Fw: Re: slide of my talk

From: Yuji SHIRASAKI <yuji.shirasaki-at-nao.ac.jp>
Date: Sun, 14 May 2006 10:43:24 -0700


I forgot to send to the voql list.

Forwarded by Yuji SHIRASAKI <yuji.shirasaki-at-nao.ac.jp>

----------------------- Original Message -----------------------
 From:    Yuji SHIRASAKI <yuji.shirasaki-at-nao.ac.jp>
 To:      Arnold Rots <arots-at-head.cfa.harvard.edu>
 Date: Sun, 14 May 2006 09:17:44 -0700  Subject: Re: slide of my talk

Dear Arnold,

Thank you your comments,

My suggestion on the coordinate frame is

  1. default coordinate frame should be defined on each service, and it should be exposed by metadata interface.
  2. recommend to use "FK5" or "ICRS"? as a default if there is no strong reason to use another frame.
  3. user need to specify the coordinate frame, if he/she does not know which frame is a default for the table in advance
  4. user does not need to specify the frame, if he/she know which frame is a default for the table.

You are saying that to include the frame in a query is a very small effort, but it is a little verbose for a user who know which frame is a defualt as he/she uses the table very frenquently.

And it is possible to inform at the user interface level which frame is a default, so user will no be confused if he/she uses the smart user interface.

In case of the simulation data, data might be queried based on the cartesian coordinate. The query might be someting like this:

        Select *
        From   table
        Where  Region('CIRCLE 0 0 10')

This is a query searching data on XY plane withradius of 10 (arbitrary unit). In this case specifying the frame is almost meaningless.

Yuji Shirasaki

> Yuji,
>
> I haven't had time to go through your entire talk, but there is one
> point that I would like to raise.
>
> You are suggesting that the coordinate frame can be omitted under
> certain consitions.
> I think that is a bad idea, for the same reasons why you were arguing
> that a table alias should always be specified: users get confused when
> it is sometimes needed, sometimes not. And it is a very small effort
> to just include it to begin with, especially as I suspect most users
> will not fashion the queries themselves, but through an interface. I
> have no problems with the interface presenting the user with defaults,
> but it should include the frame explicitly in the query.
>
> The larger picture is that we hope to include more than galactic and
> extra-galactic astronomers to the VO. In the past week we have had
> discussions with solar people on coordintes. Invariably, if you allow
> defaults in queries, people are going to assume that the default is
> what *they* think is the most "natural" choice. That has the
> potential for large-scale confusion and it is much better to state
> from the beginning that a frame needs to be specified.
>
> - Arnold
>
> Yuji SHIRASAKI wrote:
> >
> > Dear VOQLers,
> >
> > I have uploaded a slide that will be presented at the
> > VOQL meeting. Please look at the following URL:
> >
> > http://www.ivoa.net/internal/IVOA/InterOpMay2006VOQL/VOQL-YujiShirasaki-ver1.pdf
> >
> > It is still under preparation, but I wellcome your comments and/or
> > questions on the contents, especially from those who cannot attend
> > the meeting.
> >
> > It is worthwhile to start the discussion even before the meeting.
> >
> > I also uploaded update versions of ADQL and ADQL-Core schema.
> >
> > See you soon.
> >
> > Yuji Shirasaki.
> >
> --------------------------------------------------------------------------
> Arnold H. Rots Chandra X-ray Science Center
> Smithsonian Astrophysical Observatory tel: +1 617 496 7701
> 60 Garden Street, MS 67 fax: +1 617 495 7356
> Cambridge, MA 02138 arots-at-head.cfa.harvard.edu
> USA http://hea-www.harvard.edu/~arots/
> --------------------------------------------------------------------------
>
>
> __________ NOD32 1.1537 (20060514) Information __________
>
> This message was checked by NOD32 antivirus system.
> http://canon-sol.jp/product/nd
>
>
>
>
> __________ NOD32 1.1537 (20060514) Information __________
>
> This message was checked by NOD32 antivirus system.
> http://canon-sol.jp/product/nd
>

-- 
Yuji SHIRASAKI <yuji.shirasaki-at-nao.ac.jp>




__________ NOD32 1.1537 (20060514) Information __________

This message was checked by NOD32 antivirus system.
http://canon-sol.jp/product/nd




__________ NOD32 1.1537 (20060514) Information __________

This message was checked by NOD32 antivirus system.
http://canon-sol.jp/product/nd


--------------------- Original Message Ends --------------------

-- 
Yuji SHIRASAKI <yuji.shirasaki-at-nao.ac.jp>
Received on 2006-05-14Z19:44:03