I have to question, after this latest CEP support fiasco, my continued involvement in WCG. This is the second time in a year that, in my opinion, the WCG support system has just plain refused to provide complete answers on a significant support issue. What a marketing nightmare to then leave the CAs hanging out in the cold with no Tech backup.

I understand that the science is important, but there are no results unless you have crunchers. I call that a partnership. If my 18.5 years doesn't rate a response to a simple straight-forward question...............

There a lot of people in WCG working to get the job done. Why is there nobody home?

BTW my question was (paraphrase): Why are certain (identified) series of CEP WU still being sent out since we have been told they were aborted by the server?

Steve -- Please see Keven Reed's response and apology for this here.

The techs usually keep things going pretty well here IMO. Think we can give them a wee break with this one guys. It was just a bad time for the problem to happen. Its ironed out now.

Steve, I hope I didn't sound that way. As I also crunch at a number of other projects and WCG is definitely one of the best run projects. Now that doesn't mean it couldn't be better but the room for improvement is very significantly less that the possible downside. The vast majority of DC projects are academia based. They probably have to rely on very limited university funding and grants from the outside that are not limited and are for a limited term. As such, I find that many of these projects are run with a very minimal staff. The top techie is often a grad student or research assistant. They may have one, in some cases two, folks that are developers - folks that update/improve the website and stuff behind the scenes at the projects. In some cases, the grad student techie has to do that too. Students will usually have to put their coursework and studies first and understandably so. I've seen a number of news posts at projects along the lines of "I'm gone for the next week so if the servers crash or there's a problem in the meantime, I'll address it when I return".

We're fortunate that WCG is run by a group of folks that really do care. Because WCG is run by IBM, it's purely an expense to them. They don't make any revenue off it. I'm sure they use it as a showcase with their customers about their grid computing capabilities. Still, you're looking at probably 6-9 folks that work on WCG at IBM and I doubt that many of them are full-time - they have other work they do too. In today's economy, we're lucky that it's still running if you ask me. Clearly WCG has to have a friend in high places, at least for now. Now I'm not trying to make any excuses, just trying to show the other side of the coin ( there are no one-sided coins).

I agree that communications, as good at that may be compared to most projects, could be improved. WCG has grown rapidly in the four and a half years it's been around. Only SETI is bigger. Most WCG crunchers never even bother with the forums so those of us that do are a minority. Still, the point of Member News and Known Issues was to be official sources from WCG. Most of us won't always realize if a problem is just with our own machines or with one paticular project at WCG and with several projects running at the same time, looking thru all the recent threads in each project's forum is not a reasonable expectation. I believe the original expectation was that us crunchers would use the support@worldcommunitygrid.org email address and though I don't know any actual numbers, I suspect a lot of problems get reported here in the forums. Maybe that's wrong on our part but it has to reduce duplication. Any of us should be able to rely on the Member News and Known Problems forums. If we don't see somthing there, we should be able to presume that the admins are not aware of the problem or the problem is unique to the user and/or their machine. Once the admins have multiple reports of some problem/issue from us crunchers, it should be acknowledged in Known Issues as a possible problem that they are aware of and are investigating. We would simply then have to monitor Known Issues for updates. Since we can't create threads in those two forums, we open them in the project specific forums or in Website Support or BOINC Agent Support. Putting a link in the thread there to their acknowledgement and updates in Known Issues would be simple. If they need more information, they could still ask in the forum with the original thread while the rest of us wait for an answer in Known Issues. This would allow us to quickly figure out if we're seeing a problem that is not specific to us and if the admins are aware of it. It would make for even better communication too. If there aren't multiple reports of the same problem in a forum, it would certainly be reasonable for the admins to presume it is specific to the user or their machine, at least at first and those would be prime fodder for the CAs to help with. If one turns out not to be user/machine specific, chances are the admins would get enough information to at least narrow things down.

I think it's a very reasonable expectation for us crunchers to expect to be able to rely on Known Issues and Member News for being aware of serious problems, changes, outages and such.