On Thu, Jan 14, 2010 at 6:23 PM, Lynn W. Taylor <[email protected]> wrote:
> Jord,
>
> I certainly understand what you're talking about.  I've been ridiculed for
> my small contribution in terms of credit.  Doesn't bother me, because
> despite my enthusiasm for BOINC and s...@home, I'm more representative of
> the average cruncher than the typical forum participant.

It's not that I am being ridiculed for my small amount of credit, even
though at one time on the same Seti forums, someone I tried to help
with his BOINC problems wondered what I could possibly know about the
program, since my credit was only 1/3rd of what he had accumulated
within his short time there. I simply put him in his place by helping
him anyway, leaving most of the comments about my lousy credit to
others.

> First, I wouldn't assume that Rom has become a credit hound, I would assume
> that he is using credit as intended (as a measure of computing) and that
> it's more about the amount of work that can be done with a GPU.

I gave Rom as a little (yet ill conceived probably) example, as he's
closer to the development than any of us mere mortals are. Having
rewritten the initial email some 7 times, I left that part in as for
me it did matter and it was my only option to put the thread title in
context. ;-)

> Second, quite some time ago I was told that roughly 8,800 participants at
> s...@home had posted to the forums out of something around a million
> participants (according to BOINCstats.com).

Ah yes, but as ever, BOINC is so much more than just Seti. So how many
people are active at all those other projects that aren't Seti, and
how many of them do it for the science? Let's just take Milkyway as an
example.

It's true that those that do it for the credit do science as well, but
it's also proven and true that they'll leave one project for another
when the 'pay' gets less. Or they'll flame just about anyone who
doesn't see things their way. You can have an opinion about this, as
long as it's the same as ours, or we'll make your life miserable
everywhere we can.

On Thu, Jan 14, 2010 at 5:12 PM, Adrian Worley <[email protected]> wrote:
> As long as the credits are earned and not fraudulent, why does it matter
> that a few forums are full of kiddies. After all, the credit represents work
> done, and that is why we are in it isn't it?

Since those credits gotten at other projects follow no clear
definition of credit (as said by David in this thread), how can you be
sure they are "earned and not fraudulent"? The problem between
projects is that their applications do the science in no standardized
way, there is no guide line that says/shows how all BOINC projects
should get to the same credit per bit of science done. Most
problematic is that none of the 'workarounds' are mandatory, BOINC
instead gives the projects the choice to use them.

So one project still uses the benchmarks times time method, another
has the credits fixed on the server, a third will hand it out by hand
as and when admin has time and a fourth will follow yet another
method. In the end, when you compare the projects, you'll find that
they all "pay" differently. Which one is correct?

And that's based on the CPU only... I am not even going to add the GPU
option in there as that's a hornets nest in itself.

I know I can ignore all those threads on credit, yet this crap will
bring itself into any thread that's out there. There's not enough
"filter" positions in the forum software to put everyone in there.
There's no option to filter specific threads that hold words like
credit or RAC (which in itself isn't a word).

And although I know that my initial post on this was focused around
the credit shambles, I also tried to point out that the design goal of
the BOINC program goes lost. Luckily you all missed that or didn't
want to react to it.

In the past, when comments came on slow-downs and such, the developers
would try to fix that. Everything for idle time. But with the newer
hardware being incorporated the view changed. You got slow downs?
We'll change our advice as a fix is too difficult, too complicated to
include in the present code without a whole rewrite, it needs specific
drivers (so point to the GPU makers), or we'll put it on the to-do
list for a future version of BOINC. And when that future version comes
on the horizon, make sure to move it to a next future version.

I know that's not honest.
I know that David, Rom and Charlie do everything in their power to make amends.
I know that their to-do list is huge.
I know that some things just can't be done, either at this time, or ever.
I know life isn't fair.
I know I would do and probably still will do everything in my power to
help them, if I could, if I had the knowledge.
Just not now.


-- 
-- Jord.
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.

Reply via email to