[Dune] [#563] GenericReferenceElement<double, 5> does not work at all (Attachment added)

[Martin Nolte]

Hi Sven, hi Oliver, hi all,

it was not my intention to blame anyone, here. I was just under the impression 
that people generally think that the generic geometry stuff is the major 
source of errors turning up during this transition. It is true that this code 
is very new and may still contain a lot of bugs. But this should not be an 
excuse for writing a bug report without double checking.

I just raise this point, because I always try to react as promptly as possible 
to the error reports. In the last days, there were two such reports that were 
plain wrong. The source of the YaspGrid 4d problem as well as the bug report 
we are talking about were not due to a wrong generic implementation. In the 
first case, the old numbering was simply not known (and Peter promised last 
year in Berlin that he would document this numbering -- no offense meant), and 
in the second case, the error was the enormous size of the old reference 
element. Since I considered them to be my mistakes, I tried my best to fix 
this as soon as possible -- what a waste of time.

So, let me apologize if I hurt anyone's feelings with my comment. I just 
wanted to point out that hese things will definitely make me react much slower 
to such bug reports...

Yours,

Martin


Sven Marnach wrote:
> Hi Martin,
> 
> DUNE Mailing List schrieb am Di, 16. Jun 2009, um 16:08:28 +0200:
>> As you can see, the problem is not the generic reference element,
>> but the old one (8.6MB). If you want us to take errors in the
>> generic geometry stuff serious, please don't make such false
>> accusations. I know this was not intended, but still it cost my
>> precious time... (and reading assembly language is not that hard)
> 
> I just found the error, too, but you closed the report just before I
> clicked the submit button. :)
> 
> Obviously, I made some mistake because I thought the first version I
> posted would raise the error -- that's why I assigned it to the new
> reference element.  Please do not take a bug report as an
> "accusation".  Bug reports are meant to help finding errors.
> 
> Sorry for the bogus report anyway -- I will try to be more careful in
> the future.
> 
> Greetings from Stuttgart,
>     Sven
> 
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune

-- 
Martin Nolte <nolte at mathematik.uni-freiburg.de>

Universität Freiburg                                   phone: +49-761-203-5642
Abteilung für angewandte Mathematik                    fax:   +49-761-203-5632
Hermann-Herder-Straße 10
79104 Freiburg, Germany