Apart from Killer Whales what on earth is ORCA?

Windows instructions tell us to edit the installer with ORCA.

How on earth is anyone supposed to configure and manage the systems with such an abysmal standard of documentation?

Has anybody heard of 'user-friendly'?

Come on if you want 400,000 people to move to BOINC you need to seriously and rapidly address all the concerns that keep getting posted and seemingly totally ignored. The first essential is to supply along with the executables a description and set of configuration instructions, secondly if any additional tools are needed then they should either be with the package or readily available (through the documentation). Expecting people to trawl through Google or whatever to find arcane relevant texts is not acceptable.

---

I've updated the documentation so it specifically state Microsoft Orca, which is a tool that is included with the Windows Installer SDK.

More information can be found on the MSDN site:
http://msdn.microsoft.com/

http://support.microsoft.com/default.aspx?scid=kb;en-us;255905

Warning though, Ocra isn't really all that user friendly, you can use any MSI editor though.

---

----- Rom
BOINC Development Team, U.C. Berkeley
My Blog

Okay then.

So why should we have to get this sdk? What does it do? How does it work? What does it cost? Why do we have to go through this unknown process to make your package work?

Why not just provide proper documentation, the proper tools then we can use them. We are not all developers.

What is an MSI editor and the same questions apply as above? How do we use it? What parameters are needed?

---

> Apart from Killer Whales what on earth is ORCA?
>
> Windows instructions tell us to edit the installer with ORCA.
>
> How on earth is anyone supposed to configure and manage the systems with such
> an abysmal standard of documentation?
>
> Has anybody heard of 'user-friendly'?

I suspect that a description of the underlying problem (instead of "what on earth is ORCA") might have gotten a different answer from Rom.

---

We haven't seen a fan of the code side of BOINC decide to take a stab at documenting it yet ...

Though we are seeing a nice collection of GUIs being made that seem to have all of the essential features we would like to have .... :)

So, maybe some one will step up and start to write a good delveloper site that does a better job of documenting how to do BOINC from the developer's perspective.

---

> So why should we have to get this sdk? What does it do? How does it work?
> What does it cost? Why do we have to go through this unknown process to make
> your package work?

Unless you are trying to mass deploy BOINC using Active Directory, SMS, or some other management system, you really don't have to worry about this stuff. SDK's from Microsoft are generally Free.

As the bottom of the http://boinc.berkeley.edu/win_install.php states, it is only needed if you wish to change the default behavior of the Windows Installer while deploying across a lot of machines.

For the average person, just executing the executable will work just fine.

> Why not just provide proper documentation, the proper tools then we can use
> them. We are not all developers.
>
> What is an MSI editor and the same questions apply as above? How do we use
> it? What parameters are needed?

Do you really want to customize the installer?

---

----- Rom
BOINC Development Team, U.C. Berkeley
My Blog

> > What is an MSI editor and the same questions apply as above? How do we
> use
> > it? What parameters are needed?
>
> Do you really want to customize the installer?

Nick said something on another thread about having a disk array that was very large, and the installer complained that he didn't have enough free disk space.

The value for free space was negative, so it looks like a large integer that wrapped.

I think that's the origin of this thread. I could be wrong.

---

I'm looking into that bug, from the inital looks of it, it looks like it is an MSI bug, and I'm not sure if I can work around it at present.

Probably the easiest hack I can think of to work around the issue is to create a file that is a little bit larger than the absolute value that the Microsoft Installer is complaining about.

It is a hack, but is should at least work around the issue.

---

----- Rom
BOINC Development Team, U.C. Berkeley
My Blog

On a slightly different note...

ORCA??? ORCA???? That is the name of a whale (ok a killer one, large relative of the Dolphin/porpoise).

ORCA is a whale. ORCA is big, and produces BIG babies. No wonder the MSI is so big =PPP

Why is it called a KILLER whale in english? (Yeah, I know, they like to kill other whales, but I think we could've come up with a slightly better name anyway.)

The "killer" whale from the Free Willy movies swimmed to Norway and died here. It was a sad end to a sad life, but at least he had a few happy months in our waters - never showing any of his killer instincts to all those people patting him on the nose. He liked being with people - and vice versa. Oh, well... :)

---

> > So why should we have to get this sdk? What does it do? How does it
> work?
> > What does it cost? Why do we have to go through this unknown process to
> make
> > your package work?
>
> Unless you are trying to mass deploy BOINC using Active Directory, SMS, or
> some other management system, you really don't have to worry about this stuff.
> SDK's from Microsoft are generally Free.
>
> As the bottom of the http://boinc.berkeley.edu/win_install.php states, it is
> only needed if you wish to change the default behavior of the Windows
> Installer while deploying across a lot of machines.
>
> For the average person, just executing the executable will work just fine.
>
> > Why not just provide proper documentation, the proper tools then we can
> use
> > them. We are not all developers.
> >
> > What is an MSI editor and the same questions apply as above? How do we
> use
> > it? What parameters are needed?
>
> Do you really want to customize the installer?
>
>

Not particularly but I supposed it may be able to set parameters or things in the process. Since there is nothing local from the gui I had assumed there was a preference setting capabilty somewhere other than on the web. Some of the architecture design decisions do not appear to be particularly wise. I do not like the lack of control over what BOINC does whish is why it has been tried several times and rejected. If someone in the team bothered to read why people don't like BOINC (far more of them than its fans) then perhaps they would rethink how it is packaged. I have seen nothing that is going to persuade me to use it after classic is switched off, along with hundreds of thousands of others. Getting information from these BBs is like a slow drip feed. Little bits come through after days of waiting, and STILL the project team do not provide proper documentation. BOINC might be fine for hackers who have nothing else to do but experiment and play with computer systems but some of us merely want something that can be switched on left alone, but be managed as required at a local level.

After two days of wondering I now think that the question relating to ORCA has been answered and it would appear that it won't do what it vaguely might. However if the installation parameters are adjustable then perhaps some documentation on what is adjustable, why and how would also have helped. I am not aware of development kits from Microsoft that were ever free, never mind the learning curve.

The lack of marketing is creating a massive problem. After being told last year that I would need to compile it myself to regain classic control functionality, it seems that a command line version is available. But, having downloaded it last week with no documentation or settings information I am still in the dark.

Back to classic yet again for the remainder of its life until perhaps the project team start addressing the ommissions.

Thanks for answers received to date.

---

> > So why should we have to get this sdk? What does it do? How does it
> work?
> > What does it cost? Why do we have to go through this unknown process to
> make
> > your package work?
>
> Unless you are trying to mass deploy BOINC using Active Directory, SMS, or
> some other management system, you really don't have to worry about this stuff.
> SDK's from Microsoft are generally Free.
>
> As the bottom of the http://boinc.berkeley.edu/win_install.php states, it is
> only needed if you wish to change the default behavior of the Windows
> Installer while deploying across a lot of machines.
>
> For the average person, just executing the executable will work just fine.
>
> > Why not just provide proper documentation, the proper tools then we can
> use
> > them. We are not all developers.
> >
> > What is an MSI editor and the same questions apply as above? How do we
> use
> > it? What parameters are needed?
>
> Do you really want to customize the installer?
>
>

Not particularly but I supposed it may be able to set parameters or things in the process. Since there is nothing local from the gui I had assumed there was a preference setting capabilty somewhere other than on the web. Some of the architecture design decisions do not appear to be particularly wise. I do not like the lack of control over what BOINC does whish is why it has been tried several times and rejected. If someone in the team bothered to read why people don't like BOINC (far more of them than its fans) then perhaps they would rethink how it is packaged. I have seen nothing that is going to persuade me to use it after classic is switched off, along with hundreds of thousands of others. Getting information from these BBs is like a slow drip feed. Little bits come through after days of waiting, and STILL the project team do not provide proper documentation. BOINC might be fine for hackers who have nothing else to do but experiment and play with computer systems but some of us merely want something that can be switched on left alone, but be managed as required at a local level.

After two days of wondering I now think that the question relating to ORCA has been answered and it would appear that it won't do what it vaguely might. However if the installation parameters are adjustable then perhaps some documentation on what is adjustable, why and how would also have helped. I am not aware of development kits from Microsoft that were ever free, never mind the learning curve.

The lack of marketing is creating a massive problem. After being told last year that I would need to compile it myself to regain classic control functionality, it seems that a command line version is available. But, having downloaded it last week with no documentation or settings information I am still in the dark.

Back to classic yet again for the remainder of its life until perhaps the project team start addressing the ommissions.

Thanks for answers received to date, which have been from client participants and not the team!

---

not that this will accomplish anything, it appears some people like to gripe.

> I do
> not like the lack of control over what BOINC does whish is why it has been
> tried several times and rejected.
what kind of control are you looking for, it does FAR more than classic was ever able to do?

> If someone in the team bothered to read why
> people don't like BOINC (far more of them than its fans) then perhaps they
> would rethink how it is packaged.
they thought long and hard about it, this is what was decided.
They have explained this decision numerous times, and it appears that everyone here agrees with it...

> Getting information from these BBs is like a slow drip
> feed. Little bits come through after days of waiting, and STILL the project
> team do not provide proper documentation.

it is not their job to provide documentation. Paul Buck's site has pretty much been endorsed by all BOINC projects though.
BOINC is OPEN SOURCE, if you have questions, you get to dig through code. That is not my realm either, but everything appeared extremely self evident to me.
The other questions get answered here, or at whatever site the question is concerning.

> BOINC might be fine for hackers who
> have nothing else to do but experiment and play with computer systems but some
> of us merely want something that can be switched on left alone, but be managed
> as required at a local level.

which is exactly what you have.

>
> After two days of wondering I now think that the question relating to ORCA has
> been answered and it would appear that it won't do what it vaguely might.
> However if the installation parameters are adjustable then perhaps some
> documentation on what is adjustable, why and how would also have helped. I am
> not aware of development kits from Microsoft that were ever free, never mind
> the learning curve.
there are some, but you need to get on their developers list. and yes, nothing is easy.
but again, that information was provided for the DEVELOPERS that use BOINC, not the typical user like me. again, it is a part of being OPEN SOURCE. you can do the same thing with all versions of windows and have it configured exactly how you want and with all the updates and SPs. but i have no clue how you would do so, same with BOINC.

>
> The lack of marketing is creating a massive problem. After being told last
> year that I would need to compile it myself to regain classic control
> functionality, it seems that a command line version is available. But, having
> downloaded it last week with no documentation or settings information I am
> still in the dark.

i seriously doubt that you were actually told you would HAVE to compile it.
since i started boinc with CC 4.09, the typical install came with both boinc_gui AND boinc_cli, or did you never bother to look at the install directory?
there are command-line switches, just like classic.
now with 4.25, this is different. the name appears to have changed to boinc.exe. bypass the boincmgr,exe if you want...
>
> Back to classic yet again for the remainder of its life until perhaps the
> project team start addressing the ommissions.

again, there are NO ommissions...*sigh*
>
> Thanks for answers received to date, which have been from client participants
> and not the team!
>
sorry that we were unable to explain things to a manner that you could understand. if we lose ALL the classic users as you suggest (extremely doubtful) those here already seem to be doing a good job of keeping up with the WU queue. and we will ALWAYS be getting new users and hosts. so those unwilling to change (and if you are, why are you even with SETI to begin with? if a signal is found, there will be such a profound societal change) won't be that big a loss.
those that do switch i think will find an enjoyable experience here.

Micah

---

S@h Berkeley's Staff Friends Club

> BOINC might be fine for hackers who
> have nothing else to do but experiment and play with computer systems but some
> of us merely want something that can be switched on left alone, but be managed
> as required at a local level.

The part that I find most amusing:

Nick is complaining that BOINC can't be managed in (presumably exactly) the same way that he managed classic.

Yet, here we're reading that "some of us merely want something that can be switched on left alone."

... and if you load BOINC, sign up for a project, and leave it alone it works very, very well.

The vast majority of complaints I've seen have nothing to do with leaving it alone and everything to do with managing it fairly aggressively or monitoring it extremely closely.

---

> not that this will accomplish anything, it appears some people like to gripe.
>
> > I do
> > not like the lack of control over what BOINC does whish is why it has
> been
> > tried several times and rejected.
> what kind of control are you looking for, it does FAR more than classic was
> ever able to do?
>
> > If someone in the team bothered to read why
> > people don't like BOINC (far more of them than its fans) then perhaps
> they
> > would rethink how it is packaged.
> they thought long and hard about it, this is what was decided.
> They have explained this decision numerous times, and it appears that
> everyone here agrees with it...
>
> > Getting information from these BBs is like a slow drip
> > feed. Little bits come through after days of waiting, and STILL the
> project
> > team do not provide proper documentation.
>
> it is not their job to provide documentation. Paul Buck's site has pretty much
> been endorsed by all BOINC projects though.
> BOINC is OPEN SOURCE, if you have questions, you get to dig through code. That
> is not my realm either, but everything appeared extremely self evident to me.
> The other questions get answered here, or at whatever site the question is
> concerning.
>
> > BOINC might be fine for hackers who
> > have nothing else to do but experiment and play with computer systems but
> some
> > of us merely want something that can be switched on left alone, but be
> managed
> > as required at a local level.
>
> which is exactly what you have.
>
> >
> > After two days of wondering I now think that the question relating to
> ORCA has
> > been answered and it would appear that it won't do what it vaguely might.
>
> > However if the installation parameters are adjustable then perhaps some
> > documentation on what is adjustable, why and how would also have helped.
> I am
> > not aware of development kits from Microsoft that were ever free, never
> mind
> > the learning curve.
> there are some, but you need to get on their developers list. and yes, nothing
> is easy.
> but again, that information was provided for the DEVELOPERS that use BOINC,
> not the typical user like me. again, it is a part of being OPEN SOURCE. you
> can do the same thing with all versions of windows and have it configured
> exactly how you want and with all the updates and SPs. but i have no clue how
> you would do so, same with BOINC.
>
> >
> > The lack of marketing is creating a massive problem. After being told
> last
> > year that I would need to compile it myself to regain classic control
> > functionality, it seems that a command line version is available. But,
> having
> > downloaded it last week with no documentation or settings information I
> am
> > still in the dark.
>
> i seriously doubt that you were actually told you would HAVE to compile it.
> since i started boinc with CC 4.09, the typical install came with both
> boinc_gui AND boinc_cli, or did you never bother to look at the install
> directory?
> there are command-line switches, just like classic.
> now with 4.25, this is different. the name appears to have changed to
> boinc.exe. bypass the boincmgr,exe if you want...
> >
> > Back to classic yet again for the remainder of its life until perhaps
> the
> > project team start addressing the ommissions.
>
> again, there are NO ommissions...*sigh*
> >
> > Thanks for answers received to date, which have been from client
> participants
> > and not the team!
> >
> sorry that we were unable to explain things to a manner that you could
> understand. if we lose ALL the classic users as you suggest (extremely
> doubtful) those here already seem to be doing a good job of keeping up with
> the WU queue. and we will ALWAYS be getting new users and hosts. so those
> unwilling to change (and if you are, why are you even with SETI to begin with?
> if a signal is found, there will be such a profound societal change) won't be
> that big a loss.
> those that do switch i think will find an enjoyable experience here.
>
> Micah
>

The add-ons to classic make it extremely effective and easy to operate especially as they don't actually interfere with the core functionality. yes classic on its own was pretty dire, but the very sophisticated suite of add-ons make this argument pointless. Especially as BOINC will not work with less than 4gb of disk space. As far as documentation goes there isn't any. If they want me to run something for them then I would expect a something more forthcoming thna to be expected to find some strange un-connected and unrelated web site elsewhere. If it is that good and fills a gap why isn't it part of SETI? Look at the classic site. Lots of links to add-ons, further documentation and information, FAQs and so on.

I have never said I am unwilling to change but if the darned thing won't work then it won't work, if the team cannot be bothered to provide documentaiton adn a proper method of carrying over comparable 'credits' then something is amiss isn't there? Or is there a suggestion that 80,000 BOINC users are doing more SETI than 400,0000 live classic users?

I run several copies of the classic CLI, not the gui. When I looked at BOINC last year there didn't appear to be anything other than the video version (or ot was abysmally explained). When asking about alternative versions (remember the classic CLI is not the same as the screensaver version) it explicitly said that people can compile their own. When questioned about this no answer was forthcoming. So perhaps if the team had looked at what was in use last year and bothered to explain how the functionality could be replicated with the new version and provided proper documentation on how to set it up and stopped remote control over client disk space then all might have been different. As it is anybody reading the concerns should immediately see the marketing failures.

I think the point is being missed in all the peripheral detail, which is a bit of 'while I am complaining there are these things that need attention as well', is that the full and proper documentation, settings, configurations, locations, etc MUST be publicised, instead of trying to filter through thousands of message threads and strange headed queris and so on. Maybe the information is out there but if we can't find it or it is obscure how are we to be able to spot an alien electromagnetic signal?

---

> The add-ons to classic make it extremely effective and easy to operate
> especially as they don't actually interfere with the core functionality. yes
> classic on its own was pretty dire, but the very sophisticated suite of
> add-ons make this argument pointless.

Classic didn't have any add-ons at first either. There are several add-ons for BOINC already and I'm sure a lot more are being worked on.

> Especially as BOINC will not work with less than 4gb of disk space.

huh? BOINC is running fine on 2.5 GB of space here and 4.3 on my laptop. You must have your disk space settings wrong.

> As far as documentation goes there isn't any.
> If they want me to run something for them then I would expect a something more
> forthcoming thna to be expected to find some strange un-connected and
> unrelated web site elsewhere. If it is that good and fills a gap why isn't it
> part of SETI? Look at the classic site. Lots of links to add-ons, further
> documentation and information, FAQs and so on.

Ever heard of google? Search for "boinc documentation". It will take you to Paul's web site which has massive amounts of documentation on BOINC itself as well as on all the projects using it. The fact of the matter is that programmers often don't make for good documentation. Since they know the system from the inside out, they will assume that many features are "obvious" or common sense. Therefore it is better to let someone else who isn't as familiar take care of that. That is the point of an open-source project. The community helps out where it can. Some write good code, others can document it so that the rest of the world can understand.

> I run several copies of the classic CLI, not the gui. When I looked at BOINC
> last year there didn't appear to be anything other than the video version (or
> ot was abysmally explained). When asking about alternative versions (remember
> the classic CLI is not the same as the screensaver version) it explicitly said
> that people can compile their own. When questioned about this no answer was
> forthcoming. So perhaps if the team had looked at what was in use last year
> and bothered to explain how the functionality could be replicated with the new
> version and provided proper documentation on how to set it up and stopped
> remote control over client disk space then all might have been different. As
> it is anybody reading the concerns should immediately see the marketing
> failures.

BOINC has matured a lot since last year. It is working well enough now for 80,000 users to figure out how to run it on almost 200,000 computers.

---

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

>
> The add-ons to classic make it extremely effective and easy to operate
> especially as they don't actually interfere with the core functionality. yes
> classic on its own was pretty dire, but the very sophisticated suite of
> add-ons make this argument pointless.

i completely disagree. what add-ons specifically are you wanting a comparable BOINC add-on for? and notice that these are ADD-ONS. they were NOT AROUND IN MAY 1999, most came much, much later. so at this same point in the timeline, you would only have the cli or screensaver, and one or two add-ons.


> Especially as BOINC will not work with
> less than 4gb of disk space.

also completely untrue. i have it running on a system where the boot drive is only 4gb, BOINC gets the same % as all my other systems. there are 5 projects running on that system.

> As far as documentation goes there isn't any.

there is tons, you just appear to refuse to look at it when we point the way.
there are 52 pages of messages in this forum, and Paul is approaching 800pages at his site.

> If they want me to run something for them then I would expect a something more
> forthcoming thna to be expected to find some strange un-connected and
> unrelated web site elsewhere.

how is that different then the beginning of classic? and all you NEED to know to RUN is right here on the main page.

> I have never said I am unwilling to change but if the darned thing won't work
> then it won't work, if the team cannot be bothered to provide documentaiton
> adn a proper method of carrying over comparable 'credits' then something is
> amiss isn't there? Or is there a suggestion that 80,000 BOINC users are doing
> more SETI than 400,0000 live classic users?

actually, yes, there is that suggestion. it has also been discussed in many of the threads you are posting in.
we, the public, are attempting to provide the information to make it work, but credits WILL NOT be converted. There WILL be a snapshot of the total wu's you contributed to Classic, but as the developers and admins have stated, they cannot be compared. and many thanks to Ingleside, Rom, and Matt for answering those questions so often.
>
> I run several copies of the classic CLI, not the gui.
as did i, using setiqueue for local proxy and setidriver on each system

> When I looked at BOINC
> last year there didn't appear to be anything other than the video version (or
> ot was abysmally explained).

you get all of it, what you use is up to you, again, did you look at the install dir, or just start complaining without ever installing?

> When asking about alternative versions (remember
> the classic CLI is not the same as the screensaver version) it explicitly said
> that people can compile their own. When questioned about this no answer was
> forthcoming.
probably because only about 5 people have actually compiled it? and probably only 1 or 2 have successfully done so in windows.

> So perhaps if the team had looked at what was in use last year
> and bothered to explain how the functionality could be replicated with the new
> version and provided proper documentation on how to set it up and stopped
> remote control over client disk space then all might have been different. As
> it is anybody reading the concerns should immediately see the marketing
> failures.
>
how many times can we say it? they are developers and scientists, not not business marketing majors. any marketing skill should be set solely their grant writing proposals.
as you constantly bring up the 80,000 BOINC users, it would seem obvious that 80,000 of us easily understood the new system, or were able to get our questions answered either by reading the fora or asking a question in said fora.
so you wish to change everything because 1 person out of 80,000 doesn't understand? what kind of sense would that make?

> I think the point is being missed in all the peripheral detail, which is a bit
> of 'while I am complaining there are these things that need attention as
> well', is that the full and proper documentation, settings, configurations,
> locations, etc MUST be publicised, instead of trying to filter through
> thousands of message threads and strange headed queris and so on. Maybe the
> information is out there but if we can't find it or it is obscure how are we
> to be able to spot an alien electromagnetic signal?

WE aren't supposed to find it, we are just crunching data. i certainly know i wouldn't know what it looked like if one of my machines found it, that is why it is sent back to berkeley.
as for documentation, it is really straightforward. and if you haven't been able to figure it out in 2 minutes by looking at the gui (i know, i know, you want to run cli, but look at the mgr gui once), then with all of your posts here, any questions appear to have been answered.

most of the classic add-ons are just duplicate ways of doing the same thing, if you need the functionality of one, list it and we will try to help you get the same type of result.

Micah

---

S@h Berkeley's Staff Friends Club

Thanks for the nice words guys...

One of the points about a project like this is that those of us that can help are allowed to help. I can write pretty well (though I am also a software developer, of course I am also mentally ill, so that may explain it ... an insane programer ... we know he must be insane, he writes clearly ...)

Also, there are people that have looked at my material and have made MANY contributions, in some cases I coiuld take what they gave me verbatum, though most required some editing, but the point is, I do that instead of asking the half a dozen developers to work on documentation rather than the program itself. So, I fill in the gap.

Questions people ask me in the FAQ I answer along with ones here. If good answers come up I steal the answers and post them also. Just today I grabbed one for additional examples for the credit process, and Matt added material so I can fix the place where we talk about the name of the SETI@Home WOrk UNit.

The statistic sites could also be singled out as doing something that, by your definition, should be done by the project. To be honest, I think they should take it out of the project site ... why? because what is there does not justify the cost of supporting it with the project servers. What is there is so weak that it does not hardly qualify as information.

---

so if i can get our school to deploy this it can mass install on all 200 comps. they have deepfeeze though so is there a way to install it on the servers at the school where deepfreeze is off or is that what it does anyways

---

~boinc its not the credit or satisfaction its the screensavers~
S@h Berkeley's Staff Friends Club ©members

> >
> > The add-ons to classic make it extremely effective and easy to operate
> > especially as they don't actually interfere with the core functionality.
> yes
> > classic on its own was pretty dire, but the very sophisticated suite of
> > add-ons make this argument pointless.
>
> i completely disagree. what add-ons specifically are you wanting a comparable
> BOINC add-on for? and notice that these are ADD-ONS. they were NOT AROUND IN
> MAY 1999, most came much, much later. so at this same point in the timeline,
> you would only have the cli or screensaver, and one or two add-ons.
>
>
Yes! The point is about incremental development. So why go backwards instead of forwards? BOINC may well have included equivalents of these add-ons, but comparing them side by side I know which is easier. Especially as the add-ons do not actually interfere with the core cli operation. I use as well as the x86 cli, seti-spy, seti-stash, seti-watch (which has a serous memory leak), seti-log, and occasionally seti-mon. I run 2 instances on a Xeon, with set affinities, and another on a XP2 PC. I encapsulated (until I started Seti stash) all of this within a series of my own batch scripts. No problem, I am full control of what is happening, so I am happy. I leave the things to run themselves, so again I am happy. If a problem occurs (apart from internet and berkeley network or power probs) then I can fix it easily. Nothing in BOINC seems to replicate the above. If someone in the team can provide a documented answer with the package to allow the above scenario to be adapted then fine, BUT, they don't.

> > Especially as BOINC will not work with
> > less than 4gb of disk space.
>
> also completely untrue. i have it running on a system where the boot drive is
> only 4gb, BOINC gets the same % as all my other systems. there are 5 projects
> running on that system.
>

The response I got was that a wu would not download because of insufficient disk space!! I merely reported what it told me. I don't doubt that it runs (esp as it has now actually started to do that after getting 4.2) but it wouldn't, illustrating a serious issue with the preferences system.


> > As far as documentation goes there isn't any.
>
> there is tons, you just appear to refuse to look at it when we point the way.
> there are 52 pages of messages in this forum, and Paul is approaching 800pages
> at his site.
>
Where, why is this NOT on the seti front or download page, and why isn't it with the package of downloads?

How does anybody expect arcane complex queries to be resolved by having to trawl around the net (no I am not) or read through those 52 pages (tens or hundreds of MBs) of messages to see if one is or isn't relevant? Come on get real!

> > If they want me to run something for them then I would expect a something
> more
> > forthcoming thna to be expected to find some strange un-connected and
> > unrelated web site elsewhere.
>
> how is that different then the beginning of classic? and all you NEED to know
> to RUN is right here on the main page.
>
> > I have never said I am unwilling to change but if the darned thing won't
> work
> > then it won't work, if the team cannot be bothered to provide
> documentaiton
> > adn a proper method of carrying over comparable 'credits' then something
> is
> > amiss isn't there? Or is there a suggestion that 80,000 BOINC users are
> doing
> > more SETI than 400,0000 live classic users?
>
> actually, yes, there is that suggestion. it has also been discussed in many of
> the threads you are posting in.
> we, the public, are attempting to provide the information to make it work, but
> credits WILL NOT be converted. There WILL be a snapshot of the total wu's you
> contributed to Classic, but as the developers and admins have stated, they
> cannot be compared. and many thanks to Ingleside, Rom, and Matt for answering
> those questions so often.
> >
> > I run several copies of the classic CLI, not the gui.
> as did i, using setiqueue for local proxy and setidriver on each system
>
> > When I looked at BOINC
> > last year there didn't appear to be anything other than the video version
> (or
> > ot was abysmally explained).
>
> you get all of it, what you use is up to you, again, did you look at the
> install dir, or just start complaining without ever installing?
No I look before installing if it isn't there then inevitably complaints get generated. I can only see a boinc.exe!!!!! and lots of other stuff. No docs so that is the problem. Why doesn't the mgr gui have a help text?


>
> > When asking about alternative versions (remember
> > the classic CLI is not the same as the screensaver version) it explicitly
> said
> > that people can compile their own. When questioned about this no answer
> was
> > forthcoming.
> probably because only about 5 people have actually compiled it? and probably
> only 1 or 2 have successfully done so in windows.
>
Perhaps that was an interpretation but at the time I also recall that it stated that there was only a gui version. Since I and hundreds of thousands of others run the cli, what were we to think? It isn't in the teams interest to rely on the screensaver, as the graphics wastes cpu cycles. So yes it can be switched on or off but that wasn't forthcoming or explained last year and in fact still isn't.

The solution to queries is not to ignore them or tell them to find it for themselves. The solution, which is good customer service practice is to help them. If something is common then it should go and obviously titled into a FAQ list. As well as that an up front statement should also be made.


> > So perhaps if the team had looked at what was in use last year
> > and bothered to explain how the functionality could be replicated with
> the new
> > version and provided proper documentation on how to set it up and
> stopped
> > remote control over client disk space then all might have been different.
> As
> > it is anybody reading the concerns should immediately see the marketing
> > failures.
> >
> how many times can we say it? they are developers and scientists, not not
> business marketing majors. any marketing skill should be set solely their
> grant writing proposals.
> as you constantly bring up the 80,000 BOINC users, it would seem obvious that
> 80,000 of us easily understood the new system, or were able to get our
> questions answered either by reading the fora or asking a question in said
> fora.
> so you wish to change everything because 1 person out of 80,000 doesn't
> understand? what kind of sense would that make?
>
Hmmm I think that reinforces lots of perceptions. Developers and scientists are very important, but so is the social and operational expertise. If something is developed that few want then what is the point. Why did VHS eliminate Betamax? Perhaps your same argument should be turned round and ask why are the 400,000 active classic users still there? That is what will identify what needs to be resolve. What would you rather have 400,000 happy number crunchers or 80,000?


> > I think the point is being missed in all the peripheral detail, which is
> a bit
> > of 'while I am complaining there are these things that need attention as
> > well', is that the full and proper documentation, settings,
> configurations,
> > locations, etc MUST be publicised, instead of trying to filter through
> > thousands of message threads and strange headed queris and so on. Maybe
> the
> > information is out there but if we can't find it or it is obscure how are
> we
> > to be able to spot an alien electromagnetic signal?
>
> WE aren't supposed to find it, we are just crunching data. i certainly know i
> wouldn't know what it looked like if one of my machines found it, that is why
> it is sent back to berkeley.
> as for documentation, it is really straightforward. and if you haven't been
> able to figure it out in 2 minutes by looking at the gui (i know, i know, you
> want to run cli, but look at the mgr gui once), then with all of your posts
> here, any questions appear to have been answered.
>
> most of the classic add-ons are just duplicate ways of doing the same thing,
> if you need the functionality of one, list it and we will try to help you get
> the same type of result.
>
> Micah

Yes I know. But the point was intended not to mean us participants (you read it too literally). The comparison was made with the difficulties of us finding documentation and answers. If we make it difficult to get our our own answers to something we have created then scientifically are we sure we could spot something significant that an alien intelligence has done? But, as an electronic communication engineer I am more than aware of how to find signals and just glad I don't have to do the analysis by hand, the issue is what signal are we looking for? How do we know that we are looking for the right thing? In human terms we perhaps may be trying to spot the oncoming headlights while looking at the sun. Perhaps the spectrum we need to be looking at is not what we are? Who knows, until we stumble on how the aliens actually structure their communications. Thinking holistically or latarelly generates such profound questions. We need to ensure that the same is applied to other matters including presentation and architectures.

In any event I am not going to compromise my classic statistical position unnecessarily. Moving to BOINC will do that unequivocally, which is something else that generates opposition. The failure to acknowledge that (wu comparability and so on) is a serious problem. Ultimately all these complaints may well have been avoided by a simple transference of credit equivalence in the charts, and not just a list of the numbers but a confirmed and continuously up to date(!!) statement of classic standings, even if BOINC wus are not being run. Simple marketing and presentation!

---