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

Re: Elliotte Rusty Harold on Web Services

• To: xml-dev@l...
• Subject: Re: Elliotte Rusty Harold on Web Services
• From: Mike Champion <mc@x...>
• Date: Fri, 31 Jan 2003 13:49:55 -0500
• In-reply-to: <20030131125153.G2173@w...>
• References: <003801c2c946$e5b34520$6401a8c0@cavnarjohnson><3E3AAB07.4080403@d...> <20030131125153.G2173@w...>
• User-agent: Opera7.0/Win32 M2 build 2637

On Fri, 31 Jan 2003 12:51:53 -0500, Mark Baker <distobj@a...> wrote:


> In the former, the data that is sent is just data, not a request for
> action.  POST is the action.  If it were sent with HTTP PUT, it would
> mean something entirely different.

Back to our canonical disagreement :-)  My position is that the "interface" 
is the totality of the contract between producer and consumer; the 
difference between "just data" and "request for action" is pretty fuzzy at 
best. (LISP being the usual example, I guess).  The "action" is whatever 
happens on the server, however that got invoked.  Call it a "method", call 
it a "request", call it a "suggestion" ... some code will parse the "just 
data" and decide to do something.

Disallow POST and insist that all GETs be safe and PUTs idempotent, and I 
would agree that REST is less of a security hole, but it's gonna be hard to 
do application integration that way.

• References:
  • RE: Elliotte Rusty Harold on Web Services
    • From: "Cavnar-Johnson, John" <JCavnar-Johnson@s...>
  • Re: Elliotte Rusty Harold on Web Services
    • From: Rich Salz <rsalz@d...>
  • Re: Elliotte Rusty Harold on Web Services
    • From: Mark Baker <distobj@a...>

• Prev by Date: RE: malfunctioning, evil adult as XML
• Next by Date: Are There URIs in Heaven?
• Previous by thread: RE: Elliotte Rusty Harold on Web Services
• Next by thread: Re: Elliotte Rusty Harold on Web Services
• Index(es):
  • Date
  • Thread