Interoperability Minutes 2007-09-19

From MemberWiki

Jump to: navigation, search

Full minutes: /member/wiki/Interoperability_Minutes_2007-09-19

Contents

OpenAjax Alliance Interoperability Committee meeting minutes 2007-09-19

Attendees

  • Gideon Lee <glee(at)openspot.com>
  • Jon Ferraiolo <jferrai(at)us.ibm.com>
  • Ted Thibodeau <tthibodeau(at)openlinksw.com>
  • Coach Wei <coach(at)nexaweb.com>
  • Bertrand Le Roy <bleroy (at) microsoft.com>
  • David Boloker <boloker(at)us.ibm.com>
  • Howard Weingram <weingram (at)tibco.com>
  • Larry Koved <koved(at)us.ibm.com>
  • Naohiko Uramoto <uramoto(at)jp.ibm.com>
  • Todd Kaplinger <todkap(at)us.ibm.com>
  • Frederik De Keukelaere <EB41704(at)jp.ibm.com>
  • Michael Steiner <msteiner(at)us.ibm.com>
  • Sumeer Bhola <sbhola(at)us.ibm.com>
  • Suresh N. Chari <schari(at)us.ibm.com>
  • Haik Sahakian <Haik.Sahakian(at)fmr.com>
  • Sachiko Yoshihama <sachikoy(at)jp.ibm.com>
  • Mike Pittaro <mikeyp(at)snaplogic.org>
  • Gustavo Munoz <gustavo.munoz(at)jackbe.com>
  • Stanislav Malyshev <stas(at)zend.com>

Original Agenda

Minutes

Topic: Introductory comments

Jon: We have an exceptionally high turnout today because we will be discussion Hub 1.1 with people from InteropWG, now defunct CommHub TF and Security TF. CommHub TF has now merged with InteropWG. Therefore, the Security TF people are special guests today.

Topic: InteropFest

Jon: Reminder - press release next week. quotes need to be in by tomorrow

Bertrand: MS will be participating

Jon: Do you want to be listed in the press release?

Bertrand: Yes

Jon: Provide a quote?

Bertrand: Probably not.

Jon: I'll send you the information.

Topic: Face-to-face meeting

Jon: Big week next week. AJAXWorld mon-wed. OpenAjax f2f thurs. Mobile Ajax workshop on friday. Our generous hosts MS need to know who will be attending. Is there anyone here who will be attending that hasn't added their name to the registration page?

(one person says yes and then adds his name to the registration page)

Jon: I included a link to the f2f agenda in the agenda for this meeting. If anyone has suggestions about missing topics or other changes, please tell me.

Topic: OpenAjax Registry

(Jon asks everyone to navigate to /Registry and subdirectories. Explains the assessments/ subdirectory and the search_globals.html tool which loops recursively through the window object to look for all JavaScript objects and properties and then allows comparisons to determine what changes were added to the JavaScript environment.)

Bertrand: What does the logic do?

Jon: Starts with window object. for/in loop to get all properties. If a property is an object, then recursive go through that object.

Bertrand: Should be easy enough to add logic to look at the built-in types and add lookups for HTML extensions.

Jon: Please send in the code.

Bertrand: Will do.

Jon: Feedback on this approach?

Howard: Have companies do this themselves.

Bertrand: Yes, in everyone's interests to submit accurate information.

Howard: Self-police. We aren't trying to do this for everyone in the industry.

Jon: Yes. I was thinking that toolkit vendors would submit an HTML file along with their proposal registry entry.

Howard: Some toolkits have many instantiations. Don't want to put a burden on ourselves to study all of those permutations.

Jon: How about we encourage them to use the tools.

Howard: At some level, it is like the InteropFest.

Jon: OK, here is a proposal. I'll take this discussion and update the Registry wiki page to reflect this discussion then send an email telling everyone to review the diffs. Any objections to this approach?

(no objections)

Jon: Next proposal is that we start with some easy well-behaved toolkits to see if the Registry process works for them. Then add a toolkit such as Prototype that extends core objects. Then look at MS, which is likely to be more complex.

Bertrand: How about json.js in addition to Prototype.

Jon: Yes, good idea.

Topic: Hub 1.1

Jon: This is largely a repeat of things we have talked about many times, but this time we are all together so I will present this again. The target features for Hub 1.1 are to extend pub/sub beyond a single frame to cross-frame, add a framework for secure mashups, and add mediated comet support per the work in the CommHub task force. The proposed process is to have volunteers work on experimental open source. This is the same process that we used for Hub 1.0. People produced experimental open source. When the open source was ready, we talked about it and decided what we liked and didn't like. We consolidated contributions from multiple sources, such as some of the Tibco/Dojo contribution with some of the Nexaweb contribution. What do people think?

(no comments)

Jon: The SMash team has already submitted their source code to the open source project. I am expecting them to transform the SMash code into a form such that it will be upwardly compatible with Hub 1.0. When that is at a point, we will look at it just like any other contribution.

Jon: The timeframe I have in mind is to have working code by the end of 2007. If there isn't enough volunteer effort, I will probably have bandwidth after the horrible week next week to do some of this work myself.

Topic: Version string issue

Bertrand: Don't feel too strongly one way or the other. Not sure it is worth ...(lost what he said)

MikeP: Is the issue whether we think people will want to do version number comparisons?

Ted: ODBC has had success with four numbers separated by dots followed by arbitrary text. The version comparison is rarely needed in early phases but is used more in later phases.

(others - thumbs up on ODBC's success in this area)

Frederick: What is the purpose of the free text?

Jon: I can only guess what Adam had in mind and assume that most often it would be used to say alpha N or beta M or build N.

Frederick: Human readable?

Ted: Yes. But the leading numbers can be used for comparison.

Jon: Here is what I propose. Ted sends links to ODBC specs. We agree in principle to adopt the ODBC approach pending detailed review. But no real decision until we review the ODBC specs. Any objections?

(no objections)

Personal tools