| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Suggestions / Feedback
Thread: Ideas on having survey to find out which research project to pick next. |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 54
|
|
| Author |
|
|
NixChix
Veteran Cruncher United States Joined: Apr 29, 2007 Post Count: 1187 Status: Offline Project Badges: 
|
Members do get mailed when a new project launches based on their communication preferences in settings [which under no condition but an individual emergency should be circumvented], NOT when beta starts as beta cycles can take months... Don't announce projects if there is no / not enough work, was the last much heard complaint.  I agree with Rob's points above (emphasis added by me). Cheers   |
||
|
|
twilyth
Master Cruncher US Joined: Mar 30, 2007 Post Count: 2129 Status: Offline Project Badges: 
|
On average do you know how many projects requests does WCG gets? I know that there are technical things that they need to develop on IBM side to make sure that it works. I'm curious about that as well. Given the cost of time on a supercompter I would think that WCG is deluged with requests. According to this 2009 Forbes article , people paid $1/core/hour. Even if the price has dropped to a half or less of that, it's still pretty expensive when you need thousands of years, never mind hours of time for your project. I doubt that anyone at WCG keeps track of rejected requests, but I'm sure they have some rough idea of how many they get and why most are rejected. My guess would be that the biggest issue is having enough work that needs to be done to justify the prep WCG requires. My guess for the second reason would be that most researchers don't have the expertise required to write for massively parallel processing. But those are just guesses. It would be interesting to know what the facts are, even if it's just a guestimate. Another problem might be the fact that all results have to be public domain. So even though there is an accelerating trend toward doing experiments in silico rather than in the lab, I can see that this could also be a barrier for some research teams.   |
||
|
|
vlado101
Senior Cruncher Joined: Jul 23, 2013 Post Count: 226 Status: Offline Project Badges: 
|
Another problem might be the fact that all results have to be public domain. So even though there is an accelerating trend toward doing experiments in silico rather than in the lab, I can see that this could also be a barrier for some research teams. Well that does make sense, but if you download either the BOINC or WCG program and go to select new projects there does not really seem to be shortage of them. Sure projects like Malariacontrol, GPU GRID, or Asteroid@home might have their own dedicated resources to just working on this project, but I don't think they keep their results private as they do publish their work (If I am mistaken please let me know). I guess I would have thought that if you are a university that can get access to multiple computer working like a supercomputer and that people volunteer to do this then it would be win for them. Yes you have to work with the techs at WCG to see if its compatible, but how different is that from them doing it themselves as you see some of the other BOINC projects. It seems like either the researches don't care (time constraints or looking for funding) or don't know about WCG because if all they have to do is to see if its compatible, then submit their idea and then focus on their work, while WCG takes care of the actual technical side of things then I can't see why they would not want to. That in addition that they see that volunteers here have already helped many projects (19 so far). Hopefully word gets out because this process is working. Yes there are issues to be hammered out, but a completion of 19 projects from different organizations is no small feat.  |
||
|
|
twilyth
Master Cruncher US Joined: Mar 30, 2007 Post Count: 2129 Status: Offline Project Badges:
|
For projects with direct commercial applications, no one is going to use distributed computing because of potential data exposure. Anyway, it doesn't matter for commercial projects since they can either buy or rent the processing power that they need. If you're Merck, you're not going to expose anything about what you're doing, not even what you're working on.
----------------------------------------All of the dist. comp. projects out there are purely academic. There isn't a single one with a commercial purpose - at least all of the volunteer programs. There are a view that pay you to run their software but in one case you only get paid via prize drawings and in the other, I've never gotten very much more than a couple hours worth of work. But if you can give me an contradictory example, I'd be interested. The point here though is that we don't know what sort of interest WCG generates among researchers and it's likely we'll never know since WCG rarely deigns to tell us jack s***. |
||
|
|
Coleslaw
Veteran Cruncher USA Joined: Mar 29, 2007 Post Count: 1343 Status: Offline Project Badges: 
|
twilyth, when you say distributed computing projects, do you really mean ALL Distributed Computing? Yes charityengine is like a drawing. There is also majestic-12 that donors would get payouts if they chose based on their contribution percentage over a given time. That project just recently closed its doors to new members. However, current members can still get a payout each quarter. Then all the crypto currency generators (yes the Bitcoin generation) is all distributed computing.
----------------------------------------As far as commercial entities, well there is again Majestic-12. There was AQUA@home which I believe was commercial. BitcoinUtopia is commercial since it is nothing more than a money generating project and is a for-Profit. And these are just a few. So, I think it would be easy to argue that not all of dist. comp. project are purely for academic.     [Edit 1 times, last edit by Coleslaw at Jun 23, 2015 2:38:06 AM] |
||
|
|
twilyth
Master Cruncher US Joined: Mar 30, 2007 Post Count: 2129 Status: Offline Project Badges:
|
Thanks Coleslaw. I should have been more specific. I only consider something a DC project if it's doing useful work. Mining cryptos doesn't really fall into that category. Also I specifically limited my comments to volunteer projects. If you're getting paid, you're no longer a volunteer.
----------------------------------------The only exception to this was the brief 5 months or so that Ripple Labs was giving people XRPs based on how much WCG work they did. That arguably falls into a gray area. The ones I had in mind were things like Charity Engine and Slicify, both of which I mentioned in my previous post but by description rather than by name. CE probably qualifies since they are supposedly doing useful work but you won't necessarily get paid anything. Slicify doesn't qualify since in that case, you are paid based on the hours your machine is used. I did a quick search and it seems that there are now one or two more organizations like Slicify that pay by the hour. edit: looks like slicify might now be defunct. Couldn't get to their home page. [Edit 2 times, last edit by twilyth at Jun 23, 2015 5:14:08 AM] |
||
|
|
Coleslaw
Veteran Cruncher USA Joined: Mar 29, 2007 Post Count: 1343 Status: Offline Project Badges:
|
Well Majestic-12 was originally trying to develop something new. So it would most certainly qualify and the donors didn't originally get any monetary compensation. That came later from one of the employees divying his own shares. Now many of the older donors are share holders. However, it was a commercial entity from the start.
----------------------------------------Doing useful work is very subjective. Can't really argue or debate when such terms are thrown in there. |
||
|
|
vlado101
Senior Cruncher Joined: Jul 23, 2013 Post Count: 226 Status: Offline Project Badges:
|
The only exception to this was the brief 5 months or so that Ripple Labs was giving people XRPs based on how much WCG work they did. That arguably falls into a gray area. I would have to agree that I had mixed feelings about Ripple Labs as well. When you introduce money into equation it can sometimes bring out the worst in people even if it is aimed at a noble cause. I also do not have issue supporting academic projects. Its important to do research that is not solely aimed at profit, but advancing pure knowledge as well. I guess its a bit of a discouragement that there is not really enough of those to go around or that they face such walls to actually do this. However, I do like WCG alot more than going through projects at BOINC. I don't think each project should be its own world, but a new person seeing WCG can see that 19 projects were completed and read about their publications. Otherwise they would need to know a specific BOINC project and follow that exclusively and do research on their own. Even I wish I knew about WCG earlier than I joined and contributed to those first projects. |
||
|
|
kurioryuu04
Cruncher Joined: Jun 24, 2015 Post Count: 1 Status: Offline Project Badges: 
|
I understand the situation where organizations and universities have to use WCG to complete complex tasks. The one project I think would be very helpful to the world is mapping the ocean floor. That is a project I would get behind in a heartbeat.
|
||
|
|
Still Alive
Advanced Cruncher Germany Joined: May 27, 2013 Post Count: 69 Status: Offline Project Badges: 
|
What about projects like Fight Malaria @ Home (www.fight-malaria.org) or Docking @ home (project of a university)?
----------------------------------------You are already running with Boinc and I think that it would be suitable candidates for WCG.  AMD 3900X - 21 Threads @ 4,00 GHz - 24/7 Samsung Xcover4 - 4 Threads @ 1.43 Ghz - 24/7 |
||
|
|
|