When is classic going to be shut down? Does centralCommand have a plan? It seems setiBoinc is now demonstrably more efficient than setiClassic (e.g. the same size work units are being validated on the fly, not accumulating in some digital heap somewhere). Classic hit the silly 2B mark. So what's the next step? Let's breath some life into this project.

---

May this Farce be with You

I can't believe you're wanting to shut down palindrome central. :)

I can't believe you're wanting to shut down palindrome central. :)

I can't believe he wants to shut down boinc so soon.

---

I can't believe centralCommand doesn't post a strategic project plan.

I can't believe how boring the classic message boards are; they are each enchanted with the combination of digits they are getting as 'scores' and want to live in the past rather than move to boincSeti.

I can't believe that despite the amount of computer effort the community has provided and the manpower/hardware expended on this project, there is very little, very rare accountability. Most scientists like to advertise their science (science here means ET/astroscience and not distributed computing, which is merely a means to an end.)

I guess I'm just an aetheist because I care too much to be an agnostic.

---

May this Farce be with You

They DO have a plan. It goes something like this:

1) Make sure seti/BOINC can handle a sudden influx of 100k+ users
2) Send an email to all classic users informing them of BOINC
3) One month later, shut down classic

#1 is what has proven more difficult than expected. Every time they fix a bottleneck, another one crops up. It does appear that they are running pretty smoothly right now however if you read the technical news you will see that the current setup is a small, temporary band-aid. If 100k users suddenly signed up and started doing work units, the upload file server could fill up and/or fail. This would be a bad thing. They need another storage solution before classic goes anywhere. Maybe with the donations that came in this month...

---

A member of The Knights Who Say NI!
For rankings, history graphs and more, check out:
My BOINC stats site

They DO have a plan. It goes something like this:

1) Make sure seti/BOINC can handle a sudden influx of 100k+ users
2) Send an email to all classic users informing them of BOINC
3) One month later, shut down classic

#1 is what has proven more difficult than expected. Every time they fix a bottleneck, another one crops up. It does appear that they are running pretty smoothly right now however if you read the technical news you will see that the current setup is a small, temporary band-aid. If 100k users suddenly signed up and started doing work units, the upload file server could fill up and/or fail. This would be a bad thing. They need another storage solution before classic goes anywhere. Maybe with the donations that came in this month...

Add as #0 - fix a few BOINC useability issues. This work is currently in progress.

---

BOINC WIKI

I can't believe how well Seti is running at the moment.

---

IAS - Where Space Is Golden!

[quote]They DO have a plan. It goes something like this:

1) Make sure seti/BOINC can handle a sudden influx of 100k+ users
2) Send an email to all classic users informing them of BOINC
3) One month later, shut down classic

#1 is what has proven more difficult than expected. Every time they fix a bottleneck, another one crops up. It does appear that they are running pretty smoothly right now however if you read the technical news you will see that the current setup is a small, temporary band-aid. If 100k users suddenly signed up and started doing work units, the upload file server could fill up and/or fail.

I think what they are waiting for is for that More Sensitive version of SETI BOINC to make it out of BETA. Those Work Units are I hear 10 Times the Size and take about 10 times as Long to Crunch. Even with the Influx of 100,000 more Crunchers plus the 200,000 we already have (300,000 Total) we would still only be Uploading the amount of Units that 30,000 Crunchers used to do. If the New SETI BONC BETA becomes the Standard we are going to need 2 Million Crunchers to produce the same Work Units Completed that 200,000 used to do.

---

They DO have a plan. It goes something like this:

1) Make sure seti/BOINC can handle a sudden influx of 100k+ users
2) Send an email to all classic users informing them of BOINC
3) One month later, shut down classic

#1 is what has proven more difficult than expected. Every time they fix a bottleneck, another one crops up. It does appear that they are running pretty smoothly right now however if you read the technical news you will see that the current setup is a small, temporary band-aid. If 100k users suddenly signed up and started doing work units, the upload file server could fill up and/or fail. This would be a bad thing. They need another storage solution before classic goes anywhere. Maybe with the donations that came in this month...

What you are saying is that they don't have a plan. I can't believe a bunch of adults are rubbing their worry beads waiting for some solution to the disk problem (it it is a problem) to drop on their heads.

Remember, every system has a bottleneck. What is needed to make such a statement sensible is an engineering statement of what capacity the current (or planned?) system should be/has.

---

May this Farce be with You

I think what they are waiting for is for that More Sensitive version of SETI BOINC to make it out of BETA. Those Work Units are I hear 10 Times the Size and take about 10 times as Long to Crunch. Even with the Influx of 100,000 more Crunchers plus the 200,000 we already have (300,000 Total) we would still only be Uploading the amount of Units that 30,000 Crunchers used to do. If the New SETI BONC BETA becomes the Standard we are going to need 2 Million Crunchers to produce the same Work Units Completed that 200,000 used to do.

How about keeping this version of boincSeti running without significant modifications for a year or so.

Recycle the classic hardware into running the next version of seti, call it seti3, possibly as a separate boinc project. A cleaver person might enable seti2 and seti3 to share the same science database. This way the 'standard' size wu's will continue to run as seti2 (I'm assuming they want to continue running the lower resolution wu's as screens) and the seti2 system will stablize, absorb new users, etc.

Then Seti3 transitions off the alpha hardware and can feel the growing pains as the number of users who can go to longer work units migrate over there, just like people jumped ship off seti1 to move to seti2.

Finally, the real gonzo's can work on seti4, whatever that is going to be, using the alpha hardware. Maybe a seti4 client might be tuned to work primarily on 64 bit multi-processors using instrument limited telescope resolution.

---

May this Farce be with You

I think what they are waiting for is for that More Sensitive version of SETI BOINC to make it out of BETA. Those Work Units are I hear 10 Times the Size and take about 10 times as Long to Crunch. Even with the Influx of 100,000 more Crunchers plus the 200,000 we already have (300,000 Total) we would still only be Uploading the amount of Units that 30,000 Crunchers used to do. If the New SETI BONC BETA becomes the Standard we are going to need 2 Million Crunchers to produce the same Work Units Completed that 200,000 used to do.

How about keeping this version of boincSeti running without significant modifications for a year or so.

Recycle the classic hardware into running the next version of seti, call it seti3, possibly as a separate boinc project. A cleaver person might enable seti2 and seti3 to share the same science database. This way the 'standard' size wu's will continue to run as seti2 (I'm assuming they want to continue running the lower resolution wu's as screens) and the seti2 system will stablize, absorb new users, etc.

Then Seti3 transitions off the alpha hardware and can feel the growing pains as the number of users who can go to longer work units migrate over there, just like people jumped ship off seti1 to move to seti2.

Finally, the real gonzo's can work on seti4, whatever that is going to be, using the alpha hardware. Maybe a seti4 client might be tuned to work primarily on 64 bit multi-processors using instrument limited telescope resolution.

I don't believe that they are going to keep running the current software, but are going to switch over to the new one.

---

BOINC WIKI

in a response to the title:

The age old argument...

I say......Seat lid UP. It's easier to lower it than to lift it.

I can't believe centralCommand doesn't post a strategic project plan.

Hmm... didn't Technical News for 04.09.2005 mention something about not shutting down "classic" before:

1; BOINC-client v5 is released.
2; Seti_Enhanced is released.

As for timeline, for both of these it depends on how many more bugs pops up, and how long it takes to fix them...

in a response to the title:

The age old argument...

I say......Seat lid UP. It's easier to lower it than to lift it.

If ya learned to aim better rather than spraying around willy-nilly, it wouldn't matter if the seat was up or down ...

---

I can't believe centralCommand doesn't post a strategic project plan.

Hmm... didn't Technical News for 04.09.2005 mention something about not shutting down "classic" before:

1; BOINC-client v5 is released.
2; Seti_Enhanced is released.

As for timeline, for both of these it depends on how many more bugs pops up, and how long it takes to fix them...

Maybe; that comment was made in the heat of the battle and every battle plan changes. Plus there were so many contingiencies referred to.

So why not make a clear statement of key milestones and expected time line.

My other point is don't fix it if it ain't broke. It seems that the current boinc is producing results, which I will assume are useful but I have no evidence of that, and so throwing another rock in the water should be met with some trepidation. For example, how many clients will be rendered useless by Seti_Enhanced? If this would be a large number, is there a work around.

I suppose I could continue, but perhaps what I would like (among many things) is a link to a page that describes the project plan and current status, updated monthly. The news page is fun because it is almost recent and casual. Matt's slog is folksy and lay back. CC needs to balance it out with a proper progress page.

Finally, @archer and @mmciastro, what is the answer if one were a eunich?

---

May this Farce be with You

I can't believe how well Seti is running at the moment.

"That is why you will fail, young Skywalker...." >:-(

I BELIEVE!!!! >:-) (-:<

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

Maybe; that comment was made in the heat of the battle and every battle plan changes. Plus there were so many contingiencies referred to.

So why not make a clear statement of key milestones and expected time line.

But as you just said yourself, "every battle plan changes"...

As for Seti_Enhanced, Matt Lebofsky mentioned this even before all the resent problems with data-server:
We are going to be reviving the SETI@home beta site soon (if it isn't up already). There's some last minute science changes that should be tested before we final throw the switch and start shutting down SETI@home classic.

My other point is don't fix it if it ain't broke. It seems that the current boinc is producing results, which I will assume are useful but I have no evidence of that, and so throwing another rock in the water should be met with some trepidation. For example, how many clients will be rendered useless by Seti_Enhanced? If this would be a large number, is there a work around.

"Classic" + SETI@Home/BOINC gets around 1.5M results/day, but current seti-Arecibo-recorder can max record 271k wu/day meaning 1.084M results/day. Due to RFI/outages the actual is less, meaning if all "classic"-users migrates 30%-50% of the users will not get any seti-work...
Well, there is currently a 5-month-backlog of recorded tapes, but this will not last long...

When Seti_Enhanced is released on the other hand, they're expecting on average 75% reduction, this gives under 400k results/day and is less than SETI@Home/BOINC is currently getting.

Seti_Enhanced has the same disk/memory-requirements, and deadline is set so a 100 Mflops-computer running 10% of the time is fast enough. If users wants to keep their slowest computers crunching for 1+ months on same result is another matter. Oh, and wu with very-high angle-range is nearly as fast as with current seti-application.

Interesting.

One comment: if the classic users would switch and find they cannot get enough work to process, I hope they would join one or more of the other projects. Especially if they have the slower machines. In that sense, the longer 'we' keep them on classic, the more the other boinc projects suffer.

---

May this Farce be with You

At this point we're basically just waiting on the latest release of BOINC, which includes newer user-friendly features (like not having to deal with an account key). This should make transitioning over from Classic even easier.

We may also wait for the SETI Enhanced client, but this isn't really a gating item (though it would ensure that our hardware could handle the influx of 300K+ new users, and that we would have plenty of work to send out to the new users so they aren't left hanging out to dry).

The last internal discussion we had on this matter looked something like: new client released by mid-october. Immediately following would be a mass e-mail saying we are shutting down Classic in a month, here's how to install BOINC, etc. Then around mid-november, the switch is thrown.

- Matt

---

-- BOINC/SETI@home network/web/science/development person
-- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude

At this point we're basically just waiting on the latest release of BOINC, which includes newer user-friendly features (like not having to deal with an account key). This should make transitioning over from Classic even easier.

We may also wait for the SETI Enhanced client, but this isn't really a gating item (though it would ensure that our hardware could handle the influx of 300K+ new users, and that we would have plenty of work to send out to the new users so they aren't left hanging out to dry).

The last internal discussion we had on this matter looked something like: new client released by mid-october. Immediately following would be a mass e-mail saying we are shutting down Classic in a month, here's how to install BOINC, etc. Then around mid-november, the switch is thrown.

- Matt

Thanks for the info! :-)

---