal failure..."

I do feel that both these projects require close watch, as they have had significant growing pains. There are still times where new target runs are not fed to the community fast enough to keep it properly fed.

I wouldn't call either project install and forget, but I personally consider the tsc cause to be more "worthy" and from experience with attitudes in other projects as well as a post or two above, I suspect others would have similar feelings.

Anyway, good luck in whatever you choose and best wishes.

Share this post


Link to post
Share on other sites
Damn I have no clue what the write-up would say.  PW...can you help on that?

Good one! You ask a non-native speaker to write an invitation for the project... :P

Anyway, not many words... A big banner saying sort of "SR has switched projects to reset competition and renew interest. Download the client at (link)... Join our team at (link)..."

The works. Small and functional.

BTW, once (and if) we go to D2OL, please resume your responsibility as Team Captain.  I'm sure you're a better one than I would be...

There's no need for that... Unless you're already planning for an early retirement! :(;)

The moment I'm back from uni, I'm installing the client. Sorry, too busy yesterday.

Share this post


Link to post
Share on other sites

I cannot believe you are worrying about having a competative 'TEAM' and will only go for the project where you have a good competative team!

You should be doing it because you find the cause good, the mentality of the people behind the cause are sound and the client doesn't bugger up your computer ;)

Not because lots of people may join it so your team will get in the high rank.

A team of 2 crunching small numbers to a cause they want to seems a better team to be in that what you have put forward.

A team only changing to another project because they have a change of getting into the high rank :(

Share this post


Link to post
Share on other sites

Good point. I believe the counter-argument is that we're also trying to raise the profile of SR in the online community. We can't (as)* effectively do that if we're spread across half a dozen different projects. The final decision on which project/team to support will always come down to your personal morals, ethics, priorities, alliegances, etc.

You could always offer to start an SR team in the distributed computing project of your choice. Especially if you feel it has something unique and substantial to offer over the existing projects SR has teams for. This could reduce SR's profile in any one project, but if it brings more people in (who weren't previously on an SR team), it seems worth it.

*Even if we have 500 teams of a few people each, that's 1500 people who will feel more connected with SR as a result. They might visit the site/contribute to the forums more as a result. We could have a few core projects that most of us support, but others could contribute to whatever project they like, and still feel they're waving the SR banner. There aren't really that many DC projects.

I like the idea that f-a-d lets you choose where your cycles are going from a list, but it all goes towards a single team. D2OL doesn't give you that choice, but works on a variety of projects. TSC sounds pretty socially responsible, too. I'm disturbed that the D2OL client takes up so much memory (up to 180 MB), though.

Share this post


Link to post
Share on other sites
I cannot believe you are worrying about having a competative 'TEAM' and will only go for the project where you have a good competative team!

I don't believe this was addressed to me but anyway I feel like responding here. This is called healthy competition. It's a small but valid reason to have this program running all the time. Competition makes as bind more machines and make the client run 24/7.

You should be doing it because you find the cause good, the mentality of the people behind the cause are sound and the client doesn't bugger up your computer ;)

Really? Then stop running SETI and join us at D2OL. Crunch for the real deal!

Not because lots of people may join it so your team will get in the high rank.

A team of 2 crunching small numbers to a cause they want to seems a better team to be in that what you have put forward.

A team only changing to another project because they have a change of getting into the high rank :(

Oh dude, don't you understand this is a further incentive? It's for the "sport" let's say. To have something to chat about.

Do you prefer America vs. Europe threads in the Bar & Grille? If yes, bring it on... :angry:

Share this post


Link to post
Share on other sites

Prof. It's just the way the thread was heading.

Nope Europe vs America threads, couldn't care less about them. We ALL know that the UK is the best ;)

I do understand why the competition i good and how it can benefit the project (they get more done). It the using that point as the argument to do that project over another...

Seti, I believe to be a viable project for me.

No I won't be joining D2OL I prefer the FaD so I am putting computing time towards that.

D2OL seems to have been chosen due to the probability of being able to get more people to join in so SR gets higher in the ranks (which of course in itself is not a bad thing). Nothing debating the reason for choosing it, ie. the funamental Science behind it.

In the end though it doesn't matter as it all turn out for the better in the end. Some project get's work done and other project get noticed by more people.

Share this post


Link to post
Share on other sites
I cannot believe you are worrying about having a competative 'TEAM' and will only go for the project where you have a good competative team!

You should be doing it because you find the cause good, the mentality of the people behind the cause are sound and the client doesn't bugger up your computer ;)

I think the point (from the top of the thread) was that the SR community effort for the previous work was dying off slowly, and the idea was to re-ignite the SR community by switching to a NEW project where we could easily attain a top ranking and get both more visibility AND new member participation. People like to do things in groups that WIN, that's just human nature. Few people want to play for the last-place ball club, even if they like the manager (the equivalent of the "cause" in this case). SR has put a LOT of machine cycles into the previous competition, and we are still stuck at 75th place - we started late. We should be able to get a top 10 or top 20 place with a new competition that we begin on the ground floor, and THAT should cause even more SR members to contribute (everyone wants to play for a "winner"). In the end, it's just human psychology, but if the END RESULT is to get more people to contribute more processor cycles, then regardless of which of several worthy causes we work on, the world is a better place...

Future Shock

Share this post


Link to post
Share on other sites

Lest this be interpreted as an endorsement of UD, I am not overly interested in which project is actually chosen. In the end, all of this is voluntary anyway, so it is hard to imagine that strong disagreements will see (or need to see) any resolution.

But it should be pointed out that the contention that interest in UD has waned doesn't exactly square with the facts. By my count, yesterday the team had its fifth highest point total for a one day period since the creation of the team.

Certainly it is true that only about a third of the team actually contributed points this week, but that could easily be symptomatic of a lack of interest in these projects generally. Who's to say that a lot of people will join a new project. And the major complaint on here doesn't seem to be the initial number of volunteers who joined the UD project, but the number who joined after it was up an running. If this complaint makes sense, then the real issue that we are facing is either a lack of interest in these programs generally or our inability to recruit effectively. Neither of these decisively points towards any particular path.

If there isn't much interest in these projects, we will fail no matter what. There isn't, therefore, much point in debating that premise. Instead, the real question we should be asking in evaluating any project with only an eye towards building the strongest team possible is which project to we expect to keep interest after we have been up and running for a while. Simply "resetting the competition" isn't obviously going to do this. And while it may be true that starting before most others will allow us to assume a competitive posture (thus creating interest), it is also true that we are likely to quickly reach a point where we cannot improve our rankings at all. If, however, we start late, as we did in UD, it is possible to improve the team ranking more dramatically.

Ultimately, the decision is pretty much a wash. I seriously doubt that a new project will tap into a massive amount of previously undiscovered interest, but by the same token it probably will get more people than currently are on the UD team to join the ranks. Better recruiting will help all the way around.

Share this post


Link to post
Share on other sites

NEWS FOR DUAL PROCS:

I just finished installing the client on my dual-PIII machine, and you need to do a double install for two clients to get it to use dual-processors - so the client itself isn't multiprocessor capable (total bullshit for a "grid" application IMHO). So do one install, then do another install into a DIFFERENT DIRECTORY (i.e., "c:\program files\SengentD2OL-Node 2"). Register it normally, then go to "My Nodes" in the D2OL web page and add the new node to your node collection as normal. It takes a few minutes for the stats pages to show newly added nodes...

Future Shock

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now