OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

oiic-formation-discuss message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [oiic-formation-discuss] Profiles


On Tue, Jul 29, 2008 at 6:18 PM, Dave Pawson <dave.pawson@gmail.com> wrote:
> 2008/7/29 Peter Dolding <oiaohm@gmail.com>:
>
>> I see this dividing as more negative than good.
>>
>> Sections in the main standard fine.  Where the sections have to be
>> compatible with everything else in the standard.
>>
>> When you break it completely off to its own free will like w3c and
>> done many times over you need bigger and bigger processing engines to
>> handle the differences between them.
>
> I don't read the W3C document like that Peter. There's a pretty picture
> of lots of bits making up a whole (the standard). No reason ODF
> shouldn't stay as a single document and recognise classes of use
> as hinted at? I think it's fact, whether the main TC recognise it or not.
>
You missed seeing what I am seeing explained better over the XHTML.
>>
>> The logic is how to make a mess.  Remember XHTML is a profile off HTML itself.
>
> Beg to differ. XHTML 1.0 was a re-write of html in XML.

Why did HTML need rewriting to XML.  Ok its was not standard XML.   So
you ended up with 2 formats instead of one.

Profiles become dividing for dividing sake as it did with HTML and XHTML.

http://www.w3.org/html/  Look closer and you can see my issue.  XHTML
is 4.0 HTML current HTML is 4.01.   Profiles risk drifting away from
standard.   This has happened many times over the complete w3c kept
standards.

This is why I am saying look at opengl.  Core standard and define
groups made from segments inside the main standard.   So as the main
standard updates so does it.   No sync issue.  These sync issues are
all over w3c.
>
>>
>> This path has to be followed with massive care.   Better to reject it
>> and be wrong then take it and have the standard fragment into a bigger
>> and harder to process mess.
>
> Agreed. The ideas seem to me to quite usable.
>>
>> Opengl ES what is a sub profile of Opengl it has to be what is in the
>> main opengl standard.
>
> Thats in the definition? E.g.
> A profile is a subset of the technology that supports a particular
> functional objective
>
> So a profile could be ODF on disk. Stop. ID all the spec paras that
> address this (and no more)
> and these become the profile? Simplistic, but I could see that working.

Its how it bound to the standard is the issue.  Opengl is directly
linked no dift.

>>  This has kept opengl clean.  W3C models should
>> be avoided since it has proven its self as a mess maker.
>
> So the lessons are there to be learned. Don't call up this document,
> just take the ideas from it and use them for ODF.
>
Profile word has been abused by different groups.   So needs clear
define what one.

Peter Dolding


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]