[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: URIs harmful

• To: "Dare Obasanjo" <dareo@m...>,<xml-dev@l...>
• Subject: RE: URIs harmful
• From: "Danny Ayers" <danny666@v...>
• Date: Wed, 24 Jul 2002 22:00:17 +0200
• Importance: Normal
• In-reply-to: <8BD7226E07DDFF49AF5EF4030ACE0B7E06621E35@r...>

>I just want to say I completely disagree with your post. 

Ok, why?

The 
>reasoning is pretty suspect especially around the "insisting that 
>it should always be dereferenceable breaks the web - every site 
>must be available 24/7" part.  Websites being available 24/7 
>breaks the web??? On what planet? 

As you quoted, I said *insisting* on this [URLs always being dereferenceable] breaks the web. I don't see any flaw in that reasoning - if the system requirement is 100% on some parameter and you only have 99.9999% then that system doesn't meet the requirements, i.e. is broken. I don't think we can rely on every URL responding to a http method, on this planet at least.

Our designs (and specifications) should take this into account. The planned non-functionality of a non-dereferenceable URI is easy to work with in a systematic fashion; the non-functionality of a URL that *must* be dereferenceable asks us to deal with undefined situations. 

Cheers,
Danny.

• References:
  • RE: URIs harmful
    • From: "Dare Obasanjo" <dareo@m...>

• Prev by Date: [ANN] NetGlue Web-Service & Plug-in Software
• Next by Date: RE: URIs harmful
• Previous by thread: RE: URIs harmful
• Next by thread: RE: URIs harmful
• Index(es):
  • Date
  • Thread