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

Re: The triples datamodel -- was Re: SemanticWeb per

• To: John Cowan <cowan@c...>
• Subject: Re: The triples datamodel -- was Re: SemanticWeb permathread, iteration n+1
• From: Henrik Martensson <henrik.martensson@b...>
• Date: Thu, 10 Jun 2004 22:54:00 +0200
• Cc: XML Developer List <xml-dev@l...>
• In-reply-to: <20040608105635.GB4386@c...>
• References: <40C31EFC.2090203@k...> <p06010202bce8dc0073d2@[192.168.254.88]> <40C33FD3.3050308@c...> <40C35952.1080209@k...> <40C36038.6020900@c...> <1086642052.3611.17.camel@l...> <40C52D2F.7080005@c...> <40C5411D.5050905@z...> <891ACD72-B90B-11D8-B347-000A95CCC59E@x...> <1086676833.3719.154.camel@l...> <20040608105635.GB4386@c...>

On Tue, 2004-06-08 at 12:56, John Cowan wrote:
> Henrik Martensson scripsit:
> 
> > For the record, if two systems exchange information, and changes in one
> > system mean that the exchange format has to change, the systems are
> > _not_ loosely coupled. There has been a tendency in this thread to
> > equate "loose coupling" with "ignoring changes in couplings", but this
> > is wrong. Loosely coupled systems must still play by the rules they have
> > agreed on.
> 
> In a really loosely coupled system, like Walter's, the "sender" has no
> clue that the "receiver" even exists, still less is there any agreement
> between them.  The sender has published something, and the receiver must
> make do as best he can.

There is still a contract. If I publish an interface to a service, then
I must stick to that interface. If I want people to continue using my
service, I have better not break backwards compatibility too often.

With XML documents, as with software APIs, you can often add new things
without breaking backwards compatibility. Then again, just as with
software APIs, quite frequently you can't.

/Henrik

• References:
  • Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
    • From: Robert Koberg <rob@k...>
  • Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
    • From: Elliotte Rusty Harold <elharo@m...>
  • Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
    • From: "Thomas B. Passin" <tpassin@c...>
  • Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
    • From: Robert Koberg <rob@k...>
  • Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
    • From: "Thomas B. Passin" <tpassin@c...>
  • Re: The triples datamodel -- was Re: SemanticWeb permathread, iteration n+1
    • From: Henrik Martensson <henrik.martensson@b...>
  • Re: The triples datamodel -- was Re: SemanticWeb permathread, iteration n+1
    • From: "Thomas B. Passin" <tpassin@c...>
  • Re: The triples datamodel -- was Re: SemanticWeb permathread, iteration n+1
    • From: Rick Marshall <rjm@z...>
  • Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
    • From: Michael Champion <mc@x...>
  • Re: The triples datamodel -- was Re: SemanticWeb permathread, iteration n+1
    • From: Henrik Martensson <henrik.martensson@b...>
  • Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
    • From: John Cowan <cowan@c...>

• Prev by Date: Re: The triples datamodel -- was Re: SemanticWeb permathread, iteration n+1
• Next by Date: RE: Meta-somethingorother (was the semantic web mega-permathread thing)
• Previous by thread: Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
• Next by thread: Re: The triples datamodel -- was Re: Semantic Web permathread, iteration n+1
• Index(es):
  • Date
  • Thread