[Developers] About WITSML data loading in OD

[Bert Bril]

hi Nauman,

> I wanted to ask if someone has already developed WITSML data loading utility
> for OD.

To be perfectly honest I had to Google for WITSML. It's a POSC
diversification:
http://www.witsml.org


> Is there any such utility available?

Not to my knowledge, but hey, I didn't even know what WITS and WITSML were.


> And is OD exploring this direction to load Well log in this format in
> future?

There may be people interested in doing this, by I very much doubt it.
POSC models are nice for reference purposes, to get some ideas so to
say. When it comes to actually using them ... I don't think so.

And then the usage. Some 15 years ago I had to give courses in the 'DAE'
 (Data Access and Exchange) API specified by POSC. Apart from me
disliking the interface very (indeed very) much, as far as I know (FWIW,
see above) this whole effort was a complete failure.

With WITSML they have adapted to the new times (XML is hot). Who knows.
But after glancing through the material I feel the signs of disaster
already:
* Big (IMO oversized) data model impossible to implement without
large-scale efforts
* No (good) implementation of the specification
* The interface is still as ugly as before: it looks very similar to the
DAE interface.

<philosohpicalmode>

I think the problem I have with the POSC approach is that it violates a
very important principle:
"Don't specify what you can't implement"
Rather, they specify until they are blue in the face and then see API's
and object libraries as an extra, nice-to-have.

I think this is part of a complex problem sometimes called 'analysis
paralysis':
http://en.wikipedia.org/wiki/Analysis_paralysis

The Open Source way is to make sure there is a good executable
implementation that can be shared by all. This implementation *defines*
the model. This hurts the soul of true informaticists (data modellers in
particular), and that's why there is still so much money going from
large oil companies into that approach. The decision makers have a
degree in informatics, or are advised by people with degrees in
informatics. And if it fails, for the Nth time, can anyone blame you for
 doing the 'only true and good' thing?

</philosohpicalmode>


All in all, I'd say the chances are low.


Bert

-- 
-- Bert Bril / OpendTect developer at dGB
-- Nijverheidstraat 11-2, 7511 JM Enschede, The Netherlands
-- mailto:Bert.Bril at opendtect.org , http://opendtect.org
-- Tel: +31 534315155 , Fax: +31 534315104