<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><div>Hi DAL,</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="gmail-">> > sec 4.3.1:<br>
> > A reserved namespace is proposed:<br>
> ><br>
> > "The ivo prefix is reserved for functions that have been<br>
> > defined in an IVOA specification."<br>
> ><br>
> > I think that's not a bad idea, but I believe that in<br>
> > at least some cases current practice violates it; the DaCHS<br>
> > service at <a href="http://dc.g-vo.org/tap" rel="noreferrer" target="_blank">http://dc.g-vo.org/tap</a> defines e.g., ivo_healpix_center,<br>
> > ivo_interval_overlaps, ivo_healpix_index, ivo_apply_pm.<br>
> > I've heard Markus defend that practice elsewhere, so I guess<br>
> > he should comment.<br>
><br>
> Well... As usual we don't really have a process for how to phase in<br>
> these things, and "IVOA specification" isn't a well defined term<br>
> either (WD? PR? REC? Note?).<br>
<br>
</span>I don't think this would be hard to do, if we wanted ivo_ to mean<br>
"defined in an IVOA specification" as per the current text.<br>
The rule in practice could be that if the intention is for<br>
a given UDF to end up in a REC, then it can have the ivo_ prefix.<br>
So ivo_ is fair game for inclusion in any standards-track<br>
document (WD, PR, REC, or pre-WD, or sketch on a napkin of what<br>
a WD might look like one day), and not otherwise.<br>
It might result in some ivo_* UDFs getting provisionally defined<br>
and then later retracted or changed, but I don't think that's<br>
a disaster.<br>
<span class="gmail-"><br>
> The way I'd like this to work is: If two or more implementors agree<br>
> on a function pattern (which was the case for the healpix functions,<br>
> but admittedly not for apply_pm), they'd use ivo_ and ideally put up<br>
> their proposal for discussion here.<br>
<br>
</span>I mildly prefer the in-a-standards-track-document rule I've outlined<br>
above, but I don't have very strong views.<br></blockquote><div><br></div><div>I'm replying with the simple idea of reminding that</div><div><br></div><div><a href="http://wiki.ivoa.net/twiki/bin/view/IVOA/AdqlFunctionPrefixes">http://wiki.ivoa.net/twiki/bin/view/IVOA/AdqlFunctionPrefixes</a><br></div><div><br></div><div>gives some hints on the startup of this process.</div><div><br></div><div>I agree that simply saying "specification" is a bit unclear.</div><div>If we keep the above wiki page clean, a reference to it</div><div>may be a (temporary) solution.</div><div><br></div><div>Letting ivo_ prefix available to implementors agreeing</div><div>each other doesn't feel right to me, it sort of bypasses</div><div>the working group collaboration or discussion.</div><div><br></div><div>Cheers,</div><div> Marco</div></div></div></div>