[Nexus-developers] [Nexus] NeXus API 4.0.0

Tue May 29 15:59:01 BST 2007

I hate to embarrass myself, but an anecdote is in order.  The biology nexus
file format is, unfortunately, extensible and has many different named block
types.
It's entirely possible that there could be a block type called "Muon" or
"Neutron"
or "XRay" or perhaps something like the pre-existing "DISTANCES" block that
contains a matrix of numbers.  This is in fact what misled me into
mistakenly
packaging the wrong NeXus library for Debian.  I was completely confused by
the name, did not realize the capital X was crucial, and thought that the
XRay
part was just a special extension to the more generic and extensible Nexus
file format.  This was of course a bad assumption but I suspect I will not
be
the last to make it.  Incidentally, does anybody know a good reason why we
cannot solve the problem by combining the two libraries some day?  e.g.
unifying the formats and adding just a new block type or a few new types
so that one format can subsume two seperate and incompatible ones without
two much headache?  I am hoping for any solution since, for example, we
are already packaging both libnexus and libNeXus for Debian at the moment
because there is substantial interest in both libraries.  And I have pretty
much
finished the packaging for the NeXus so far.  If the NeXus committee does
not
make a disambiguating name, packagers will have no choice but to make one
up for themselves and this couldlead to a bit of fragmentation perhaps in
both
library's official name.

Best regards,

Rudi

On 5/29/07, Nick Maliszewskyj <nickm at nist.gov> wrote:
>
>  An unfortunate coincidence, particularly since we have no trademark. It's
> a bit
> late in the game to switch names, as both collaborations are older than
> ten years
> and have some name recognition. Does the UnUsuaL capitalization help us
> stand out?
>
> In any case, when I use Google to search for "nexus data format" I
> do get both nexuses (nexii ?) but each format has a different context.
> Ours is associated with neutron,x-ray, and muon data, and the other
> is associated with systematic biology. I think it's reasonable to assume
> that the end-user searching will be subject to some confusion, but
> should still select the correct format according to its context.
>
> Perhaps littering our links with neutron/x-ray/muon will help out
> the search engines to better steer our users to the right place...
>
> Nick
>
> Rudi Cilibrasi, Ph.D. wrote:
>
> Agreed, it seems an excellent release.  However, I would also
> ask if the name itself for the library is "set in stone"? Unfortunately,
> there is another scientific data format called the "Nexus" file format
> and it is used in bioinformatics. You can read about it here:
>
> http://awcmee.massey.ac.nz/spectronet/nexus.html
>
> The confusing part is that the other Nexus file format has an
> extendable block format.  It can very well be misunderstood to contain
> an X-Ray or other radiation type data block.  It already contains many
> other data block types and so this will probably continue to be a source
> of confusion for all newcomers to either file format potentially, since
> most people type nexus into the search box anyway both results come
> up intermingled and thus suggesing that the two libraries are the same
> format when in fact they are totally unrelated.
>
> The bioinformatics Nexus format has been around about 10 years now I
> think.
> I just
> checked the NeXus mailing lists and noticed that that use of the term has
> also
> been around for about 10 years, so I'm not sure which use came first  So
> probably eventually
> one of the two file formats will need to change names I think; otherwise
> there will be no end of confusion due to the similarity between
> Nexus
> and
> NeXus
> and the confusing similarity of scientific file format domains.
>
> I look forward to any ideas for resolution of this problem.  Best regards,
>
>
> Rudi
>
> On 5/26/07, Ray Osborn <ROsborn at anl.gov> <ROsborn at anl.gov> wrote:
>
>
> On 2007/05/25 8:27, "Nick Maliszewskyj" <nickm at nist.gov> <nickm at nist.gov>wrote:
>
> > The NeXus technical committee is pleased to announce the official
> > release of version 4.0 of the NeXus API. The NeXus API provides a
> > unified interface for writing hierarchical data files in HDF4, HDF5,
> > and XML formats. Version 4.0 resolves a number of bugs reported in the
> > 3.0 release by providing updated makefiles, support for the gfortran
> > compiler, and an improved test suite.
>
> Congratulations for all the work done in getting this release out - I
> think
> it's been a very well conducted process.  Could I put in a plea though
> that
> someone updates the documentation on the API wiki page to include the new
> routines?
>
> Obviously not until after the Memorial Day (or Whitsun or Pfingsten...)
> holiday.
>
> Have a great weekend,
> Ray
> --
> Dr Ray Osborn                Tel: +1 (630) 252-9011
> Materials Science Division   Fax: +1 (630) 252-7777
> Argonne National Laboratory  E-mail: ROsborn at anl.gov
> Argonne, IL 60439-4845
>
>
>
> _______________________________________________
> NeXus-developers mailing list
> NeXus-developers at nexusformat.org
> http://lists.nexusformat.org/mailman/listinfo/nexus-developers
>
>
>
>
> ------------------------------
>
> _______________________________________________
> NeXus-developers mailing list
> NeXus-developers at nexusformat.orghttp://lists.nexusformat.org/mailman/listinfo/nexus-developers
>
>
> --
> ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
> o Dr. Nicholas C. Maliszewskyj
> o Center for Neutron Research
> o National Institute of Standards & Technology
> o 100 Bureau Drive, Stop 8562
> o Gaithersburg MD 20899-8562
> o nickm at nist.gov     Phone: (301)975-3171    Fax: (301)921-9847
> ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
>
>

-- 
Democracy is two wolves and a sheep deciding what to eat for lunch.
Liberty is a well-armed sheep contesting the vote.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.nexusformat.org/pipermail/nexus-developers/attachments/20070529/c13ac48c/attachment.html 