On Feb 19, 2010, at 10:11 AM, Peter Murray-Rust wrote:
> Taking a step back...
  ..
> That's the totality of the formality. Our terms are vague. I believe 
> deliberately.

That is not the issue at hand. The Blue Obelisk wiki has historically stated

http://blueobelisk.sourceforge.net/wiki/index.php?title=Open_Standards&oldid=2060

> In general a proprietary format, even if published (MDL, SMILES, etc.), is 
> not Open as there is no community process for its development

and still says things like:

http://blueobelisk.sourceforge.net/wiki/Open_Data_in_Chemistry/Tools_and_data_formats_for_chemical_data_handling

> Molecular data
> 
> Allowed but discouraged are vendor specific formats (like .skc in case of 
> ISIS Draw or SMILES). 

Therefore the Blue Obelisk has specific but unstated reasons for saying that 
certain formats are not open. It is these reasons that I am pushing the Blue 
Obelisk people to resolve.

Yes, my view, and a view shared by almost everyone I've talked to about this 
discussion, including those with experience in free and open source projects, 
is that for at least the last 15 years SMILES and the MDL formats are 
sufficiently open that they do not hinder any progress or project development 
in this field.

In addition, my experience from organizing the Bioinformatics Open Source 
Conference is that many speaker presenters have no idea of what "open source" 
really means. They would submit talks which were "for academic use only". By 
pointing to specific guidelines we were able to make changes in their release 
policy, including helping them justify to their advisors, etc. the usefulness 
of making the software truly free.

Having at least rough guidelines is useful. Otherwise the mantra is meaningless.

> Some of that has been imperceptible change and simply changes of opinion (it 
> now seems that most people regard MDL and SMILES as "open" - that was 
> definitely not true when they were launched).

How was SMILES not open? Everything I've read and heard about suggests 
otherwise. The only complaint you've stated is that the full details of 
Daylight's canonicalization algorithm was never published, but that has nothing 
to do with the use of SMILES as an structure exchange format - which is why 
OpenSMILES does not need to include a canonicalization algorithm.


I've been listening to the podcasts from the Software Freedom Law Center. They 
are very informative. One of their points is that they view the copyright 
breakers (of the GPL) of today as the free software contributors of the future.

Just because MDL was bad in the early 1980 doesn't mean Symyx of today is still 
bad.


                                Andrew
                                [email protected]



------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Blueobelisk-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/blueobelisk-discuss

Reply via email to