Re: Postel's law

On May 28, 9:55 am, Marshall <marshall.spi..._at_gmail.com> wrote:
> On May 28, 5:17 am, Ed Prochak <edproc..._at_gmail.com> wrote:
>
>
>
> > On May 23, 1:28 am, David BL <davi..._at_iinet.net.au> wrote:
>
> > > Have you heard of Postel's Law?
>
> > > "Be liberal in what you accept, and conservative in what you
> > > send."
>
> > > I can imagine it being applied to many things. Eg file formats, APIs,
> > > compilers, databases ...
>
> > > I think it generally leads to unnecessary complexity and sweep errors
> > > under the carpet.
>
> > > Comments?
>
> > I always understood this in the context of standards (ad hoc and
> > otherwise) to mean essentially backward compatibility.
>
> Well, no, not really. Postel's law is applicable to protocols
> that have only a single version.
>
> > The discussion about HTML is interesting, but I see problems with HTML
> > as less of an issue as the compatibility problem of all the other
> > browser issues with JAVA, Java script, and uncounted plug-ins. As
> > broken as it is, HTML still mostly works.
>
> I think what we're talking about here, though, is the difference
> between "mostly works" and "works." Did you read the link
> Gene supplied?

I did. I just thought the discussion was too limited to HTML. Especially since this is a group on database theory, not document format theory.
>
> On May 23, 10:21 pm, Gene Wirchenko <ge..._at_ocis.net> wrote:
> > Joel Spolsky said it well:
> > http://www.joelonsoftware.com/items/2008/03/17.html

Good article.
 Should we conclude this was a result of the law of unintended consequences?
 Or is HTML development an example of misapplying a rule (Postel's Law)?
Something for comp.risks?

  Ed Received on Thu May 29 2008 - 13:33:48 CEST