RE: TS clarification (was: [rohc] The discussion on slope(s) (red ux))

[zhigang.c.liu at nokia.com]

Lars-Erik,

Just a few points/questions:

- Which "sophisticated programming language" did you use when implementing
RFC 3095?

- RE the hair-splitting on "/", it reflects the awkward attempt of changing
section 4.5.3 of 3095 without admitting it. And I guess you will have to 
redefine the meaning of "modulo" in section 4.5.3 as well? Of course, you
want call these as minor issues now ...

- Do you still remember the "direct appeal" you sent me in December 2004,
to identify "text parts of RFC that you say would have to be twisted
hard if there was no slope". Well, this is one of them.

- I wasn't alone on my opinion in March/April 2004. Pawel Szymanski and 
Kamal Deep Singh agreed with me. Not surprisingly, the discussion stopped
at that point.

- If I hadn't insisted on looking into the technical issues (especially 
with examples I provided), this discussion would have been buried 
sometime ago. It's obvious that some people are inventing new solutions
as they see my examples (e.g. the 0 slope for video). However, they never 
care to look at the existing solution I presented with the examples, 
which by-the-way has been discussed in detail during the development
of 3095 and which doesn't contradict 4.5.3. Why people do that? Don't ask me.

- Please read my updated proposal and give your technical comments.

BR, Zhigang

> -----Original Message-----
> From: ext Lars-Erik Jonsson (LU/EAB)
> [mailto:lars-erik.jonsson at ericsson.com]
> Sent: 24 January, 2005 03:13 AM
> To: Liu Zhigang.C (Nokia-NRC/Dallas); rohc at ietf.org
> Subject: RE: TS clarification (was: [rohc] The discussion on slope(s)
> (red ux))
> 
> 
> > > 2) In general, I strongly disagree that "/" is normally used for 
> > >    integral division. In all sophisticated programming languages,
> > >    integral division has its own operands, while "/" is 
> the operand
> > >    generating a decimal result. 
> > 
> > Lars-Erik,
> > 
> > Have you ever written code in C/C++? I don't know if they are 
> > qualified as "sophisticated programming languages" you mentioned
> > above, but "/" below generates an integer. 
> > 
> >    unsigned int i = 10 / 3;
> > 
> > BR, Zhigang
> 
> Sure, I have written lots of C code, and no, C/C++ does not qualify
> as sophisticated languages in my mind, especially not when it comes
> to type handling. 
> 
> Anyway, this is just a minor side issue and a detail in the exact
> wording of the TS encoding clarification. Unfortunately, I have to
> agree with you that you seem to be all alone with your opinion on
> this subject, and I am sorry to say that we will thus not be able
> to get complete WG agreement on the issue. However, we do have rough
> consensus to go with the resolution now present in the implementer's
> guide (although we might adjust details in the wording).
> 
> Rgds,
> /L-E
> 

_______________________________________________
Rohc mailing list
Rohc at ietf.org
https://www1.ietf.org/mailman/listinfo/rohc