Re: the case for machine-readable error fields

On Tue, Aug 04, 2009 at 09:16:23PM -0400, Robert Haas wrote:
> On Tue, Aug 4, 2009 at 8:36 PM, Joshua D. Drake<jd(at)commandprompt(dot)com> wrote:
> >
> >> So, we are just trying to whip into shape explain diagnostics which are
> >> in JSON or XML, and now you want us to exclude XML from this one because
> >> you don't like it? Can we please try for some consistency?
> >>
> >> Sorry to break it to you, but there are plenty of people and businesses
> >> who want XML. And I certainly don't want to have to master every data
> >> representation model out there. XML has far more traction than anything
> >> else that's comparable in my experience.
> >>
> >> The fact that Greg is prepared to suggest CSV, with its obvious serious
> >> deficiencies, as being *better* than XML, makes his whole argument
> >> highly suspect IMNSHO.
> >
> > From a business perspective, XML is the only viable option for output.
>
> Wow, I feel like it's time for a bench-clearing brawl!
>
> My serialization format kicks your serialization format's butt!
>
> This doesn't have a whole lot to do with the original topic of this
> thread, which unless I missed something had only to do with extending
> the FE/BE protocol, but it definitely makes for lively conversation.
> Anyone want to vote ASN.1 for world domination? Can we set up some
> kind of cage match between the dueling standards?

I love standards. There are so many to choose from :)

Cheers,
David.
--
David Fetter <david(at)fetter(dot)org> http://fetter.org/
Phone: +1 415 235 3778 AIM: dfetter666 Yahoo!: dfetter
Skype: davidfetter XMPP: david(dot)fetter(at)gmail(dot)com

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate