As the 'moderator' of XML-DEV I deliberately remain publicly neutral about
whether new XML lists should be created or not (just as I remained neutral
about compt.text.xml). However this posting suggests a revised purpose for
XML-DEV, that I'm not in favour of.
>With above two additional mailing lists, we will have four tiers in the XML
>development community:
>
>W3C - for standards
They actually call them Recommendations and (I think deliberately) do not
use 'standard'
>XML-DEV - for implementation and incubation of standards
This has never been the primary or even secondary purpose of XML-DEV and I
wouldn't want it to become that. The procedures for developing standards
lies with the bodies - ISO, IEEE, W3C. They have well defined mechanisms
which may or may not be enhanced through public mailing lists. It happens
that part of what has happened on XML-DEV has included the generation of
new ways of doing things *** paralleled by working code ***. [I suspect
that SAX would have less successful (or taken a longer time to become
successful) if there had not been an implementation which any parser-writer
could use with minimal work.]
No list remains static in a fast-moving field and unlike (say) a newsgroup
XML-DEV has deliberately a minimalist 'charter'.
'XML-DEV is a list for (XML) developers'.
This implies that the postings are related to development of the use of XML
(not XML itself). Where other specialist lists exist (e.g. XSL or RDF) I
would gently point a posting in that direction. [Recently there was a
discussion about XSL that should have been on the XSL list rather than
XML-DEV - I was away for a few days and didn't catch it.] There is no list
for XLink or XPointer and questions on these are welcome here.
The initial motivation 18 months ago was that XML-DEV was a place to
develop working code alongside the developing XML spec. [I have a fear of
specs which are too complex to implement.] A large number of parsers were
developed during this time and several problems were identified in the
emerging spec (e.g. PEs). There were also naturally announcements of new
developments in the use of XML - new tools, new markup languages (i.e. DTDs
or similar approaches). There were also questions *at developer level*
about how to implement the spec(s).
'Developer' is interpreted widely. It can include people new to XML
although I do not expect XML-DEV to be used for introductory questions.
Many of these can be answered by reading the XML FAQ and others could be
posted to comp.text.xml or XML-L if appropriate. However technical
questions on particular pieces of software are often appropriate here.
Later these may develop their own mailing lists.
'Developer' includes but is not limited to:
- those writing XML-based tools
- those making serious use of XML
- those making serious use of XML-based tools
- those developing new DTDs (or other XML 'applications')
- those wanting to work out how to do new things with XML (this is why
there has been a lot of discussion on names.)
If some of these result in a group developing a NOTE or a de facto way of
doing things, fine. But it's not the primary purpose. I note that many of
the progenitors of XML and XML tools post to this list so I assume it fills
a useful purpose for them.
XML-DEV has been used for announcements and - assuming they are compatible
with the above ideas - I'm quite happy for them. [They don't cause huge
bandwidth at present.]
I agree that there has been a considerable amount of general discussion
recently [and I probably contributed in small part]. My current feeling is
that some of the developers/implementers are going through a phase of
taking stock but that there may be a lot of activity 'beneath the surface'.
There are many directions in which to go and only some are achievable.
Almost always the vision is greater than the reality of implementation.
Personally I favour postings that suggest simple ways of doing simple
things rather than all-embracing solutions that require a lot of buy-in
from everyone. XML was difficult enough. XSchema has been created as
deliberately simple at this stage. It sticks very closely to the concept of
the DTD and doesn't require large conceptual leaps. When, I hope, the
discussion on Element-oriented programming starts to coalesce I shall argue
strongly for a minimalist approach.
If you or others wish to create new lists - fine. But the position of
XML-DEV should be seen as roughly what is outlined in this posting.
P.
Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg