On Wed, Feb 17, 2010 at 12:58 AM, Geoffrey Hutchison <[email protected]> wrote:
> One of my (unstated) goals with Open Babel is to support as many file formats 
> as possible. Leave no chemical data behind and process it as accurately as 
> possible.

Let there be no mistake here... I do not like to leave behind any data
either! The reason why format specifications are important to me (open
or not, see below), is that most of them cause *modeling error* ...
and that's get pushed in the your statistical analysis, so, in order
to do the latter properly, and fairly judge the quality of the
modeling, you *must* know the error involved in reading the data from
disk.

Some format are better than others, have less limitatations, more
flexible: CML is very flexible, and information loss is just minimal;
(Open)SMILES does not handle coordinates, and will never handle
conformational variance properly...

That's the kind of thing that matters to me...

> I support MDL, CDX, MOL2, and other formats which might be considered 
> "non-open" by Egon.

... and to a much lesser extend if a format is open or not! So,
offense taken here!

I brought the whole discussion up, just I feel the current
*guidelines* are not enough for me to decide what is open and what is
not... it is not me who made claims in the Blue Obelisk wiki (well, I
hope not)...

I cannot find good arguments to say that Daylight SMILES, MDL, ... are
not as Open as some of the specification that we deem Open...

I am only fairly sure about OpenSMILES... the rest is to me undecided yet.

> No offense to Egon, but he's not part of the OB project.

No offense taken on not being OB contributor. In a multi-universe
where I'd have a subtle knife not to just move between worlds, but
also back and forth in time, I'd be a happy OB contributor...

> (Sadly, Blue Obelisk = BO, which is an easy typo...)

I am still loving to see a proper OpenBabel plugin for Bioclipse...

> I like Peter's comments about "rough consensus and running code." That's very 
> much in line with my view. Tests are also welcome.

I recapitulate my note on this, and take OpenBabel as a lead here...
OpenBabel *is* the "rough concensus and running code" for MDL/Symyx
molfile, Daylight SMILES, CDX, ...

And this is my point all along... Open Source is making a strong point
for existing formats (which is good), and as such, makes existing
rules on Open Specifications/Standards less useful...

Egon

-- 
Post-doc @ Uppsala University
Proteochemometrics / Bioclipse Group of Prof. Jarl Wikberg
Homepage: http://egonw.github.com/
Blog: http://chem-bla-ics.blogspot.com/
PubList: http://www.citeulike.org/user/egonw/tag/papers

------------------------------------------------------------------------------
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________
Blueobelisk-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/blueobelisk-discuss

Reply via email to