Does OPM target industry prior to academia? Could you please elaborate on that?
I am not sure that question makes any sense. The question assumes a conflict, while OPM (at least for me) is very much about bridging gaps between academia and industry. I would rather say that OPM seeks to enable direct collaboration between industry and academia. To my reality (academia) having an entirely open software/specification ecosystem is extremely important. It's in fact the only possible way to make our work reproducible by other researchers and interested people [1<http://en.wikipedia.org/wiki/Reproducibility>,2<http://reproducibleresearch.net/index.php/Main_Page>]. Fully agree. OPM certainly wants to address that. It saddens me if you have another impression. Even more, it is an attempt to enable academic research at the source code level, building on each others code (which is arguably a leap further than simply reproducing). Ideally, I would love to see academic papers where commits to OPM is part of the academic work, so that the next researcher easily can build on top of the previous work also at the implementation level. Going back to that new dependency opm-core <- opm-parser... It's the first time I see *-core depend on something else. As a developer I would expect the core of anything to be self-sufficient. Agree, actually I have already argued that we should have file IO stuff in opm-core. I see it as a natural part of a core library. Whether that view prevails and opm-parser is integrated in the core repo remains to be seen though. There may be other factors and views weighing in. A format would help describe complex reservoirs with pure OPM code without the need for external (proprietary) tools. And that is where the problem is at. A proper answer to that issue is a bit more involved than what I have time for now. I will try to provide my view on it schematically though. I fully agree with your conclusion, we should not need proprietary tools. However, we are not there now. If you want to address industrially relevant problems (which I really do believe academic communities addressing petroleum technology should do), you already are in need of a lot of input to even get started. You need a PVT model, which is typically exported in a format a commercial simulator can use. You need a geological model and you cannot realistically make it without a geomodelling package (all geomodelling packages support Eclipse formats, actually quite a few only supports Eclipse). To address this the Norne model, a current full field reservoir model with everything it entails, has been made available for academic use. All input are in ASCII format, human readable and editable with your favourite text editor. Years before that, the Gullfaks model was made available. You may argue that this is not enough, but it is what we have been able to achieve so far. On the output side, there are binaries, but at least libecl, a library for reading and writing these binaries is openly available to academia. Moreover, we have been able to provide a full visualisation solution through Resinsight that in my opinion already out-competes any proprietary offering for its intended usage (quite an accomplishment don't you think?). The really missing part of the puzzle on the output side then is 2D plotting, typically of production curves. There has been discussions, but frankly I am restrained by the 24 hour a day limit (as many of us are), so do not expect us to have a polished solution on that this year. If we had tried to design and implement our own file format, we would probably not be able to test our implementation on anything realistic for years. Personally, I am a firm believer in having useful code early, and planning the development so that you can start testing your implementations as you go along. Cheers, Alf The information contained in this message may be CONFIDENTIAL and is intended for the addressee only. Any unauthorised use, dissemination of the information or copying of this message is prohibited. If you are not the addressee, please notify the sender immediately by return e-mail and delete this message.Thank you
_______________________________________________ Opm mailing list Opm@opm-project.org http://www.opm-project.org/mailman/listinfo/opm
