2009/12/8 Peter Murray-Rust <pm...@cam.ac.uk>: > Thanks Egon, > Here's my take. > There are very few standards which are Open in the sense that there has been > or is communal Open authoring of the standard. Even where there is community > input the result is usually moderated by a small group. If the results of > discussion and the documents (at all stages) are openly visible then that is > probbaly as close as we can get in practice. > > I've been involved in a number of "standards". > > SAX. The only true Open standard, developed on XML-DEV within a month. A > complete meritocracy > > XML: input from anyone through XML-DEV but final spec by a group of ca 20 > (including paying members) > > CIF. Regular meetings of committee with papers made open for publication and > comment. Final decision by committe vote > > InChI. Originally developed by NIST staff > > ISO12260 (I think). Closed meetings selected by ISO and interested parties. > ISO standards not formally Open > > CML. Dveloped by PMR and Henry Rzepa. Continuously available publicly on > Sourceforge. All decisions by benevolent dictatorship. Opportunity for > anyone to contribute through mailing list. > > IMO Open standards have to have a process. It could be useful for BO to > develop ideas for the meta-process for standards. > > In practice I would see an Open standard as one that is: > * published > * redistributable without permission > * has a process which can be understood and where representation can be made > > I would regard CML, CIF, Java, W3C specs as Open; I would not regard the > following as Open: > * Bruker binary FID format (proprietary, spec not published) > * Marvin XML format > * Daylight fingerprints > * Daylight SMILES (as opposed to Open SMILES).
There is some confusion here. These are not standards. There is no complete spec for example. From the point of view of promoting interoperability and avoiding vendor lock-in, a standard is either a standard or it's not. I think the BO should promote standards, period. I don't find the idea of an "open standard" useful in the context of BO. > If you only use meritocratic specs there will be very very few that are > useful. > On Tue, Dec 8, 2009 at 9:32 AM, Egon Willighagen > <egon.willigha...@gmail.com> wrote: >> >> Hi all, >> >> in reply to a recent private, off-line discussion on what the Blue >> Obelisk means with 'Open Standard', and what the Open Standards in >> chemistry are, I like to ask for clarification. >> >> >> -- >> Peter Murray-Rust >> Reader in Molecular Informatics >> Unilever Centre, Dep. Of Chemistry >> University of Cambridge >> CB2 1EW, UK >> +44-1223-763069 > > ------------------------------------------------------------------------------ > Return on Information: > Google Enterprise Search pays you back > Get the facts. > http://p.sf.net/sfu/google-dev2dev > > _______________________________________________ > Blueobelisk-discuss mailing list > Blueobelisk-discuss@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/blueobelisk-discuss > > ------------------------------------------------------------------------------ Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev _______________________________________________ Blueobelisk-discuss mailing list Blueobelisk-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/blueobelisk-discuss
