[Message Prev][Message Next][Thread Prev][Thread Next][Message Index][Thread Index]

[vsnet-chat 4387] Re: Orthography of V-numbered variable stars?




Thorsten Lange wrote:

>It is very stupid to discuss the number of leading zeros in a star name. You
>can write a simple computer program which is able to use "AndV000123", "V123
>And", "ANDV123" and "And V00123" for the same star.

Oh no I can't! ;)

and if (like me) you have to use some database management programs or some
scripts to do your data processing it tends to be that these are no way
near as flexible data entry format wise as the fortran and/or C that more
capable people are wont to use.

Similarly, I believe it is now an IAU data archiving decision to drop the
use of things like "999.9999" in blank fields and only use spaces.  With
some of the import progs I have to use this means that I cannot easily (if
at all in some cases) strip null fields, as some instantly convert blanks
to 0 for numeric fields, which is not useful if 0 can be a valid value.
On the other hand a numeric field full of nines is invariably meaningless
in terms of the data and readily ignored

Okay, this boils down to my inadequacies at programming...

...incidentally, I wonder if anybody out there uses a fancy sort routing
for variables, or whether they are all done by hand (you can't sort them
alphabetically!).  In fact, I surmise that the internal numeric identifiers
in the GCVS are primarily a sort code field, hence the padded zeros.  And,
yes, you do need to sort variables "properly" from time to time... ...you
get in the habit of knowing the order, and on using an index etc invariably
tend to look for something like Per AX _after_ Per R, and not before as you
would if being purely alphabetic (well I do anyway).

>My own observation
>database does this, I implemented a function to scan the vsnet-obs mails,
>and it works for more cases than the above mentioned. E.g. novaes: "N1999#2
>Aql" = "Aql Nova 1999#2" = "AqlV1494" = "V1494 Aql" ...

That's okay if you really do have a limited number of variations and you
remember to cover for them all, and your database isn't too large with
respect to objects.  However, if something gets missed because you haven't
allowed for it, due to the nature of automation, you may never notice.  N
Aql 92...


OT: Where does this USA English usage of '#' to denote a number come from?
Is it from the USA, or from some other origin?  Cannot people in the USA
recognise arabic numerals unless they have a # in front of them? If not,
what do they think 34 means, and how does it differ from #34?????

John

VSNET Home Page

Return to Daisaku Nogami


vsnet-adm@kusastro.kyoto-u.ac.jp