Why do we write standards?

[Joshua E. Smith]

>The standard must be easily implemented.
>This, I think is the reason IP and HTTP were accepted: they were relatively
>simple protocols to implement.

Nonsense.  IP never existed on it's own, IP was part of TCP/IP from the
outset, and TCP/IP is a pretty complicated beast.  It took years to get it
tuned right.  HTTP isn't all that easy either: it takes a few hundred lines
of hell to make a robust client that handles the myriad error conditions
and all the versions.  I've never written an HTTP server, but I can't
imagine it's all that trivial either.

No, the reason these were accepted was that good, working source code was
made freely available to all comers, and they met a very real need.  It
also helped that the IETF's policy is that you cannot make it up the RFC
food chain until there are two separate interoperable implementations of
the standard.  (If W3C required just one working implementation it would be
a quantum improvement of the process.  It's a great way to fight the
natural tendency for bloat.)

(Speaking of which, I just learned that the math which drives MP3 encoders
is patented!  You can get the source code for MP3 from ISO, but you have to
agree not to compile it!  How nuts is that?  It's GIF all over again!)

-Joshua Smith


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)