W3C home > Mailing lists > Public > xproc-dev@w3.org > May 2009

Re: Feedback from the students in my XProc class

From: Norman Walsh <ndw@nwalsh.com>
Date: Fri, 22 May 2009 07:01:57 -0400
To: XProc Dev <xproc-dev@w3.org>
Message-ID: <m2r5yhqu3e.fsf@nwalsh.com>
"Costello, Roger L." <costello@mitre.org> writes:
> 1. XProc is great! Everyone sees the value of XProc and is excited
> about it. Many plan to use it in their work.

Cool.

> 2. Need better error messages from Calabash. Debugging XProc
> pipelines was frustrating. There was one error message that came up
> over and over and was particularly frustrating: null pointer found.
> In frustration one person said, "This is not ready for prime time."

Yeah. I know. NPEs are bugs though, if you can send me pipelines that produce
them, I'll try to fix those.

> 3. Need more XProc processors. Concern was expressed that Calabash
> is the only XProc processor available at the present time.

That's true too. Vojtech's Calumet is/will be available soon, if I
understand EMCs PR. And there are others coming along, I hope.

> 4. Specification going to change? Concern was expressed that XProc
> is not yet a Recommendation and may change.

It could, but I think we're darned close.

> 5. Two people had their pipelines produce the error "Multiple
> primary inputs are not allowed." I, along with the whole class
> examined their pipelines. After 15 minutes we gave up, unable to fix
> it.

Could you send those to me, please? I can't fix the bugs I don't know
about.

> 6. Lax validation working? We tried using p:validate-with-xml-schema
> with the mode option set to 'lax'. Calabash didn't seem to permit
> lax validation.

That sounds like a bug too.

> 7. Need a WYSIWYG tool.

For pipeline construction? Yes. I'm hoping someone else does that.

> 8. Need a Configuration Management (CM) tool.

For what C?

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | The perfect man has no method; or
http://nwalsh.com/            | rather the best of methods, which is
                              | the method of no-method.-- Shih-T'ao

Received on Friday, 22 May 2009 11:02:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 May 2009 11:02:42 GMT