Re: Holding connections open: an immodest proposal

Martijn Koster (
Thu, 15 Sep 1994 15:07:49 +0200

> In article <> you write:
> I was proposing that it be valid for all protocols, not just http. We could
> however use
> /hallam Non Recursive directory
> /hallam/* All the objects in the directory
> /hallam/** directory tree
> /hallam/*** All the objects in the directory tree
> So GET /hallam/* would return a multipart with all the objects in the
> directory in it.

It is rather unclear to me is what you mean by "in the directory
tree", especially in the context of "all protocols". What are the
semantics when "/hallam" is not a simple directory, but dynamically
generated by a script or so? Do you fail, or do you run a local robot
and pass it all the HTTP headers you got from the client? The latter
would be neat, but an enouremous overhead.

I'm don't think I like these special URL encodings -- why not specify
a normal URL as part of new methods? That wouldn't break existing
servers/clients, and save you from the uri hornets nest. For example:

GET /hallam Non Recursive directory
LIST /hallam All the objects in the directory
RLIST /hallam directory tree
RGET /hallam All the objects in the directory tree

> This sort of thing would not be too usefull for clients but very nice for
> other applications.

Yeah, like robots and mirrors -- will people really allow this on
their servers, even if you recursively check modification dates
against an If-modified-since? I certainly won't without out of band
bilatreal agreement with clients.

-- Martijn
X-400: C=GB; A= ; P=Nexor; O=Nexor; S=koster; I=M
X-500: c=GB@o=NEXOR Ltd@cn=Martijn Koster