[whatwg] Authoring tools

[Sander Tekelenburg]

At 06:30 -0500 UTC, on 2006-12-05, Mike Schinkel wrote:

> Sander Tekelenburg wrote:

[...]

>>>> The tools need to be standard and compatible.
>>>I can't follow this. In what sense? Tidy is Tidy.
>>>AWPS authors can incorporate it into their product.
>>>What standard or compatibility plays a role here?
>
> By "standard" I mean referenced in the HTML 5 specification. My opinion is
> the standard should point to a base level implementation for each major
> platform/development tool.

I disagree. The standard should restrict itsef to defining the *rules* for
interoperability. User Agents and authoring tools just need to comply with
that. Exactly *how* they achieve that compliance is up to them. I agree with
what Lachlan Hunt wrote in the related thread "Provding Better Tools":

"The spec already requires authors and authoring tools to produce conforming
documents, and requires user agents to use conforming parsers. It's just that
the specific implementations and implementation details are not for the spec
to determine."

[...]

> By "compatible", I mean these base implementations would need to be
> compatible with each other, i.e. the .NET parser would need to be compatible
> with the PHP parser et. al.

Other than "compatible" in the sense of "HTML5-conformant", I don't see what
sort of compatibility is applicable here.

> BTW, by "tool" I really mean developer component.

I'm not sure what you mean with that. Would that include something like Tidy?

It is the WRI's intention to develop tools (like Tidy, etc.) that can be used
by AWPSs [automated Web publishiing systems] to help them output (not just)
valid code. Indeed when such a tool is written in another language than the
AWPS, that gap will need to bridged. We'll have to see what will be the best
approach for that. It may be possible to define a common API; or it may be
necessary for the AWPS author to build that bridge himself; it may be
possible to port the tool to multiple platforms. I don't see this as the
terrain of a HTML spec though. This is about implementation.


-- 
Sander Tekelenburg
The Web Repair Initiative: <http://webrepair.org/>