IDE Minutes 2009 07-07
- Javier Pedemonte, IBM
- Lori Hylan-Cho
- Jon Ferraiolo, IBM
- Adam Peller, IBM
(1) Proposed loose wording for algorithm for finding localization files
Spec just says RFC 4647 has algorithms. Doesn't say you must any particular algorithms. Jon thinks this is the right way to go.
Jon: There are three major algorithms in 4647. I don't think we should legislate one. It will be a victory if developers implement any of them reasonably. The goal is to match user's locale with the right message bundle file.
Jon: Noticed a grammatical error. Needs to say "can use"
Lori/Javier: Sounds like a reasonable approach.
(2) Are we sure: no substitution on preload, postload, inline require?
Jon asks "are we sure"?
Everyone: Should be easy to implement after everything else is working.
Javier: Yes. Also might have inline CSS that you want to namespace.
Javier/Adam: Why did we decide this before?
Lori/Jon: Pretty sure this was just an oversight.
RESOLUTION: Should do variable substitution on inline REQUIRE, POSTLOAD and PRELOAD
(3) A few details concerning Variable Substitution chapter
(a) __WID__ characters: start with ASCII alpha, then subsequently only ASCII alpha, ASCII numeric or underscore (b) Substitution order: localization, property, __WID__, BIDI (c) Is XSS warning OK?
(discussion about Jon's reasoning for (a))
RESOLUTION: Jon's proposal for __WID__ is good
Adam: Might implement all var sub in a single pass. Do we want to allow chaining?
Jon: No. Maybe we should say SHOULD NOT combine.
Lori/Adam: Order should be implementation-dependent
RESOLUTION: Change spec to say variable substitution order is implementation-dependent and developers SHOULD NOT combine
Jon: For (c), I just wanted people to look at the XSS warning and say whether it is appropriate for the spec
Adam: About the best we can do.
Javier: I like the wording. General problem, and then here is what we provide to work around it.
(4) Any products out there with %% localization support already?
TIBCO is running into a conflict with existing products which already use similar syntax and is wondering who has implemented %% and if this can be changed (perhaps ## or ~~ or __MSG_key__ like OpenSocial).
Jon: Neither Howard or Adobe is here. Let's push this to another phone call.
(5) Outstanding editorial tasks
- Factory method example (Lori completed this one. Thanks, Lori!)
- CDATA discussion within widget chapter
- Example for <parameter> section
- Example for 'includeRef'
Jon: Lori did the first one. Thanks, Lori. If no one works on any of the above, I'll do them, but not in the next few days. The PARAMETER one has an example that uses features not in our spec and is too long
Lori: I'll work on the PARAMETER example.
(6) Finalizing the metadata spec
- What implementation experience do we need?
- Detailed spec reviews: set a deadline?
- Target dates for finalization and approval?
Jon: We have various implementations already. Aptana and open source on API Metadata side. (Jon forgot to mention Eclipse JSDT) On widget side, we have open source and Adobe, and at least one other Enterprise company. (Jon forgot to mention Rational Application Developer and other IBM products in progress and SAP)
Jon: What's the plan for finishing? I'm in the mood to finish early.
Lori: Me, too. We have been working on this a long time.
Jon: Let's discuss this with others at a future phone call, particularly Adobe and TIBCO.
Jon: We need people to review the spec carefully. I'm going to do two more complete read-throughs before we finalize.
Future phone calls
Jon: If we can get Adobe and TIBCO next week, we can discuss the last two issues. Then if no more issues, we can switch to an as-needed basis.