This bounced on the first attempt.
On Tue, 6 May 2003, Patrick Wallace wrote:
> Date: Tue, 6 May 2003 08:28:20 +0100 (BST)
> From: Patrick Wallace <[log in to unmask]>
> To: Tim Jenness <[log in to unmask]>
> Cc: Starlink development <[log in to unmask]>
> Subject: Re: TimeFrame proposals
>
> Tim,
>
>
> > I agree although I don't see a problem with using sla_DAT for the first
> > cut to get something working.
>
> I agree with that of course, as long as installing a proper operational
> solution doesn't become a "when I get round to it" sort of job.
>
> In the SOFA equivalent to sla_DAT I added a mechanism that looked for a
> suspiciously long time between the supplied date and the date of the
> last leap second that the routine knew about. That fell on its face
> once the present lull in leap seconds got under way - the last one was in
> 1999 and the next one will probably be in 2005. I changed it to a
> check against the release date of the routine, so enforcing periodic
> relinking.
>
>
> Patrick Wallace
> ____________________________________________________________________________
> Starlink/HMNAO Internet: [log in to unmask]
> Rutherford Appleton Laboratory Tel: +44-1235-445372
> Chilton, Didcot, Fax: +44-1235-446362
> Oxon OX11 0QX, UK
> ____________________________________________________________________________
>
>
|