[exprla-devel] Re: [XPL] XPL Working Draft

[reid_spencer <ras...@re...>]

--- In xpl-dev@y..., Jonathan Burns <saski@w...> wrote:
Richard Anthony Hein wrote:

> Hi everyone.  It's great that everyone seems keen again on getting
> down to
> some business.  I was starting to lose hope.  I hope now that soon 
we
> will
> have substantial gains made once we are able to answer the vital
> questions
> necessary to get to the next level:  establishing the actual
> programming
> model of XPL.  However, first we need to have a framework in place 
in
> which
> to operate.  Once this has become a reality, things will be much 
more
> organized, and we will all know where to put our ideas in order to
> ensure
> they are not lost in confusion, and will be addressed quickly.

We do need the framework, and the one below looks pretty good.

I know I've rather lost track of where my own contributions fit into 
the
big picture.
Basic data and machine architectures, fogs, patterns, interfaces,
hierarchies ...
makes my head spin. But most of it seems to fit into the 
classification
below.

Plaudits for taking the initiative.

>
> In addition, this will encourage others to contribute where they 
can,
> by
> making it easier for them to focus on their strengths, access
> information
> related to their ideas and contributions more effectively, and have 
an
> all
> around better idea of the current and ongoing state of XPL.

ABSOLUTELY.

>
> Since this is a cooperative effort, I am asking you all to look at
> this
> table of contents, and add your points to the list of items where
> appropriate (please refrain from writing up issues - just points at
> issue
> will be included).  Feel free to discuss, disseminate, and add
> anything you
> feel is necessary towards the development of XPL.  Once we have
> established
> a framework of issues from which we can easily draw information from
> and put
> information into, we will be on the way to some serious development.

We'll need to think up a way of accumulating discussions under each
heading separately.
I suppose we could elaborate our mail headers. For the time being, 
this
is probably your
job by default, Richard.

>
> You may note that this TOC is somewhat similar to the Working Drafts
> that
> the W3C issues.  This is not an accident.  The goal of this entire
> document
> is to eventually have a completed draft.  As a working draft, this
> document
> will be made available online, at our website, with links to 
messages
> which
> relate to the specific points.  It is the hope that this will
> facilitate
> easy access to ongoing discussions, while providing everyone with 
the
> overall picture of the project as a whole.  The final version of the
> draft,
> which will be submitted as a Note to the W3C, will have only the
> accepted
> points.
>

I accept the W3C submission as a goal - but my attitude is that the
framework is to
help us organize ourselves, primarily.  I won't be inhibited about
suggesting something
because it might be at odds with existing W3C specifications, or 
because
it might
cohere with a non-standard or proprietary system. We can postpone such
worries
to the endgame, I think.

So, let me see if I can fill in the blanks.

>
> =B7      Abstract

As XML is a meta-language for data, so XPL is proposed as a
meta-language for
programming languages, which operates on data described in XML, and is
itself
expressible in XML.

The principal purpose of XPL is to provide convenient programmability 
of
Web
content.

>
> =B7      Mission Statement

To define a range (ideally the range) of operations which can be 
applied
to data
which is defined in XML.

To define a semantics for combining those operations into programming
languages.

To define a syntax (ideally all-XML) which completely and 
unambiguously
represents that syntax.

>
> =B7      Vision(s)

There will be programs which operate on XML documents at the semantic
level.

XPL document types will define programming languages in which such
programs
can be written.

Programs in XPL-defined programming languages will be expressed in 
XML.
They will be transmissible in XML form to any XML-capable client,
including
XML-capable browsers.

XPL-defined programs can run either client- or server-side, where 
such a

distinction is made.

XPL-defined programs can run on any peer in a (serverless) peer-to-
peer
XML-capable protocol.

There will be a standard XPL, and standard XPL-defined programming
languages,
which are defined independently of any proprietary enabling 
technology.

>
> =B7      Scope

XPL will provide the ability to define operations in terms of any API
which can be
expressed in XML. [The Universal Glue Principle.]

> =B7      Design Goals (Accepted)
>
> =B7      Design Goals (Under Review)

It will be possible to duplicate the functionality of existing
programming extensions
to XML (e.g. Miva, Cold Fusion) , as XPL-defined languages.

Any user of an XPL-defined programming system will be able to ensure 
the
security
of the local system against errors or incursions permitted within the
XPL system.

> =B7      Requirements (Accepted)
>
> =B7      Requirements (Under Review)

Several requirement sets have been posted to the list - all these will
be taken under
review.

> =B7      Terminology (Accepted)
>
> =B7      Terminology (Under Review)

[ Big issue. Don't start on it, Burns. ]

>
>
> =B7      Semantics (Accepted)
> o      Data Model

> o      Production Rules

> =B7      Semantics (Under Review)
> o      Data Model

Trees.

> o      Production Rules

Definable in some grammar, as tree operations.

> o      Interfaces

Definable in terms of tree patterns

> o      Primitive operations

Include at least, string, regexp, boolean, artihmetic.
Relational?

> o      Resource Identifiers

Local, remote? Protected? Encrypted?

>
> =B7      Syntax (Accepted)
> o      XPL Elements and Attributes
> o      Data Types
> o      I/O
> o      Patterns
> o      Component Interaction
> o      Object Model
>
> =B7      Syntax (Under Review)
> o      XPL Elements and Attributes

> o      Data Types
> o      I/O
> o      Patterns

[ Hmmm. I'll leave this alone too, for now.]

>

>
> o      Component Interaction
> o      Object Model

Should these be under Semantics?

>
>
> =B7      Appendices
> References
> Resources
>
> Group Members and Acknowledgements
>

That's about it for now.

More stuff tomorrow.

Jonathan
--- End forwarded message ---