[fitsbits] FITS 64-bit Integer Public Comment Period

This is a reminder that the Public Comment Period on the proposal to add
support for 64-bit integers to the FITS data format is currently open here
on the FITSBITS mail list. Details of this proposal are at
http://fits.gsfc.nasa.gov/fits_64bit.html

The regional FITS committees will begin voting soon on whether to accept
this proposal, so it would be helpful to them to see more comments here on
FITSBITS (or the sci.astro.fits newsgroup) in order to judge the level of
community support for this major change to the FITS Standard. Note that the
last time a new data type was added to FITS was back in 1990 when the 32-bit
and 64-bit floating point data types were introduced.

In particular, it would helpful to hear your opinion on the following issue
that was raised in the previous discussions:

Are you in favor of adding support in FITS for BOTH 64-bit integer images
and 64-bit integer table columns? Should the FITS committees be allowed to
vote on each of these 2 cases separately (and perhaps approve one without
the other), or are these 2 cases so closely linked that they should only be
considered together?

Bill Pence
--
__________________________________________________ __________________
Dr. William Pence
NASA/GSFC Code 662 HEASARC +1-301-286-4599 (voice)
Greenbelt MD 20771 +1-301-286-1684 (fax)

On Wed, Jun 15, 2005 at 03:51:46PM -0400, William Pence wrote:

Are you in favor of adding support in FITS for BOTH 64-bit integer images
and 64-bit integer table columns? Should the FITS committees be allowed to
vote on each of these 2 cases separately (and perhaps approve one without
the other), or are these 2 cases so closely linked that they should only be
considered together?

I am in favor of allowing 64-bit quantities in both images and tables
because in my experience it is common to need to be able to convert one
to the other.

I believe the questions should be whether to allow 64-bit data quantities
and to allow 64-bit pointer offsets, but that if you allow 64-bit data
quantities you should allow them everywhere. If they are allowed in one
place, then there is no implementation advantage in not allowing them
everywhere.

-Mike

------
From: Mike Nolan
Mike Nolan +1 787 878 2612 Fax: +1 787 878 1861
Arecibo Observatory, HC3 Box 93995, Arecibo, Puerto Rico 00612

William Pence wrote:
This is a reminder that the Public Comment Period on the proposal to add
support for 64-bit integers to the FITS data format is currently open
here on the FITSBITS mail list. Details of this proposal are at
http://fits.gsfc.nasa.gov/fits_64bit.html

The regional FITS committees will begin voting soon on whether to accept
this proposal, so it would be helpful to them to see more comments here
on FITSBITS (or the sci.astro.fits newsgroup) in order to judge the
level of community support for this major change to the FITS Standard.
Note that the last time a new data type was added to FITS was back in
1990 when the 32-bit and 64-bit floating point data types were introduced.

In particular, it would helpful to hear your opinion on the following
issue that was raised in the previous discussions:

Are you in favor of adding support in FITS for BOTH 64-bit integer
images and 64-bit integer table columns? Should the FITS committees be
allowed to vote on each of these 2 cases separately (and perhaps approve
one without the other), or are these 2 cases so closely linked that they
should only be considered together?

Bill Pence

I had thought you wanted change-requests rather than votes;

Yes for both, consider together.

Peter.

On Wed, Jun 15, 2005 at 03:51:46PM -0400, William Pence wrote:

Are you in favor of adding support in FITS for BOTH 64-bit integer images
and 64-bit integer table columns? Should the FITS committees be allowed to
vote on each of these 2 cases separately (and perhaps approve one without
the other), or are these 2 cases so closely linked that they should only be
considered together?

I advocate three separate votes (BITPIX 64, TFORM nK and TFORM 1Qt).

Still lacking of a strong *intrinsic* reason to support BITPIX 64.

The Q pointer vote should also include the signed/unsigned stuff I'll
post about separately.


Also I'm somewhat worried by the limited number of responses (to the
public comment solicitation and in general. I solicited opinions in my
national community and got so far none). Too few FITS "developers"
around ?

--
----------------------------------------------------------------------
is a newsreading account used by more persons to
avoid unwanted spam. Any mail returning to this address will be rejected.
Users can disclose their e-mail address in the article if they wish so.

On Wed 2005-06-15T15:51:46 -0400, William Pence hath writ:
Are you in favor of adding support in FITS for BOTH 64-bit integer images
and 64-bit integer table columns? Should the FITS committees be allowed to
vote on each of these 2 cases separately (and perhaps approve one without
the other), or are these 2 cases so closely linked that they should only be
considered together?

Yes to the first.

Caution, the second question has two basically opposing clauses.

No, I think that separate voting is a bad idea because, yes, I think
they should only be considered together.
But I would be willing to compromise on that for political expediency.

--
Steve Allen WGS-84 (GPS)
UCO/Lick Observatory Natural Sciences II, Room 165 Lat +36.99858
University of California Voice: +1 831 459 3046 Lng -122.06014
Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m