headers
Ian Hickson | 4 Sep 2009 22:19
Picon

Re: [hybi] ws: and wss: schemes

On Fri, 4 Sep 2009, Julian Reschke wrote:
> Ian Hickson wrote:
> > >
> > > Because that's how URI and thus URLs are defined.
> > 
> > The ws: and wss: URLs are IRIs; why would we limit them to URIs? I'm 
> > not especially interested in ASCII-only URIs at this point. These URLs 
> > are only ever going to be used in contexts that accept full IRIs.
> 
> But that's not who registering an URI scheme works. Check the relevant 
> RFCs. Essentially you register the *URI* scheme, and get IRIs based on 
> the mapping rules defined in RFC 3987.

That's what I thought, but then I got feedback saying I had to register an 
IRI scheme if I wanted to use IRIs.

I've no interest in making ws: and wss: URIs. Only IRIs.

If I define the syntax to be a subset of the full URI syntax, how does it 
ever get extended to be a subset of the full IRI syntax?

What should I put in the spec to make you happy and to make the use of ws: 
and wss: IRIs fully well-defined?

> > > > > Furthermore, it still doesn't answer what the semantics of these parts
> > > > > are. What do "ihier-part" and "iquery" represent in a ws URI?
> > > > This is defined by the RFC 3987, no? Surely we wouldn't want IRI
> > > > components to have different meanings in different schemes?
> > > If you can point to a section in RFC 3987 which defines more than the
> > > syntax, and can state that that also applies to "ws", then, great...
(Continue reading)

Permalink | Reply |
headers
Julian Reschke | 5 Sep 2009 08:43
Picon
Picon

Re: [hybi] ws: and wss: schemes

Ian Hickson wrote:
> ...
> That's what I thought, but then I got feedback saying I had to register an 
> IRI scheme if I wanted to use IRIs.
> 
> I've no interest in making ws: and wss: URIs. Only IRIs.
> 
> If I define the syntax to be a subset of the full URI syntax, how does it 
> ever get extended to be a subset of the full IRI syntax?
> 
> What should I put in the spec to make you happy and to make the use of ws: 
> and wss: IRIs fully well-defined?
> ...

The point is not to make me happy, but to do the right thing.

Just define a URI scheme; use of ws IRIs will be defined automatically 
in terms of RFC 3987 (IRI experts, please correct me if I'm wrong).

> ...
>> The RFC reference is immutable. Just paste the content in your source 
>> file, and change the anchor attribute value.
> 
> My source file is an HTML document, so I don't think that would work well.

The source file that you feed into xml2rfc is an XML file using the 
RFC2629bis syntax. You control that file. Put into it what you need.

> ...
> I've read this, but as far as I can tell, "Always UTF-8" and "See IRI" are
(Continue reading)

Permalink | Reply |
headers
Martin J. Dürst | 8 Sep 2009 12:16
Picon
Gravatar

Re: [hybi] ws: and wss: schemes


On 2009/09/05 15:43, Julian Reschke wrote:
> Ian Hickson wrote:

>> I've read this, but as far as I can tell, "Always UTF-8" and "See IRI"
>> are both complete and accurate ways of addressing this.

Ian, you are famous for pointing out your low trust in implementors. 
Given that, I'd tend to spend a few more words on this, to help them 
along a bit better.

>> Since apparently neither of these options satisfies you, could you
>> state exactly what literal text would satisfy you?
>> ...
>
> I already pointed you to RFC 5092 as relatively recent example, see
> <http://tools.ietf.org/html/rfc5092#section-8>.

A newer example (the RFC numbers are confusing, because RFC 5092 goes 
back to RFC 2192) would be the XMPP IRI/URI spec, see 
http://tools.ietf.org/html/rfc4622.

Also, please note that there may be nothing like an "IRI only" scheme. 
There is always the chance that in some strange context, a conversion to 
ASCII (i.e. to URIs) happens.

Regards,   Martin.

--

-- 
#-# Martin J. Dürst, Professor, Aoyama Gakuin University
(Continue reading)

Permalink | Reply |
headers
Ian Hickson | 5 Sep 2009 12:46
Picon

Re: [hybi] ws: and wss: schemes

On Sat, 5 Sep 2009, Julian Reschke wrote:
> Ian Hickson wrote:
> >
> > That's what I thought, but then I got feedback saying I had to 
> > register an IRI scheme if I wanted to use IRIs.
> > 
> > I've no interest in making ws: and wss: URIs. Only IRIs.
> > 
> > If I define the syntax to be a subset of the full URI syntax, how does 
> > it ever get extended to be a subset of the full IRI syntax?
> > 
> > What should I put in the spec to make you happy and to make the use of 
> > ws: and wss: IRIs fully well-defined?
> 
> The point is not to make me happy, but to do the right thing.

My idea of "the right thing" and your idea of "the right thing" have very 
little in common, so really, for me, it boils down to making you happy.

> Just define a URI scheme; use of ws IRIs will be defined automatically 
> in terms of RFC 3987 (IRI experts, please correct me if I'm wrong).

That's what I thought I had done; since you disagree that I had done that, 
please provide the copy that you believe would do the job, so that we can 
stop dancing around playing "warmer/colder" with this text.

> > > The RFC reference is immutable. Just paste the content in your 
> > > source file, and change the anchor attribute value.
> > 
> > My source file is an HTML document, so I don't think that would work
(Continue reading)

Permalink | Reply |
headers
Kristof Zelechovski | 5 Sep 2009 21:53
Picon
Gravatar

RE: [Uri-review] [hybi] ws: and wss: schemes

Ian, you are already going out of your way to make HTML back into XML.  You
are swimming against the current.  A noble exercise, I must admit, but
hardly worth the effort in this case :-)
Cheers,
Chris

-----Original Message-----
From: uri-review-bounces <at> ietf.org [mailto:uri-review-bounces <at> ietf.org] On
Behalf Of Ian Hickson
Sent: Saturday, September 05, 2009 12:46 PM
To: URI; hybi <at> ietf.org; uri-review <at> ietf.org; public-i18n-core <at> w3.org
Subject: Re: [Uri-review] [hybi] ws: and wss: schemes

> The source file that you feed into xml2rfc is an XML file using the 
> RFC2629bis syntax. You control that file. Put into it what you need.

There's no such file; the XML is generated by a script and posted straight 
to the xml2rfc Web service. I have every intention of keeping this as 
automatic as possible; I already have to go out of my way to make the 
references to [WEBADDRESSES] and [HTML5] work, I really don't want to have 
to start doing the same for IETF documents when I don't actually have to.
Permalink | Reply |
headers
Julian Reschke | 5 Sep 2009 13:08
Picon
Picon

Re: [hybi] ws: and wss: schemes

Ian Hickson wrote:
> ...
>> Just define a URI scheme; use of ws IRIs will be defined automatically 
>> in terms of RFC 3987 (IRI experts, please correct me if I'm wrong).
> 
> That's what I thought I had done; since you disagree that I had done that, 
> please provide the copy that you believe would do the job, so that we can 
> stop dancing around playing "warmer/colder" with this text.
> ...

I'm not going to write the spec for you.

A definition of a URI describes it's ASCII-based syntax, it's purpose, 
and how it's used in the protocol. I think that's pretty clear from RFC 
4395.

 > ...
>> The source file that you feed into xml2rfc is an XML file using the 
>> RFC2629bis syntax. You control that file. Put into it what you need.
> 
> There's no such file; the XML is generated by a script and posted straight 
> to the xml2rfc Web service. I have every intention of keeping this as 
> automatic as possible; I already have to go out of my way to make the 
> references to [WEBADDRESSES] and [HTML5] work, I really don't want to have 
> to start doing the same for IETF documents when I don't actually have to.

Well, you asked how to do it, and I explained how it can be done. 
Apparently you're not happy with the answer, but that's all I can do 
from here.
(Continue reading)

Permalink | Reply |
headers
Joseph A Holsten | 5 Sep 2009 01:51
Gravatar

Re: [hybi] ws: and wss: schemes


On Sep 4, 2009, at 3:19 PM, Ian Hickson wrote:

> On Fri, 4 Sep 2009, Julian Reschke wrote:
>> Ian Hickson wrote:
>>>>
>>>> Because that's how URI and thus URLs are defined.
>>>
>>> The ws: and wss: URLs are IRIs; why would we limit them to URIs? I'm
>>> not especially interested in ASCII-only URIs at this point. These  
>>> URLs
>>> are only ever going to be used in contexts that accept full IRIs.
>>
>> But that's not who registering an URI scheme works. Check the  
>> relevant
>> RFCs. Essentially you register the *URI* scheme, and get IRIs based  
>> on
>> the mapping rules defined in RFC 3987.
>
> That's what I thought, but then I got feedback saying I had to  
> register an
> IRI scheme if I wanted to use IRIs.
>
> I've no interest in making ws: and wss: URIs. Only IRIs.
>
> If I define the syntax to be a subset of the full URI syntax, how  
> does it
> ever get extended to be a subset of the full IRI syntax?
>
> What should I put in the spec to make you happy and to make the use
(Continue reading)

Permalink | Reply |
headers
Julian Reschke | 5 Sep 2009 08:47
Picon
Picon

Re: [Uri-review] [hybi] ws: and wss: schemes

Joseph A Holsten wrote:
> ...
> The only scheme I can think of that was defined as an IRI was XMPP 
> [RFC4622]. It actually makes more sense when you start with IRIs. If 
> that's what you need, please just do that.
> ...

Actually, that RFC *registers* a URI scheme; see 
<http://tools.ietf.org/html/rfc4622#section-3>.

> ...

BR, Julian
Permalink | Reply |
headers
Joseph A Holsten | 5 Sep 2009 09:56
Gravatar

Re: [hybi] ws: and wss: schemes

Julian Reschke supposedly wrote:
> Joseph A Holsten wrote:
>> ...
>> The only scheme I can think of that was defined as an IRI was XMPP  
>> [RFC4622]. It actually makes more sense when you start with IRIs.  
>> If that's what you need, please just do that.
>> ...
>
> Actually, that RFC *registers* a URI scheme; see <http://tools.ietf.org/html/rfc4622#section-3 
> >.

It does, and so should websocket. But every other scheme RFC defines  
the URI first and foremost, then describes how to map IRIs. If  
websocket will be an IRI scheme first and foremost, defining it in  
terms of ihier-part and iquery makes sense. Then just adapt the text  
from RFC4622 sections 2 and 3.

The first sentence from <http://tools.ietf.org/html/rfc4622#section-2.8.1 
 > seems quite similar to Ian's preference with websocket:
"If a processing application is presented with an XMPP URI and not  
with an XMPP IRI, it MUST first convert the URI into an IRI by  
following the procedure specified in Section 3.2 of [IRI]."
--
Joseph Holsten
Permalink | Reply |
headers
Julian Reschke | 5 Sep 2009 10:34
Picon
Picon

Re: [Uri-review] [hybi] ws: and wss: schemes

Joseph A Holsten wrote:
> Julian Reschke supposedly wrote:
>> Joseph A Holsten wrote:
>>> ...
>>> The only scheme I can think of that was defined as an IRI was XMPP 
>>> [RFC4622]. It actually makes more sense when you start with IRIs. If 
>>> that's what you need, please just do that.
>>> ...
>>
>> Actually, that RFC *registers* a URI scheme; see 
>> <http://tools.ietf.org/html/rfc4622#section-3>.
> 
> It does, and so should websocket. But every other scheme RFC defines the 

Yes.

> URI first and foremost, then describes how to map IRIs. If websocket 
> will be an IRI scheme first and foremost, defining it in terms of 
> ihier-part and iquery makes sense. Then just adapt the text from RFC4622 
> sections 2 and 3.

But there's no registry for IRI scheme, as far as I can tell.

So you always define the URI scheme, and then, when needed, talk about 
how IRIs are mapped.

> The first sentence from 
> <http://tools.ietf.org/html/rfc4622#section-2.8.1> seems quite similar 
> to Ian's preference with websocket:
> "If a processing application is presented with an XMPP URI and not with
(Continue reading)

Permalink | Reply |

Gmane