[fitsbits] FITS on Wikipedia

Rob Seaman wrote:
.. google FITS: ..

http://en.wikipedia.org/wiki/FITS

from which we find, for instance, that "FITS is the most commonly used
digital file format http://en.wikipedia.org/wiki/File_format in
astronomy http://en.wikipedia.org/wiki/Astronomy." ..

This URL was a pleasant surprise for me. Its description of FITS is
terse, but precise, and contains the necessary hints about the
generality of the design, capabilities and usage of FITS, while still
acknowledging that simple FITS image files are very common. The only
name that I see in a brief glance at the history page is that of a
graduate student at Manchester, Mike Peel:
http://en.wikipedia.org/wiki/User:Mike_Peel
I say 'Well done, Mike Peel!'.
Did anyone else contribute to this Wikipedia page? If so, please
identify yourselves, so that you too can get some thanks and praise.

-=-=- Suggestions -=-=-

* Probably the Wikipedia text should contain a URL for the FITS MIME RFC:
http://www.rfc-editor.org/rfc/rfc4047.txt
* Probably the section about tables in FITS should at least hint at the
full generality of BINTABLE, especially the concept of (variable length)
multi-dimensional arrays in fields.
* Citation of HDF, CDF and NetCDF in the "See also" section is
appropriate. Perhaps GeoTIFF ( http://en.wikipedia.org/wiki/Geotiff )
should be added to that list of data formats that are reasonably
comparable to FITS. I suggest that terse comparisons of the features of
these formats with the features of FITS be added to the links; Wikipedia
is a a truly 'open' project, so these descriptions could be added by
anybody in the FITS community.
* Wikipedia is multi-lingual. The FITS community is multi-lingual. It
may be appropriate for members of the FITS community to produce
translated versions of this first FITS Wikipedia page.
* Perhaps the Wikipedia URL should be cited in some of our FITS web
pages, along with the MIME-code URL.

-=-=- On planting descriptions of FITS -=-=-

From the beginning of FITS, long ago, one of our great goals has been
to plant precise descriptions of the format in places that we judge will
have a chance for long-term survival, motivated by the need to support
FITS as an archival data format. The primary place that we chose was
one of the major refereed standard journals of astronomy, Astronomy &
Astrophysics, a journal that is likely to survive as paper copies (or
electronic equivalents) in several hundred libraries around the World
for the next few centuries. We (the FITS community) have also planted
a wide variety of FITS-related documents in many secondary institutional
archives that are likely to survive for at least the next few decades.
Finally, and most recently, the registration of FITS as a MIME-code RFC
embeds a precise description of the format in the machine-readable
documentation of the Internet. I expect that MIME RFCs will still be
readily available in 2032, and I wouldn't be surprised if they prove to
survive to 2107. This idea of planting FITS descriptions in archival
places was a key personal motivation for me as I was promoting the MIME
RFC project in the 1990s, in addition to the obvious current operational
need for official MIME codes for FITS. I consider this Wikipedia page
to be yet another place where a precise description of FITS has been
planted. While the Wikipedia project is often criticized, it is now
clear that it has succeeded in creating an invaluable general
information resource, a resource that is likely to survive for decades
at least, and perhaps for a century or more.

My own mental model in this matter is to imagine that some astronomy
professor in some distant time tells an astronomy graduate student to
compare some data recently acquired with related archival data. The
professor might or might not be knowledgeable about FITS. FITS might or
might not be in regular use in that distant time. There is no assurance
at all that any of our current software environments will survive in
that era, so the graduate student might have to crack our FITS files
with custom software. I hope that the FITS header will contain a
precise bibliographical citation to one of our journal articles, a paper
that will aid that graduate student in designing the custom software.
I hope that a search operation finds information about FITS. The
graduate student will know astronomy, so the context of the metadata in
the headers will make sense, and will raise questions in the mind of the
graduate student that will trigger other searches, and I hope that the
answers to those questions will soon be found so that our files can be
interpreted properly.

Originally, in the early days of FITS, my concept of "a distant time"
was 40 years, long enough that the early FITS pioneers would be retired
(like me). Now, with a whole new generation of FITS pundits in the
profession, it is clear to me that "a distant time" reaches much farther
into the future, and I even dare to dream that FITS will be acknowledged
and understood in 2107, even if it is no longer in regular use. Even
though I am now (guardedly) optimistic about this, I still say that we
have a duty to plant precise descriptions of the format in multiple
places that are likely to survive, and that we have a duty to put
precise citations in all of our FITS headers.

-Don Wells