Re: [apps-discuss] [Uri-review] Provisional URI registrations

----- Original Message -----
From: "Graham Klyne" <GK-lists@ninebynine.org>
To: "Michael Wojcik" <Michael.Wojcik@microfocus.com>
Cc: <uri-review@ietf.org>; "Înes" <ah@TR-Sys.de>; <apps-discuss@ietf.org>;
<Alfred H@core3.amsl.com>
Sent: Tuesday, November 23, 2010 4:45 PM

Michael, thanks for the clarification.  I clearly skimmed the discussion too
quickly - I didn't pick up that the scheme was lacking a definition.  I think my
latest response to Alfred applies.

<tp>
Who needs a definition?  I did a quick 'literature search' and found plenty of
references to tn3270 as a URI but no definition; the examples were all of the
form
scheme :// host : port
No parameters, queries, fragments or anything fancy.  On reflection, this seems
right; tn3270 is more sophisticated than many IETF protocols and if it wants
something, it negotiates it, so no need for a complex URI to specify things in
advance.

One could write a universal URI definition for such schemes
scheme host port
means set up a connection of protocol 'scheme' to 'host', default port from
IANA, but, arguably, we already have all we need in RFC3986.

Tom Petch
</tp>

#g
--

Michael Wojcik wrote:
>> From: uri-review-bounces@ietf.org [mailto:uri-review-bounces@ietf.org]
>> On Behalf Of Alfred HÎnes
>> Sent: Monday, 22 November, 2010 15:35
>>
>> Graham Klyne wrote:
>>
>>> Having made a very quick review of that discussion (w.r.t. tn3270), I
>>> feel there is a consensus that the appropriate response to to say
>>> "that's fine, it has a definition, let's leave it there".
>> Well, that's been one of the points where opinions seem to differ.
>> I could not find a definition, only RFC 1738 "reserving the name for
>> future definition".  The users of TN3270 service so far have claimed
>> there would be a definition, but nobody has shown it so far, and some
>> indications have been reported of different, non-interoperable
>> definitions as well.  I don't want to take a position on that now.
>
> As far as I can tell, the situation is this:
>
> - There is no definition of the tn3270 scheme.
>
> - There are applications which interpret tn3270 URIs, by starting a client
that implements some flavor of TN3270.
>
> - There is no single definition of a TN3270 service, which is what the scheme
presumably refers to. There are definitions of various pieces of Classic TN3270,
definitions of "best practices" for Classic TN3270, and a definition of TN3270E.
>
> - There are various implementations of TN3270 service (client and server),
most or all of which are probably not fully compliant with the various
definitions.
>
> - Those implementations are generally interoperable, but sometimes with
caveats (eg reduced functionality).
>
> So I think it's fair to say that:
>
> 1) There is no definition of the tn3270 scheme.
>
> 2) While there's a working understanding of what specifications contribute to
a "standard" implementation of TN3270 service, there is no single definition of
that service. That makes defining the scheme more difficult.
>
> I'd also suggest that while a single definition of the TN3270 service, and a
definition and proper registration of the tn3270 scheme, are desirable, in
practice I don't believe they'd have much effect. IBM hasn't shown any
inclination to bring its client implementation into conformance with the extant
TN3270E specification (RFC 2355), for example.
>
> I don't have strong feelings about whether the "tn3270" scheme is labelled
"provisional" or "historic". Again, I don't think it will make any difference to
implementors, who will treat tn3270-scheme URIs however they like. (I'm an
implementor myself, but thus far I've only dealt with the server side, which
never sees such URIs, so it hasn't come up.)
>

_______________________________________________
apps-discuss mailing list
apps-discuss@ietf.org
https://www.ietf.org/mailman/listinfo/apps-discuss