I've been a SETI participant since 1999. It runs, it downloads and uploads, nice graphics, I'm happy - or was happy. But I had no idea of the BOINC software or any SETI changes until I recently checked my stats. No stats since March? What happened? BOINC happened. So I downloaded the infantile-sounding BOINC and ran it. Very confusing interface, not intuitive, ugly icon in my dock, it's SLOW, and the graphics that show it running (the "show graphics" button in the "tasks" menu) are so awful that they are unreadable. Why should I continue running it? The SETI people apparently thought it was a great idea to share processors across a spectrum of projects. Maybe it was, but given the lousy interface and other factors, I'm not sure I want to be involved in the SETI@home project any longer. And if Berkeley checks their user stats, I'm guessing that since BOINC, the active users/PCs have plummeted. Yes, I know all this has been discussed before, but I didn't know about any of this prior to three days ago, and it's mildly cathartic to post this. I don't want not to be involved after six years, but until SETI/Berkeley makes some big changes to re-capture the benefits of the old software, I think I'm going to sign off.

---

I think, so do I.

---

Think I am gone also. Least for a while. Until they can get a good working version. Causes to many conflicts on my PowerMac G4 system running 10.4.2 Have tried everything in the help and here to get it to work right with out causing conflict. Even tried compiling from source and still no joy. Hopefully they will get it worked out. I will miss the project for now.

I think, so do I.

See my reply on "...make Macintosh enjoyable...", full of typos as it is. Wonder if a lot of us wouldn't improve things by getting Mac programmers to add to Open Source code and just make it a client that stands alone but does what BOINC does? Would Berkerly say no? BOINC for MAC X.4.x?

Best to BOINC as they try to improve the Mac version.

---

Robert L. Hanson, Ed.D.

Think I am gone also. Least for a while. Until they can get a good working version. Causes to many conflicts on my PowerMac G4 system running 10.4.2 Have tried everything in the help and here to get it to work right with out causing conflict. Even tried compiling from source and still no joy. Hopefully they will get it worked out. I will miss the project for now.

=====
Amazing, just as I was going to remove Climate (not sure how except drag to trash library files, etc.) I went to the Wiki and saw some things that I tried. Wiki is a work in progress, so a look in there is worthwhile now and then.

1) I didn't know that the last project changed becomes the default preferences for all the rest; one doesn't need to change each project preferences (local, gen, climatenet). Finding that Climate takes the biggest share of resources I used these settings: A. Your Account>View or Edit>Use No More Than> Type 750 MB. Leave at Least 2.7 GB for the current Sulphur project. (If doing Sulphur use 2.7GB not 750 MB!) Use no more than (typo in Wiki) should be "9" of TOTAL SPACE. So type in 9% of TOTAL SPACE. I set the machine to function at 2 am to 4 am; but then did allow it to crunch when on.

2) Check all three preferences: local, gen, climate.net and make them all consistent as possible.

3) Setting space preferences are not %! but it appears there is an equivalency going on in the code. Setting Resource Allocation (that I can not find) apparently, balances out the correct numbers with Space Used. Understand that is an efficient use but it means Resource Allocation is not so important. Since I couldn't find Resource Allocation I made the Disk Use the primary tool for me.

4) I think I see now that old image of sitting in a lab next to the big IBM machine looking at my dumb terminal...I had to change preferences constantly to get it to do my crunching/programs. This client is the same...(how out of date!) and it means if you use the default 750 MB (No More Than) you will have to change it when running unknown size usage programs. It is suggested to set it to higher amounts. I just went to 4 GB...but one will have to check it from time to time. The end user who enters the data should have had a choice to make it large/small and inform the Crunchers (US) before d/l...etc., that could be improved. If you set it too low the project won't run; no space on your hard drive! That is why Climate didn't run for four months and why so many Mac folks just gave up! But, I am not sure the Wiki addressed this then. Keep up the good work as you add more on the Wiki! I even thought of rewriting the WIKI, to make it friendly, but didn't want to step on toes.

5) I have tried it and everything works for the past two hours, (the first time since installing BOINC); even Climate.net project. It seems to hold SETI until the processor is finished with Climate and then reports and d/l new data; but not always in order of last project. My preferences were way larger AND/OR smaller than in 4) before...hence, no wonder it didn't work. So, I don't have to put it on my Windows machine nor delete it from the Mac.

6) Before you leave; take a look at the WIKI...it I can make it work, like other programs, I can deal a bit with the poor interface that surely has to improve. And while I am not sure it will survive the 2 am work period I have scheduled...Clearly, today my conflicts with BOINC have ended! I WILL MAKE SURE I RESPOND TO THIS THREAD WITH MY RESULTS OF THIS SMALL EXPERIMENT! IT STILL MAY NOT CORRECT, BUT THINK THIS IS A MOVE FORWARD.

Best,

Dr. Bob

---

Robert L. Hanson, Ed.D.

Installed BOINC on my G5 10.3.9 yesterday because of an email from setiathome.

What a mess! Nothing works. I enjoyed helping with seti before this and did 1150 units of work.
I'm not into learning a whole new program with confusing interface and a ton of options. Just want to help out without having to spend time on it personally.

You can get back to me if you ever return to something that works automatically and simply.

Until then I'm deleting all traces of BOINC and setiathome I can find on my machine.

This seems like a severly retrograde step that was not thought through and debugged before release.

I wish you luck in the future but I don't have time to mess with it.

Over and OUT :-(

---

What a mess! Nothing works.

Not sure what problems you had, but you managed to correctly connect to the SETI project and download your first WU for processing. BOINC is definitely "more complex" than SETI Classic, and when I first came over, I had very much the same reaction as you - of course, at THAT point, it was command-line only, there wasn't even a GUI... It has improved a _lot_ in the last year. And all the complexity is for a reason. Instead of just a pretty screensaver and contributing some CPU cycles to SETI, now I'm attached to FIVE projects that interest me (and have Mac versions).

We're sorry to see you go. If you decide to give it another try, we'll be happy to help you with any problems you encounter, and there is also the Wiki with tons of information at your fingertips. (And yes, it is "more Windows than Mac" centric, but only because of the numbers of users, not because of any bias on the part of Paul D. Buck, who prefers Macs...)

i also have been a regular seti@home for five years and was happy with the old screensaver with its blue and red graph - i really liked it!! but having just downloaded the new software and tried to configure it, i've decided to bin it. i like to think that by doing their number-crunching for them, they should provide a simple piece of software with an easy method of configuring it, but that doesn't seem to be the case. unlucky set@home - you've lost yet another fan....

---

was happy with the old screensaver with its blue and red graph - i really liked it!!

Had you asked, we would have been happy to point you to the Wiki, which has instructions on how you can make the new screensaver look more like the old one.

I am continually surprised by how many people take one look at the new software, and without bothering to ask questions, or read the boards, or read the Wiki, just say "I don't like it" and leave. Yes - it's more complicated. It also DOES a lot more. To use a bad analogy, SETI Classic was like TextEdit. It did one thing, and did it well. BOINC is like Microsoft Word. The interface is over-complicated, it definitely takes some effort to get used to - but once you have spent even a LITTLE time with it, you can do far more than you could with TextEdit.

The problem is that there is no funding to continue SETI Classic, and certainly no funding to expand it, adding to the ability to detect fainter signals, etc. - it's past it's useful life, as great a job as it has done. The new BOINC architecture, by allowing MULTIPLE projects, is far better from a distributed computing view - and far more able to attract funding. SETI is just "riding on the coattails" of BOINC - if it weren't for BOINC, SETI would have shut down due to lack of money a long time ago. As it is, it gets to continue, and even be the largest of the many BOINC projects.

I to have been a long time user of seti classic. I run it on 3 different Macs, at least until Dec. 15 when I will no longer be able to because I cannot afford to upgrade to Mac OS X 10.3!!! Do you really think that they will ever go back to using classic? We can only hope.

---

SETI Classic is dead and about to be buried, so no, that won't happen.

Support for anything less than Mac OS X 10.3.9 is unlikely to happen but vaguely possible. However, one of the BOINC projects is currently considering dropping support for even 10.3.9, and going "10.4 only". Just too many advantages, speed and stability, to the newer OSes, and too little return for supporting the older ones.

I sympathize with everyone here - I'm trying to switch over to boinc, but not enjoying all the pain. Imagine how I feel: running IRIX, Windows and Macintosh clients, over 18781 classic work units, and watching boinc produce strange error messages (are these documented anywhere?), abort and back off downloads, and am yet to see any real results in terms of completing a single work unit. *sigh*

I will keep fighting, but my patience has limits!

-Ocelot Wreak.

---

watching boinc produce strange error messages (are these documented anywhere?), abort and back off downloads, and am yet to see any real results in terms of completing a single work unit. *sigh*

I will keep fighting, but my patience has limits!

Yes, almost all the error messages are well documented in the Wiki. (Home page, "BOINC Wiki".) And there are a lot of us here to help with whatever problems you are having. If you're fighting on your own, without asking questions (and this is your first post...), it's hard - please, feel free to ask!

I see you have 7 results that had an error downloading, then one that is now "in progress". Did you get whatever it was solved?

Walter, if you search IRIX binaires, the are at the third party site (http://boinc.berkeley.edu/download_other.php) avaiable.

---

Hi All,
Thanks for your encouragement - I'm not sure it is really a boinc problem in itself, as the climatepredictionwhatever work units download and run first shot - no MD5 checksum errors, Error 505, backoff and wait error, etc. like happens with every SETI work unit. I think it is a seti server problem.

I see the 3rd party IRIX client is available, and am about to try that next on my 10 CPU Power Challenge. I hope it goes better than the Mac SETI work units. Still no luck (from what I can see) getting a SETI work unit to run under boinc.

I will post more when I have deciphered more of what the h*ll is going on. And I have to say that using Wiki for storing your error code documentation is sub-optimal!

-wjc.

Bill Michael, SETI@home user since only 2003, says he's "continually surprised" and "sorry to see you go" (which oddly sounds like "don't let the door hit you on the way out", but I'm probably projecting). Why the surprise? Why blame the users who are baffled by BOINC? BOINC's interface IS nonintuitive. (Yes, I'm sure we're grateful there's an interface at all.) The docs accompanying the download ARE poor. Why hasn't BOINC (and by association, SETI@home) BOTHERED to fix these problems?

Not every SETI@home participant is a bithead. They may have no idea what's meant by "wiki". If BOINC doesn't work for the Average Jo/e, then it doesn't serve its purpose with SETI@home. Funding pinch? Is that participants' fault, too? Danged ungrateful users.

Learn the reason behind the continual surprise. Quit blaming participants.

Of course, if you're plenty happy having a nifty little internal-use tool for the computer lab, ignore this post. Otherwise, holler when BOINC is ready for use by every participant. Cheers, best wishes, and so long.

Bill Michael, SETI@home user since only 2003

Um, sorry - been around a lot longer than that, but lost my first account, didn't bother trying to merge it, just got another... only came to BOINC this summer...

says he's "continually surprised" and "sorry to see you go" (which oddly sounds like "don't let the door hit you on the way out", but I'm probably projecting).

Not a lot of projecting. There are some people who I _have_ flat said "don't let the door hit you" to. If someone is going to do nothing but whine and complain, and be miserable here, why would they stay? All they accomplish, besides making themselves _more_ miserable, is inflicting their misery on others.

Why the surprise? Why blame the users who are baffled by BOINC? BOINC's interface IS nonintuitive. (Yes, I'm sure we're grateful there's an interface at all.) The docs accompanying the download ARE poor. Why hasn't BOINC (and by association, SETI@home) BOTHERED to fix these problems?

BOINC isn't that hard to learn. Some 500,000 people have managed it. I am no happier about UCB's pathetic documentation than you are, because I'm sitting here trying to help people that wouldn't have a problem if it was documented correctly! The first time I tried BOINC, it was CLI-only. I gave up and went back to Classic. It's a _lot_ better now, and is continuing to improve, although slower than most of us would like.

Not every SETI@home participant is a bithead. They may have no idea what's meant by "wiki". If BOINC doesn't work for the Average Jo/e, then it doesn't serve its purpose with SETI@home. Funding pinch? Is that participants' fault, too? Danged ungrateful users.

When I referenced the Wiki here, I provided a link to it. It's also linked off the SETI home page, and in thousands of postings on the boards - it shouldn't take very much effort to find it, and once you are looking at it, it's pretty obvious what it is...

I have been one of those saying that we need a "BOINC lite", pre-loaded with SETI, enter email address and run. I got nowhere, as there weren't the resources to do it.

Learn the reason behind the continual surprise. Quit blaming participants.

The UCB-provided (lack of) documentation sucks. Been there, done that, had to actually click the mouse a few times and read a few pages, and figure it out for myself. I can't change UCB policies, or force them to communicate better. I'm not personally real happy with the SETI project at the moment, but I, unlike some of the participants who complain so much, know the difference between BOINC and SETI. BOINC works. SETI has problems. SETI under BOINC has _fewer_ problems, and more potential for growth, than SETI Classic.

Of course, if you're plenty happy having a nifty little internal-use tool for the computer lab, ignore this post. Otherwise, holler when BOINC is ready for use by every participant. Cheers, best wishes, and so long.

Half a million geeks are apparently running this in their little computer labs... Please go to the Number Crunching forum and read the posting near the top called Why classic SETI@home is closing down and other facts of life. - you don't even have to try to find it, I made it a link.

If someone is going to do nothing but whine and complain....

If that's all that's being _heard_, PLEASE try again. SETI@home participants are happy, glad, and even proud to help with the search. The requests are to _continue_ supporting OS 9 systems.

BOINC isn't that hard to learn....

Didn't say I have a problem using BOINC.

When I referenced the Wiki here, I provided a link to it. It's also linked off the SETI home page, and in thousands of postings on the boards - it shouldn't take very much effort to find it, and once you are looking at it, it's pretty obvious what it is...

How many never get that far? Can the link be included in the app, in a read-me file?

Half a million geeks are apparently running this in their little computer labs...

Down from how many participants in Classic-only days? How big is the loss?

If someone is going to do nothing but whine and complain....

If that's all that's being _heard_, PLEASE try again. SETI@home participants are happy, glad, and even proud to help with the search. The requests are to _continue_ supporting OS 9 systems.

You're taking my "general" statement, from montoring hundreds or thousands of postings, and applying it to this specific case... spend some time reading NC, you'll see the whining and complaining I'm talking about.

As for supporting OS 9 - I WOULD BE HAPPY TO! I would _love_ it! I _HAVE_ OS 9 systems, an iMac _and_ an original iBook, and I could get more credits! Somebody come up with $200K and I'll be part of the team to _do_ it! ... Unfortunately, they're not going to pay me, and reality says that I can't donate a year of my time, without pay, to write the app, and more time to support it, and one person probably couldn't do it alone anyway...

Think about it - $200,000 - to support an OS that is run by about 1/10 of the 4% of the participants who have Macs... It's just not worth it. I really, truly wish it was, but the cold hard cash isn't there, and even if it was, it could be better spent on other things.

BOINC isn't that hard to learn....

Didn't say I have a problem using BOINC.

Didn't say _you_ did. But some just _won't_ (I will not say "can't") learn how, and instead, do nothing but demand that it change, instantly, to do things the way they would like it to. Some are new Classic converts, who (quite reasonably) would like it to be simpler - but some are people who have thousands of credits, and _still_ do nothing but complain about things.

When I referenced the Wiki here, I provided a link to it. It's also linked off the SETI home page, and in thousands of postings on the boards - it shouldn't take very much effort to find it, and once you are looking at it, it's pretty obvious what it is...

How many never get that far? Can the link be included in the app, in a read-me file?

The Wiki is a voluntary set of documentation, mostly put together by one person. Dr. Anderson refuses to "legitimize" it, unless he can control the content, and Paul refuses to allow anyone to dictate what goes into it. It was a struggle to even get it linked on the home page, and that didn't happen until every other project had already done so.

Half a million geeks are apparently running this in their little computer labs...

Down from how many participants in Classic-only days? How big is the loss?

That's up from 300,000 "active" in Classic. There were well over a million who downloaded the Classic app at some point and ran it, or at least there were that many email addresses, but the most recent figure I've heard is that only 300,000 people actually returned at least 1 result in November. I don't know what the "active" figure was at the peak. And all that is based on my memory of other people's research into the numbers, so for exact figures, you'd have to check the stats sites such as Boinc Synergy, and try to figure it out.

You're taking my "general" statement, from montoring hundreds or thousands of postings, and applying it to this specific case... spend some time reading NC, you'll see the whining and complaining I'm talking about.

What I was saying was that all the chuff you're reading elsewhere is bleeding over onto this topic. The challenge with coding an app for public consumption is making it work for the public and viewing things from an unfamiliar (perhaps downright alien) perspective. An avalanche of unconstructive criticism makes getting real concerns heard all that much harder, no argument there.

Thanks for taking a stab at answering my questions. I wish those in charge would do as much. Mac OS 9 folks just want to keep on helping the project, and we're being told to get lost.