(Crossposted from private email to Ken, with permission)

>  I do not personally like the example you gave and think 
> it unnecessarily makes the scene language harder to use and 
> understand. 

	I'm working on the assumption that hand-editing is
	not the only way to describe a 3D scene.  XML I think
	is editable, but certainly not optimised for this task.
	It is however, far more generic and flexible.  XML would
	never be a replacement for POV script, only an alternative.

> And correct me if I am wrong but isn't HTML just a glorified text based
> word processing and illustration language designed exclusively for 2d
> work? 

	HTML began as a mark-up language for ASCII.  The idea was
	to encode the structure of the document (headings, titles,
	paragraphs, lists) into the document, and let the browser
	decide how to format the result.  This is not a new idea,
	LaTeX does a similar thing, a document processing system
	that to me is similar to XML in its reasoning.

	Then, when the World Wide Web became mainstream, HTML got
	polluted with tags that control the appearence of the page -
	fonts and colours, table style, buttons,  etc.

	HTML is being evolved to XML in the form of XHTML, version 
	4 of the HTML standard.  The idea of XHTML is that layout
	information is in a different file to the document.  If you
	want to change the way the document looks, you change the
	style sheet, rather than editing the data.  Therefore, 
	style information is a template for any data of a given
	format, rather than being copied into every instance of that
	kind of data.

	The document then is purely a data model of the information,
	free of nasty formatting information.  The XML design allows
	databases and other applications read and write XML data
	without being bogged down in formatting information.

	VRML has been around for quite a long time, it is a file
	format that describes many similar objects to those in 
	POV.  VRML is evolving towards XML.  The advantage is that
	you no longer need a "VRML parser", you use a generic XML
	parser and work with the internal representation of the
	data - then write the data back out through a generic XML
	emitter.

> I just don't see how it, or a variation of it, would apply to a
> 3D object oriented program such as POV-Ray - not to be confused with
> OO programming.

	XML can be used for lots of data-description or 
	data-modelling problems - including POVray.
	Is it not reasonable for POVray to be able to
	interface to a database system for complex
	scene generation?  Or, a visualisation package,
	or a CAD system?  Or an XML editor with tools for
	graphical editing of POV specific information?
	An interactive texture editing tool?  An 
	interactive CSG modelling application?
	
> > For someone with a "povray.org" email address, you are
> > behaving in a very entrenched, "old guard" kind of way.

> When I am speaking in the news groups I do so as myself and not
> representing the views of the POV-Team - which I am not even a
> member of. 

	Ken, perhaps I came across too strongly, but
	given that you started the thread as a point
	of discussion, I thought you should be more
	neutral.  Having a "povray.org" email
	address is both a privelege and a responsibility.
	There is still a perception of "official
	viewpoint" even if it unintended.

	Creating a lynch-mob environment will simply
	scare programmers away from making any kind
	of suggestion.  Clearly my perspective is quite
	foreign to many POV people, but it doesn't
	mean that I am allowed to be as critical of
	POV users, as users seem to be critical of any
	new or different idea.

> As far as entrenched and old guard let me point out as a POV-Ray
> user who is not only proficient with the program, but also as one
> who is satisfied with it's current language structure, I have every
> right to defend it. 

	You're not only defending what you're like, but you're
	attacking the unknown.  These tactics are not fair,
	and not productive.

> I think some of the changes being proposed
> jeopardize the ease of use I enjoy with the program 

	No, they don't.

> One of the key factors that has let POV-Ray
> enjoys it's success and popularity to date is the fact that
> it appeals to such a broad range of people that use if for
> very different reasons. If you lose sight of this you run the
> very real risk of alienating a large portion of it's user base.

	No, because I have never suggested removing anything
	from POVray that is already there and makes it 
	desirable to these people.

	I keep hearing this threat on the povray newsgroups
	of "alienating the user base" - but I don't quite
	understand it's basis.  We certainly need a better
	model of interaction than this polarisation.
	
	As a commercial software developer, I spend all day
	pondering the user benefit consequences of software
	design decisions.  I must admit that I find it
	strange to be accused of being an "enemy of the users".


--
Nigel Stewart (nig### [at] nigelscom)
Research Student, Software Developer
Y2K is the new millenium for the mathematically challenged.

Post a reply to this message