[bip] Bioinformatics Programming Language Shootout, Python performance poopoo'd

[Paulo Nuin]

Andrew Dalke wrote:
> On Feb 6, 2008, at 2:35 AM, Titus Brown wrote:
>   
>> I suspect that the code could use some systematic code review; "we"
>> (i.e. someone else :) could even write up something semi-formal if it
>> turns out that the results are bogus.
>>     
>
> I've got something written up.  It's mostly complete.  100 lines,  
> plus a summary I haven't figured out yes.  What should I do with it?
>
> Options:
>
>   - publish (BMC Bioinformatics has a $1,870 fee - that ain't gonna  
> happen)
>
>   - I don't see a letters to the editor column, or anything similar.   
> The closest is
>         Queries about content, submissions, or the review process  
> should be
>         directed to editorial at biomedcentral.com
>
> I really want to complain about their lack of rigorous peer review.
>
>
>    - I see there's a spot for comments at
>    http://www.biomedcentral.com/1471-2105/9/82/comments
>
> Does anyone use that?  Or is there some place else which is better?
>
>    - blog it
>   
I was thinking of blogging it. I am registered at Research Blogging and 
I was intending to make my first post there.

> Anything else?
>
> Also, does anyone want to look over my draft version?  This is a  
> public list and I would rather the comments not be widely  
> disseminated yet.
>   
I can look at it if you want.

> I'm also not sure about the politics of dissing BMC Bio's editors and  
> peer reviewers skills on reviewing this paper.  Not that I publish,  
> but how do I make it the most effective?
>   
I think as Bruce mentioned in another message, the problem is not at the 
editor or peer review level, even though I think the peer review in this 
case was extremely poor (even considering that the reviewers might not 
be experts on every language). The paper is very weak, and not only on 
what concerns to Python but the lack of a clear identification of the 
test sets, making some (or most) of the results difficult to reproduce. 
I think the authors are at fault here, and the editor/reiviewers share 
some part of the blame.

Paulo