I've been quite busy for some days so I didn't read your complaint before.
Unfortunately, we have to clean up the server every now and then, so this
particular result already has gone (from the database; the output file might
still be there).
Can you perhaps tell me some more details (e.g. the name of the result
which would start with "ft" and a number and end with "_0" - the one that
is mentioned in the upload lines) so I possibly could track this down?
There may be several reasons for invalidating a result, file corruption during upload would be the simplest one, but an accumulation of rounding errors -
which differ from platform to platform and have to be properly taken care of
during validation - might also result in a result "too far off" from the
majority of results. A "valid result" is defined by being "close enough" to
what most contributors agree upon... and I'm not yet completely happy with the current validator algorithm, thus your feedback might be helpful. Any chance?
server down?
)
> It is working great, the only problem is this wu was uploading when it
> happend
> now it has a validate error,
>
> http://einstein.phys.uwm.edu//task/559682
Hi,
I've been quite busy for some days so I didn't read your complaint before.
Unfortunately, we have to clean up the server every now and then, so this
particular result already has gone (from the database; the output file might
still be there).
Can you perhaps tell me some more details (e.g. the name of the result
which would start with "ft" and a number and end with "_0" - the one that
is mentioned in the upload lines) so I possibly could track this down?
There may be several reasons for invalidating a result, file corruption during upload would be the simplest one, but an accumulation of rounding errors -
which differ from platform to platform and have to be properly taken care of
during validation - might also result in a result "too far off" from the
majority of results. A "valid result" is defined by being "close enough" to
what most contributors agree upon... and I'm not yet completely happy with the current validator algorithm, thus your feedback might be helpful. Any chance?
Cheers.
Steffen