New list of words

[Alberto Micol]

Dear Andrea,

>  I'm sure you'll have other before Kyoto!

I'm taking it seriously :-) Here some other comments, in no particular 
order:

1. pos.frame exists, can we also add the spect.frame and time.frame?

2. instr.spec: probably not needed
    because otherwise we would also need instr.imager, and similarly for 
other kinds of instruments.
    Also,
    instr.spect.dispersion could simply be instr.dispersion
    instr.spect.order        "     "     "  instr.order
    (in fact there is no need to say instr.imager.scale, instr.scale is 
enough).

3. Transmission (an old friend of mine): some time ago Sebastien and I 
exchanged some emails
    and, if I remember correctly, we agreed that there should be a 
phys.transmission
    which could be attached to any of: instr.det, instr.filter, instr to 
distinguish
    between the transmission/sensitivity of the various components 
(detector, filter, optical elements)
    and the overall transmission (filter * detector * optical elements) 
including even the atmosphere
    if necessary.

4. instr.sensitivity; Its description reads: Instrument sensitivity, 
detection threshold
    I would say that sensitivity and detection threshold, while linked, 
are two different concepts,
    and would be better to split that one into 2 different UCDs

5. Descriptions: sometimes the description of a ucd is ambiguous, e.g.:
     a. instr.tel.focus (Q): Telescope focus
     b. instr.bandpass (Q):  Bandpass of instrument
     c. instr.pixel (S):     Pixel

     a. Is instr.tel.focus the "name" of the telescope focus (eg 
nasmith),
        or the focal length of the telescope?

     b. instr.bandpass
        How many services will make use of this ucd to describe a table 
column that contains
        the numeric transmission at a given wavelength of a given filter?
        Some other service will use this ucd to describe the name of the 
bandpass (or the filter name).
        Which one of the two is legitimate, both? (see (3) above).

     c. Another, though different, example is the instr.pixel
        I can think of a catalog associated with an image, where X and Y 
indicate
        at which detector coordinates the source was found 
(pos;instr.pixel).
        But it can be used to indicate the pixel scale of the detector 
(phys.angSize;instr.pixel)
        The fact is that instr.pixel is a Secondary word, hence its 
meaning could be broader.
        For that reason probably both above mentioned cases are 
legitimate: instr.pixel is only an
        attribute (an adjective in a grammar-speak).

     I think it is very important to be clear, especially on those words 
that could be Primary,
     because otherwise people might misuse UCDs and that would make the 
entire concept collapse.


6. Abbreviations: Why temperature is some times abbreviated and some 
other times not?
    (e.g., instr.skyTemp, phot.antennaTemp, phys.temperature) Wouldn't 
be better to be
    consistent?


7. wcs information
    pos.wcs.ctype: How would the software recognise ctype1 from ctype2 ?
                   Would it be possible to say: 
pos.wcs.ctype;arith.first and pos.wcs.ctype;arith.second?
                   Or maybe (see cdmatrix below): arith.1 and arith.2
    The same for pos.wcs.crpix, crval and naxis

    pos.wcs.cdmatrix: how to distinguish the various matrix elements?
                      what about accompaining the cdmatrix with a 
arith.1_1, arith.1_2, etc.?
    That arith.1, etc. could be useful to rank things.


8. question regarding files
    meta.file exists; how can I describe the size (eg in bytes) of a 
file? and its CRC?


Probably enough for this morning... ;-)
And, beware, I might come back with more DM issues. :-)

Ciao,
Alberto