8/9/03.
David,
As promised, herewith some more detailed comments on `Transformations
in the VO Data Model'.
* The introduction is a bit terse, so I'd wondered about additional
`motherhood' statements. Maybe something along the lines of: `there are
numerous uses of such transformations in the VO. For example, to allow
the automatic uniform presentation and comparison of equivalent
quantities stored with different representations and units, such as
wavelength/energy/frequency comparisons for electromagnetic radiation.'
On the other hand, the document is for a VO-savvy audience, so you don't
need to go over the top. I'm not sure.
As someone mentioned, you do need to address why FITS WCS is not
adequate.
* Has AST really been in use for 8 years? Tempus fugit!
* `The Mapping Class' and throughout the document. The document seems
to be presenting a set of abstract ideas about how transformations
might be handled in the VO. On the other hand the text is couched in
Object Oriented terminology and ideas, which makes it sound as though
it is describing a piece of software which either exists and you want
to adopt or you're specifying. If the document is describing a piece
of software (existing or proposed) you should say so; if it is
describing abstract ideas you should explain the use of O.O.
terminology in this context.
* `The Mapping Class', para 1, sentence 3:
- change `... one of these transformation' to `... one of these
transformations'
- there is a spurious closing parenthesis at the end of the sentence.
* `The Mapping Class', para 1, last sentence:
- change `... Mappings makes no ...' to ... Mappings make no ...'
- the footnote refers the reader to a `Package Mappings' section
which I couldn't see.
* In the first table, listing the properties of a Mapping, wouldn't
it be more natural for TranForward to occur before TranInverse?
* Note 2 to the above table, change: `" forward"' to `"forward"'
and `" inverted"' to `"inverted"'. (There were numerous similar
problems, I'll note the ones I spotted as I go along. Presumably
it was some sort of format conversion problem?)
* The second table, about the principle methods of a Mapping. In
the purpose of `Differentiate' change `specified input' to
`specified input axis'?
And I think that `principle' should be `principal'.
* `Compound Mappings', first paragraph, last sentence. The meaning
of `earlier' and `later' coordinates was not obvious (well, not to
me). I think that you mean something along the lines of: `... (where
one Mapping transform operates on one subset of the input axes
and the other Mapping transform operates independently on a different
subset)'.
* `Atomic Mappings', para 1, sentence 1:
- change `" x+y"' to `"x+y"' etc.
- change `etc., etc.' to `etc.'.
* `Atomic Mappings', last para, first sentence. Change `" packaged"'
to `"packaged"'. And again in the next sentence.
* `Atomic Mappings', last para, penultimate sentence. You've italicised
`etc.' but didn't earlier. You should be consistent.
* Appendix A. Do you need to say `Appendix A' rather than just
`Appendix', since there is only one?
* Appendix A, para 1, sentence 2: change `" *"' to "*"' and `" nin"'
to `"nin"'.
* The table in the Appendix. In the row for `TestMap' and the column
`Function' change `equalt' to `equal'.
HTH.
cheers,
Clive.
-----------------------------------------------------------------------------
Clive Davenhall Institute for Astronomy,
e-mail (internet, JANET): acd @ roe.ac.uk Royal Observatory Edinburgh,
fax from within the UK: 0131-668-8416 Blackford Hill, Edinburgh,
fax from overseas: +44-131-668-8416 EH9 3HJ, Scotland.
|