Mark, As you're feeling your oats.
> ... but > there are some inaccuracies that should not be left hanging. I wish the UTC felt that way about a) making any recommendation as a body concerning any any attempt by another body to utilize the work product of the UTC, and b) specifically making an operational suggestion that does not appear to be accompanied by operational data, or a non-naive understanding of the dns. ... > We are very curious as to the origin of this "printer company" story; where > did you hear this? That's "printer consortia". Private, with $$ to sit at the table. Like W3C now, but worse. It was my impression in 1991, at UTR#4-time, at 2.0-time, and at UTR#8-time. It [Unicode] didn't spring out of the operating system industry. So which peripheral device consortia is left? Disks? Modems? It is still my impression today. Having writ XPG/1, and XPG/4.2, and knowing most of the P1003.x particpants of the 90's, and the business and technical managements of Bull, ICL, Siemens, Olivetti, Nixdorf, IBM, Sun, SGI, and HP during the 90's -- I really don't care how much you'd prefer to believe that the non-printer lines of business considered Unicode as strategic. I watched them staff-up for the iso2022 period, and down-staff, and staff-up for the File-System-Safe period, and down-staff. "Them" means Unix vendor business core unit, not printer unit. Just because we (first Apple, then MicroSoft, then us Unix vendors) all got within a transform format of Unicode by 1996 doesn't mean that Unicode was a core value proposition. At HP HP-15 was vastly more important, and at SMI the multi-byte code-path for sort(1) was simply abominable, HP-UX 10.0 and Solaris 2.4 and 2.5 -- when we multi-byted (UTF-8'd the tty, the file system, and all the user libs and apps). > I only take the time to correct some of the items above because a mistaken > impression of the process of development of Unicode and ISO 10646 might lead > people to have a mistaken impression of the quality of Unicode and 10646, or > the organizations behind them. Jeez. I made mistakes in XPG/1. I let Nixdorff shave bits off a pid for an early SMP-which-processor identifier. I made mistakes in Spec1170 also. You guys act as if you never goofed -- and you guys actually put Klingon, Esperanto and Pharonic Egyptian in your runqueue -- ahead of living languages. We could replace Unicode as the basis for work on internationalizing the dns. It wouldn't be a lark, but at least we already know what _not_ to do when creating encoding for identifiers, and in terms of incongruity, it couldn't be worse than ACE-for-labels, something-else-for-everything-else. To quote from the report of the 1996 invitational workshop (not one was a Unix implementor) E: The IAB should encourage the IRTF to create a research group to explore the open issues of character sets on the Internet. This group should set its sights much higher than this workshop did. That was a sound recommendation. It hasn't been acted on. It should have been, and it still should be in the present. Just to keep things honest, from rfc2044: UTF-8 was originally a project of the X/Open Joint Internationalization Group XOJIG with the objective to specify a File System Safe UCS Transformation Format [FSS-UTF] that is compatible with UNIX systems, supporting multilingual text in a single encoding. The original authors were Gary Miller, Greger Leijonhufvud and John Entenmann. I replaced John Entenmann (SMI) on XOJIG in 1995, and Gary Miller (IBM) and I came out with our code-set independent architectures for our two repsective operating systems (AIX and Solaris) in 1995/6. We have bigger fish to fry than the ruffled feathers of Unicadettes, or anyone else for that matter. Anyone who wants to say they heard it from me is free to so state. Some days I look at national standards and think, is that really any worse than this mess? Eric
