|
 Picked this up at the main WCG Forum today. Just wanted to show it off here just in case anyone missed it, and for all members of 'other' projects to see the full extent of WCG members contributions. Keep up the good work at WCG - we're members of the hardest working team and we get the best feedback from our scientific users.
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
Thanks EvoDude. I did miss it, I've been so busy.
|
|
|
|
|
|
. . . we're members of the hardest working team and we get the best feedback from our scientific users. Aye, that Dr Bonneau is a top bloke alright 
|
-------------------------------
|
|
|
|
 The latest feedback from WCG on the their forum shows us fast approaching project completion with WCG contribution showing our tremendous effort to their work. Keep it up folks - they do appreciate usThere is also some other interesting info just appeared there that makes interesting reading. I'll just add the link and let you all read it for yourselves.
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
|
-------------------------------
|
|
|
|
I like the human element............these guys give information  but also manage to be fun at the same time 
|
-------------------------------
|
|
|
|
Yeah, but only the WCG forums fing - still haven'y visited the other place since I left 
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
so...i just read over there, and am curious. dr bonneau is working for isb, right? not wcg? and isb is using both wcg and grid to complete the rosetta project. however his comment here is curious...is this a comment to be seen against grid, and therefore against one of the people his own project is using to complete rosetta?  i still stand by wondering why he posts all sorts of updates at wcg and doesn't say a word at grid...
|
-------------------------------
|
|
|
|
Well since when has grid mentioned ISB?
|
-------------------------------
|
|
|
|
Current Project: Human Proteome Folding United Devices has begun a new and exciting research project -- the Human Proteome Folding Project -- in collaboration with the Institute for Systems Biology, the University of Washington, and IBM Corporation. 
|
-------------------------------
|
|
|
|
I repeat........since when?  ? No doubt you pulled that off of an official announcement. What since that initial announcement?
|
-------------------------------
|
|
|
|
regardless of whether or not grid mentions them...is it not curious that the people running rosetta completely ignore grid, yet pay so much attention to wcg? same people running the project, whether they're mentioned or not shouldn't change whether or not it would be nice for them to at least copy/paste their updates in a thread at grid forum or something... 
|
-------------------------------
|
|
|
|
Dunno....not into politics.
But partnerships are supposed to be a two way street.
Don't see/hear much from Washington Uni either
|
-------------------------------
|
|
|
|
I just think it says a lot for level of respect towards the relative crunching teams He He, I can dig at 'the other place' here 'cos, for once, I'm in the right thread. 
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
i dunno, there's something fishy about it all, especially when throwing in that comment i linked to.... even if he did think better of one over the other, posting it as he did i think is a bad step...
|
-------------------------------
|
|
|
|
Sorry PiNkY, missed that link first time through - I read it as the extra work was at World Community Grid especially as I know he does not place these posts in 'the other place'. After all, if you wanted the extra work done quickly and efficiently would you waste your time issuing it to 'the other place' They would take 3 months just getting servers ready to run it then they would crash and not be able to get any of the work back from crunchers. While all this was happening WCG members would have it all done and dusted. I accept I'm male but you know I'm still right! 
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
think i missed where you were going with your post, evo  no argument that wcg seems to get this done faster then grid (why this is true is another oddity...look at the stats between the two and explain how what the statistics show and what the chart posted shows relate to one another...they don't seem to match up at all).......no doubt that wcg will pull through the last bit of work faster then grid. but...going from my earlier post...is this not like stabbing one in the back? have someone do your work, then go say things elsewhere that puts them in a negative light? though what he says may be true.....i still say, poor show by that post...lost a bit of respect from me with that there. if fad or folding were to say similar, it wouldn't bother me...be proud about it, fine. but he represents isb, which i would think should be working equally with wcg and grid. then makes that comment.  note i'm not saying this to purposely support grid, more out of the interest of fairness...something's not right, though people still seem happy with it all. just rocking the boat a bit.
|
-------------------------------
|
|
|
|
I would have to agree with Pinky, I enjoy Dr Bonneau's posts and think it's great that we (WCG) get such feedback. But I did wonder at the apparent digs at Grid.org - or at least I read them as such. It is possible that the end users of the data, especially ISB, have had similar problems with UD like the crunchers have. My feeling from the postings by the WCG techs and UD Robbie Brewer is that WCG is more approachable - for want of a better way of putting it. If Dr Bonneau has to deal with these people then he may be venting some frustration. If he is then it is a little unprofessional - but should make for interesting reading 
|
-------------------------------
|
|
|
|
note i'm not saying this to purposely support grid, more out of the interest of fairness...something's not right, though people still seem happy with it all. just rocking the boat a bit. Yeah, I know what you mean PiNkY - my comment was slightly toungue in cheek to rock the boat a bit too I do agree that all projects involved should be treated equally but don't tell anybody I stuck up for 'the other place' 'cos I will deny every word of it! 
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
dr bonneau is working for isb, right? not wcg? and isb is using both wcg and grid to complete the rosetta project. however his comment here is curious...is this a comment to be seen against grid, and therefore against one of the people his own project is using to complete rosetta? I would suspect so. I imagine, as some others have suggested, that ISB may be frustrated with the shenanigans that have taken place at Grid.org during 2005. I personally don't see anything wrong with Bonneau’s statement. He never indicated any specific grid effort so unless you came from Grid.org and know about their troubles you wouldn’t think he was talking about them. I will however agree in lacking an understanding for why he doesn't at least cut and paste his progress posts over to Grid.org forums anyway. If people over there were regularly confronted with their diminished HPF contributions more questions may be asked and more answers received. Considering all the turmoil about productivity this year at Grid.org I know that if I had seen the discrepancies between the two grids in the progress bar sooner, while I was still at Grid.org, I would have felt appreciative. One additional thought...doesn’t the primary link from Grid.org for the HPF project go directly to WCG? Perhaps WCG really is ISB’s primary focus. 
|
|
|
|
|
'the other place' would probably delete his post as an advert for WCG. :idea: Maybe that's why they never see it over there! After all, they are leaking members to so many other projects already that they perhaps daren't let remaining crunchers see this..
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
Actually, while the thought is kind of cynical  , you may be right.
|
|
|
|
|
i still can't read that post without multiple thoughts going through my head...perhaps it's just late, but as i reread it this time, i laughed... Many grid projects run out of work and resend redundant work units to keep their volunteers from migrating or loosing interest… NOT HPF! if grid is losing people due to redundant cancer units...they must not have interest or have lost interest in hpf...  (it's late, i'll probably reread this in the morning and see something different) right now i really want to post the open-ended question of what his stance is in regards to grid.org's status, as it may help either confirm or clear up thoughts, though i'll likely fumble up the post.  'the other place' would probably delete his post as an advert for WCG. :idea: Maybe that's why they never see it over there! After all, they are leaking members to so many other projects already that they perhaps daren't let remaining crunchers see this.. lol. the posts that have made their way over from wcg have stayed so far...despite the most recent ones wcg plugs 
|
-------------------------------
|
|
|
|
Good points. If both Grid and WCG are outlets for ISB why in the world would they favor one over the other? Maybe someone needs to pose this question to the good doctor himself.
|
|
|
|
|
Good points. If both Grid and WCG are outlets for ISB why in the world would they favor one over the other? Maybe someone needs to pose this question to the good doctor himself. Not me oVo - I fully understand his attitude, if not absolutely agreeing with his methods.  WCG - the place to stay informed and in-form. 'The other place' - the place to be out-of-touch and ill-informed  Sorry folks but they'll never get sympathy or reasonable understanding from me.
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
It is your right to say how you feel...And the great thing about this forum is that it will not be deleted....By the way, I agree with you on the subject about the Grid.....
|
|
|
|
|
Oh, I understand why crunchers feel that way, that is crystal clear, but why ISB or IBS or whatever does, and shows it, and continues to work with a partner with whom they are unhappy, that I do not understand.
|
|
|
|
|
Oh, I understand why crunchers feel that way, that is crystal clear, but why ISB or IBS or whatever does, and shows it, and continues to work with a partner with whom they are unhappy, that I do not understand. Depends on how they are treated, If ISB gets cold shoulder like the members do..... If they have contract, they may not have a choice. That or they don't have anyplace else to go for the processing of thier research at this time. My speculation is that Oxford may be the cancer research provider that WCG has mentioned.  I haven't seen any details on that one yet.
|
-------------------------------
|
|
|
|
Mousies point remains valid.
Even if a party is disaffected by the other there is no reason why said person cannot post something sans the sleight.
|
-------------------------------
|
|
|
|
Yep! She has posted the question at WCG forum, will be interesting to see the response from above!
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
guess i'll just hold to the last line of that post about it being the opinions of a sleep deprived scientist, and hope a better understanding of the grid/wcg differences comes about.  a nice informative reply from viktors, though adds a bit more to the confusion assuming i interpreted right...the 6 results redundancy thing. seems disk space certainly is a problem when it comes to the rosetta project overall...
|
-------------------------------
|
|
|
|
Hi, thanks for the interest. As to the difference between UD and IBM grids i would not want to comment right now, as I work more closely with IBM, and know less about the UD side of the grid.
one additional explanation is that UD runs multiple projects and is a small company compared to IBM, thus they have devoted less of grid.org to the HPF project and more to other projects (i don't know what those projects are).
sorry I couldn't be more informative... I guess that answers all our questions as to why nothing posted at 'the other place' then. No subterfuge, no slight on 'the other place', just a simple case of whoever liases with 'the other place' doesn't bother to post the same level of information. Rather typical of everything to do with them!
|
-------------------------------
|
|
IP: [
Logged ]
|
|
|
|
Sounds more like an ignorance of the DC world to me. Read between the lines and a further sleight is revealed.
|
-------------------------------
|
|
|
|
yep...also note that with grid contributing 50% more cpu time then wcg(and thus, i would disagree with the less devoted thing, or at least keep in mind that one project has been working more then another, just not as efficiently or whatever is happening), i would think had he been keeping an eye on things at grid and making sure the projects were doing the same, rosetta would have been completed months ago... 
|
-------------------------------
|
|
|
|
posted again...not sure why i'm keeping up with it since it's not my project of choice....guess it's more the hope that something i've contributed to in the past is doing ok...
|
-------------------------------
|
|
|
|
Be more straightforward if one agency ran one project instead of it being farmed out in bits..........the left doesn't seem to know what the right is doing.
|
-------------------------------
|
|
|
|
that's kinda become my conclusion on this...
|
-------------------------------
|
|
|
|
I am just thankful that the FaD server is more reliable than their forum. Seems that all of the projects have their lil quirks.
One thing that bugs me at WCG is the trumpeting and flagwaving of IBM.
I don't give a damn for IBM...........There to get work done not kiss up to Big Blue.
|
-------------------------------
|
|
|
|
I don't mid Fad forum probs myself.... We have this one. Pulled my last WCG rig off that project after the last unit netted me a total of 67pts. Something goofy goin there. Well, more cycles for FaD. 
|
-------------------------------
|
|
|
|
Was it a long unit? 67 points indicates a unit of 2-3 hours duration.
|
-------------------------------
|
|
|
|
About 3hrs if I remember correctly. Still the first single unit I've seen that low.
|
-------------------------------
|
|
|
|
points are based on cpu time...small cpu time = small points. it's why slow machines get tons of points and small results count...same thing with grid.
|
-------------------------------
|
|
|
|
Nah, with grid, you can crunch for a week and get 0 pts. 
|
-------------------------------
|
|
|
|
About 3hrs if I remember correctly. Still the first single unit I've seen that low. Let me guess P4 circa 3.2 ghz ?
|
-------------------------------
|
|
|
|
2.8ghz...... Still trying to make up my mind is I should up this rig to 3.4ghz or just build a new box... Might just have to give AMD another go. 
|
-------------------------------
|
|
|
|
recent posting by mycrofth in the call for questions thread at wcg commenting on curve balls thrown at dr bonneau has me wondering if my questions are seen as curve balls? pardon me for what i think is being validly concerned about the project.  seems he just likes to think he's doing good and dismissing an interesting, though possibly disturbing, fact that could be important to the project overall....and if it isn't, it would be nice for them to at least explain why... am i alone in wondering if there's some sort of standard set to make sure data turned in by both grid.org and wcg are similar enough? since he couldn't answer previous question, no doubt it's another "i don't know".....and maybe there's nothing to worry about since both are theoretically using the same thing, or perhaps it's a silly concern....but i think knowing the differences between the two is something he should have known even before i asked...  tell me if i'm being silly here in wondering this... hmm...grid.org - say something that may put grid in a negative light, and you get deleted. wcg - say something that may bring out a negative side, and you get talked about elsewhere... (i suppose i am doing similar in talking about elsewhere here, though i hate to see what happens if i were to post a bit more on this over there  ) *sigh*
|
-------------------------------
|
|
|
|
The way they answer questions at both UD software running projects leads me to believe they are both a bit more commercial than they want us to know.
|
-------------------------------
|
|
|
|
Hey mousie, he'll love ya even more when you tell him you prefer Folding I considered posing one or two questions not long ago but really can't be bothered. Answers from both parties would either be dismissive or mere political nonsense. UD is mainly for the gain of United Devices WCG is mainly a vehicle for big blue to say how wonderful they are. I am starting to feel a tad jaded at the whole scene. I often feel that the actual work is secondary. But that's just me :|
|
-------------------------------
|
|
|
|
lol...  in some cases, the work can be seen as secondary to me as well, though this has kinda piqued my interest, despite it not being my project of choice. it would be good to feel reassured that the time i did put in isn't going to waste (well, i'm sure regardless of how the issue resolves, it won't necessarily be a waste, but you know what i mean  )
|
-------------------------------
|
|
|
|
I think that all you people spent too much time at the Grid...Every project has its little quirks but in the end WCG is still better then Gid was or is.....I think that we fail to realize that someone has to pay the bill for all those servers to be online...WCG has never poured "Big Blue" down my throat....Mousie I do not say this to hurt your feelings but I think that you are reading between the lines where no lines exist....
|
|
|
|
|
i may be reading between nonexistant lines...i thought about it, posted here instead of trying to jump on him there right away...left it sit for a while, then figured it'd be best to ask directly. kinda wish they had pm features there. i still don't see how him telling dr bonneau that one specific question was important, and then telling him to ignore "curve ball" questions was not telling him to ignore my question. that's just me, though. as i mentioned in my initial reply to mycrofth, this isn't something against wcg. it's me trying to understand a few things rosetta-related. i could very well ask the same questions at grid.org, but asking there would be futile. i honestly do think i have a valid concern, though lack of support makes me wonder if perhaps i've just gone nutty. not taken as hurt feelings at all...i've been in an awkward state of mind recently, so it may help explain the recent spark.
|
-------------------------------
|
|
|
|
I can see Pinky's and Scampi's point but agree with Guru - especially about payment for the technology.
Dr Bonneau's post regarding other sites serving up redundant work units is IMHO ill-considered, but surely not meant to be malicious or under-handed. I am sure that all DCZ members have done similar in the past and regretted it.
|
-------------------------------
|
|
|
|
Wether a question is considered a "curve ball" or not a question still deserves an answer.
What is and isn't a valid question is all a matter of personal opinion. If people are in place to answer questions then answer they should.....not pick and choose which questions merit response.
|
-------------------------------
|
|
|
|
Well…I’ve wanted to comment on this issue since it came up but I have been exhaustingly busy and I surprisingly found my first reaction toward the response of Pinky and Scamp to Dr Bonneau statements at WCG forums to be a bit bewildering. I sincerely did not want to aggravate the situation nor go with what I felt must surely be my sleep deprived misinterpretation. With a bit of a rest bit under my belt I’ve reread the threads at issue at WCG and I found that pinky did just fine in my opinion.  There is a legitimate question that pinky is trying to ask and it was asked clearly enough to be understood. Ultimately it does appear mycrofth misinterpreted the demeanor of pinky’s posts similar to the way I initially did. Anyway, the answer MAY be something as simple as Grid.org re-issues old work often enough to make for this discrepancy in time in and work out. That would be a lot though! I don’t know, but if you push Dr Bonneau long enough you may force him to bad mouth Grid.org if the above postulation is the case (I doubt he ever would). I’m not saying you shouldn’t try but I wonder what would you say then? I think that Vaio’s response to Dr Bonneau’s follow up… Hi, thanks for the interest. As to the difference between UD and IBM grids i would not want to comment right now, as I work more closely with IBM, and know less about the UD side of the grid.
one additional explanation is that UD runs multiple projects and is a small company compared to IBM, thus they have devoted less of grid.org to the HPF project and more to other projects (i don't know what those projects are).
sorry I couldn't be more informative... … came off being a bit sharp and put offish. No doubt there is a legitimate question to be asked with regards to this and perhaps it would have been more productive to ask it. One point to remember is that Grid.org itself sends all prospective new HPF volunteers to WCG so WCG may really be the primary focus for Dr Bonnaeu and this project. I don't know. In any regards JC’s post at the end came across to me as nothing more than negative piling on which spun the entire thread further into negative interpretations where ‘curveball’ then came in. He has edited the post to remove most of his commentary since. As far as Dr Bonneau’s statement he’d never put out old work as filler like some other projects have? I personally found his statement more as a positive affirmation that our efforts with the HPF would remain productive than a negative sleight on others. Finally I’ll say that I agree with pwrguru about IBM. I am happy any company is promoting distributed computing regardless of whether they get some name recognition out of it. To date I have not felt IBM is stuffing themselves down my throat either. 
|
|
|
|
|
I never said IBM are touting themselves endlessly..........my point was that the community admins tout IBM endlessly.
Anyways, as long as the work server works there endeth my concerns.....not interested in politics right now, just in getting the work done.
As for figures from grid.........they are meaningless anyway. They are never corrected so member stats etc incessently increase regardless of "true active" membership.
I think folding re-evaluate the "active users" figures every few months which is pretty cool.
|
-------------------------------
|
|
|
|
just a random clarification....something i've been making an effort to do is keep that my concerns are with hpf...wcg itself hasn't been around long enough for me to have a valid opinion on, imo. this is one of my main reasons for preferring folding over wcg....one has been tried and true, the other is still very fresh in my mind, not even a year old yet.
|
-------------------------------
|
|
|
|
I never said IBM are touting themselves endlessly..........my point was that the community admins tout IBM endlessly. Point taken. 
|
|
|
|
|
Reading this thread from beginning to end for the first time..... Both United Devices and IBM are competing for corporate clients in the lucrative Distributed Computing market. To a potential customer, it is obvious that they can only showcase the public DC applications, so is it any wonder that a little puff is manifest, and a slight aversion to "critisism" found. Perhaps, what many potential customers, (or more to the point, their IT departments), do not realise is that in-house DC clusters can be set up gratis, and that no specialist parallel programming expertise is needed! Such a system is provided by [email protected], details of which, can be found here.
|
|
|
|
|
can this software [email protected] be used on WCG clusters? I've got 11 running it all linked via the 100meg connection its talking about, and it seems as if it would be possible 
|
|
|
|
|
Unfortunately, in this instance the work units cannot be broken down into smaller subunits, and even if they could be decomposed, there would be the problem of generating a valid output file from the partial o/ps.
|
|
|
|
|
ok, just wondered,
hob has been talking about it in my Networking thread in the Hardware section of the forum - but using FaD
|
|
|
|
|
|
-------------------------------
|
|
|
IP: [
Logged ]
|
|
|
|
That doesn't equate to being "make work" does it? Deja vu :|
|
-------------------------------
|
|
|
|
That doesn't equate to being "make work" does it? Deja vu :| Don't think so. This plant data was mentioned in the official WCG forum some time ago. It is all new data and was added because the human work was ahead of schedule. With all the new members from easynews there now sucking up work for profit, having enough work to go around could become a problem though. 
|
-------------------------------
|
|
|
|
On the positive side............more members may justify bringing more projects online
|
-------------------------------
|
|
|
|
this is a make-or-break time for the project, they need to make sure it goes through with little hitch so not to upset its members, don't want another grid.org fiasco
|
|
|
|
|
The project is very nearly a year old and has almost 100-000 registered members.
Wonder what percentage are active. A regular update of the "active" device count will give a clearer picture.
|
-------------------------------
|
|
|
|
I had brought up the inacurate nubers via thier suggestions form. It's something they are working on for early next year. They want to make sure they don't kill any inactive members that stop in to crunch every now and then and howerver thay do it can't mess with the current points standing of the teams and active members. Tossed some ideas at them on updating ste stats engine to ignore inactive members, and got the "We'll add it to the list" feedback e-mail. They are talking of a purge of inactive users and devices. Hard to keep the stats that way. I suggested of adding "Inactive" flag to members who haven't provided any work fo 4 to 6 months. Then having the stats engine not report those in the global stats. Also mergining all the old inactive devices in to one for each user. Keeps the stats, and reduces the "extra" devices most of us have. There was a bit more to it but that's just the basics. Don't want to delete any members in case they come back, and one sure fire way to upset the members is to "misplace" some stats. Guess we'll just have to wait and see which direction they take.
|
-------------------------------
|
|
|
|
Can they not simply have two stats:
All members Currently active*
*eg In the last 3 months
|
-------------------------------
|
|
|
|
Don't know... Depends on what the database/server software will let them get away with.
If it's done corectly, other than the amount of time the stats update will take, there should be no reason I can think of that it can't be done weekly.
Most DB software has some pretty powerful things you can use to keep them organized. The biggest question is what can they do without upsetting Uninted Devices. All that copyright mumbo jumbo may limit WCG's choices on what they can do.
|
-------------------------------
|
|
|
« previous next »
Forum Index Unread Posts
|
Print
|
|