Re: User and group databases (fwd)

["Roy T. Fielding" <fi...@avron.ICS.UCI.EDU>]

>     I sent some mail around to server authors last week suggesting that those
> attending the LA IETF in March could get together a CGI BOF, with the
> possible outcome of starting a working group on server-side interfaces and
> metainformation representations.  I have had some response from other 
> public-domain server authors, but nothing from the commercial server 
> developers.  Surprisingly not many want to make their servers easy to 
> migrate away from.  :)  I will be in Cambridge Sunday through Tuesday 
> attending a W3C workshop, can we get together sometime then?

Just a word edgewise ....  although it would be nice to meet at an IETF
(since I'll be there anyway), the IETF is not an appropriate forum for
specification of API's.  What the IETF does is specify protocols on
the wire -- it avoids APIs like the plague (with one exception, which
indeed turned into the plague).

You should get this reaction if you ask the APP A.D. (Harald Alvestrand)
to schedule a CGI BOF at the LA IETF.  I suggest instead that you try to
get W3C to sponsor server-side standards -- now that they have finally
hired a server person (Anselm Baird-Smith <ab...@w3.org>), it is possible
for them to actually do what they are supposed to be doing in this area.

.....Roy

Re: User and group databases (fwd)

[Brian Behlendorf <br...@organic.com>]

On Fri, 26 Jan 1996, Roy T. Fielding wrote:
> >     I sent some mail around to server authors last week suggesting that those
> > attending the LA IETF in March could get together a CGI BOF, with the
> > possible outcome of starting a working group on server-side interfaces and
> > metainformation representations.  I have had some response from other 
> > public-domain server authors, but nothing from the commercial server 
> > developers.  Surprisingly not many want to make their servers easy to 
> > migrate away from.  :)  I will be in Cambridge Sunday through Tuesday 
> > attending a W3C workshop, can we get together sometime then?
> 
> Just a word edgewise ....  although it would be nice to meet at an IETF
> (since I'll be there anyway), the IETF is not an appropriate forum for
> specification of API's.  What the IETF does is specify protocols on
> the wire -- it avoids APIs like the plague (with one exception, which
> indeed turned into the plague).
> 
> You should get this reaction if you ask the APP A.D. (Harald Alvestrand)
> to schedule a CGI BOF at the LA IETF.  I suggest instead that you try to
> get W3C to sponsor server-side standards -- now that they have finally
> hired a server person (Anselm Baird-Smith <ab...@w3.org>), it is possible
> for them to actually do what they are supposed to be doing in this area.

I have been talking with Harald, John Klinsen, and Larry Masinter, and they
agree that while IETF traditionally does not concern itself with protocols
not "over the wire", the IETF would still make an adequate forum for a CGI
BOF, in the absence of W3C support for CGI and other server API
standardization.  About three months ago David and Marc and I received mail
from Rohit saying W3C could not expend the resources to support a CGI
standardization effort.  I will try and make the case again out there this
weekend, but if W3C doesn't have the capacity to handle it (and they
certainly have lots of other issues to handle) then I think it should be
pursued elsewhere.  Unless, of course, there is a lack of desire amongst the
server developers for forging common API's - while many users and
server-add-on developers want it.  This does go beyond CGI.

Anyways, I'll see one last time if there's a way to address CGI and 
server-side standardization issues within W3C over the next few days.  
I'm genuinely hopeful, perhaps some resources could magically appear :)
I do agree that W3C would be a better forum, but if not there, where? ISO?
Market forces?

	Brian

--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/