From 100434.3031@CompuServe.COM Wed Jul 2 16:51:48 1997 Received: from hil-img-7.compuserve.com (hil-img-7.compuserve.com [149.174.177.137]) by dkuug.dk (8.6.12/8.6.12) with ESMTP id QAA08473 for ; Wed, 2 Jul 1997 16:51:47 +0200 Received: (from mailgate@localhost) by hil-img-7.compuserve.com (8.8.6/8.8.6/2.0) id KAA13560 for SC24@dkuug.dk; Wed, 2 Jul 1997 10:51:14 -0400 (EDT) Date: 02 Jul 97 10:47:32 EDT From: Jean Stride <100434.3031@CompuServe.COM> To: ISO/IEC JTC 1/SC 24 Subject: SC22 Java study group meeting Message-ID: <970702144732_100434.3031_JHU84-3@CompuServe.COM> From: Ivan Herman, INTERNET:Ivan.Herman@cwi.nl DATE: 02/07/97 06:16 RE: SC22 Java study group meeting I was at the SC22 JSG meeting with Dave Duke and Jean Stride, representing SC24. The meeting turned out to be of one day only, instead of the announced two days. Here are some general facts and/or information which I think are of interest for SC24. Javascript (or whatever the name will be): ------------------------------------------ For whatever reason, this landed in this study group as well. However, there is an information which may be of interest. Netscape, Microsoft, and Borland could get their act together, and they ended up by a common specification for this language. It is currently in the hand of ECMA (www.ecma.ch), which has turned it into an official scripting language specification; the only problem is that they do not have an accepted name for it, so temporarily they call it ECMAscript. If the name is fixed, they plan it to get it through ISO through some kind of fast track procedure. Although I asked, it was not absolutely clear to people around the table whether the scope of the language is bound to the browsers' world, or is more general (ie, whether it would have a similar scope as, for example, Tcl, Perl, Python, younameit). However, the browser is clearly in their scope. The reason I thought this was of interest for us was VRML, and its reference and/or usage of a scripting node. Also, but this is a purely personal issue, if this language is ISO standardized, isn't it possible to reverse one of the earlier decisions of the VRML specification, and make the implementation of script nodes in this language mandatory for all compliant VRML browsers? JAVA PAS submission. -------------------- Most of the information coming from various nations were either informal, and/or not definitive yet, so all info should be taken with caution. Having said that, it seems that two nations (Austria and Japan) voted yes with comments, the others being present (US, UK, Netherlands, Norway, France) voted with no with comments. The comments were very similar to the comments of the US, ie, copyright, maintenance. We did not know of any NO vote in the hard way. Only France had some comments on the procedure itself, but this does not appear in the comments on the submission of Sun, but rather as a separate comment on the PAS procedure itself, going to JTC1. If you ask me, and seeing Sun's reply on the US TAG comments (available on the Web), my impression is that PAS submission will eventually go through, it is only a question of time. Of course, this is a PAS submission on the core technology only (Sun's reply reinforces this aspect, by the way!). (The stupid thing is, however, that it will have taken them more time to get an ISO stamp on their submissions that way, than going the VRML way right away...) Role of SC22's Java Study Group ------------------------------- There were of course discussions on this, and the results are obviously inconclusive. The resolution passed by the group asks SC22 to ask JTC1 to be the place for ballot resolution on the Sun submissions (if any), and that SC22 would delegate it to the JSG (which may then become a real working group). However, the procedure of the PAS submission may completely by-pass them, so that looked like wishful thinking to me. Having said that, there was also a discussion of what would be the scope of SC22 the JSG (or the working group resulting thereof), ie, what Java packages they would look at. The final decision is that SC22 does NOT want to infer with any other technical issues than the subject of the current PAS submission, ie, the Java Language, the class file format, the virtual machine, and some of the packages (java.io, java.lang, java.util). In other words, they will not try to interfere with java database, network, media, or graphics packages. This was not absolutely unanimous, and there were some people who wanted an overall Java group having a say in all packages, but this was voted down (we also opened our big mouths on this...) That is it... Ivan