I agree with you completely. What could be done is sort of have script with names used for J.
Then have a rule script with those names exchanged in a toggle to unicode set in a table. So in J the rule is use names. For display purposes one script version is with the names and another script those names replaced for those who want the names displayed as unicode. J would not have to bother with the unicode and the user can read it in unicode or names as wanted/needed. Maybe only translate from names to unicode for display or toggle between the two displays. Just create a separate display tool for scripts. The unicode script not valid in ijx only as ijs (possibly named iju or ijsu) On 5 Jul 2016 21:50, "Eric Iverson" <eric.b.iver...@gmail.com> wrote: > We made the decision well more than a decade ago that unicode identifiers > would be a mistake. That decision was unanimous within Jsoftware at that > time. > > It would have been just as easy to add the support then as it is now. Has > anything changed that would make us reconsider? > > I can only comment for myself. > > There are 3 main reasons I am against it: > > 1. It is a fringe area and does not warrant the effort it would take - very > little bang for buck. > > 2. It is deceptively easy at first, but is a slippery slope. As is pretty > much everything with unicode. European accented letters seem like a > no-brainer. Then CJK. Then lots of others. Then lots of special guys. APL > symbols. ETC. Glyphs that look exactly the same on paper, but that are > different code points. This takes thought and (see 1) it just isn't > warranted. > > 3. Ken left us with many fundamental ideas. One was that notation is a tool > of thought. The correllary is that notation is a way of communication. If > we limit J identifiers as they currently stand then algorithms can be > easily and effectively be communicated around the entire world. Let in > unicode identifiers and this would suffer enormously. > > Unicode is for data and the support there is pretty good. It serves no > useful purpose in identifiers and would be a serious impediment to > communication. > > For historical reasons the English alphabet has a privileged position in J > identifiers. Perhaps this is wrong in some senses, but it is enormously > practical when it comes to international communications. > > I'd be very happy to not talk about this again for another 10 years. > > On Mon, Jul 4, 2016 at 3:53 PM, Jose Mario Quintana < > jose.mario.quint...@gmail.com> wrote: > > > On the one hand, Marshal asserts that Unbox allows the use of UTF-8 based > > identifiers in a way that "is completely backwards-compatible with > existing > > J." which I find very appealing. > > > > On the other hand, you (Jsoftware) decided strongly against it because > > "the disadvantages > > strongly outweighed the advantages." > > > > Would you mind to elaborate on what the disadvantages are? > > > > > > On Sun, Jun 12, 2016 at 4:38 PM, Eric Iverson <eric.b.iver...@gmail.com> > > wrote: > > > > > We (Jsoftware) talked about unicode identifiers quite a bit years ago > > when > > > we added uft8 and utf16 support to J. We finally decided we were > strongly > > > against it. The disadvantages strongly outweighed the advantages. I > don't > > > think anything has changed in the interim. > > > > > > I doubt unicode names will be in official Jsoftware releases for a long > > > time, if ever. > > > > > > On Sun, Jun 12, 2016 at 4:30 PM, Marshall Lochbaum < > mwlochb...@gmail.com > > > > > > wrote: > > > > > > > Unbox has code to allow unicode identifiers in J, with the following > > > > rules: > > > > > > > > - All code must be UTF-8. Invalid UTF-8 causes a spelling error. > > > > - Any non-ASCII character is treated as alphabetic. Identifiers can > use > > > > these characters freely. > > > > > > > > This is completely backwards-compatible with existing J, and allows > us > > > > to use things like greek characters and code in other languages: > > > > > > > > π > > > > |value error: π > > > > π =: 1p1 > > > > π > > > > 3.14159 > > > > π_1 > > > > |value error: π_1 > > > > > > > > What do people think about this? Should it be added to jsource? > Should > > > > the rules be changed for some characters? > > > > > > > > Marshall > > > > > ---------------------------------------------------------------------- > > > > For information about J forums see > http://www.jsoftware.com/forums.htm > > > ---------------------------------------------------------------------- > > > For information about J forums see http://www.jsoftware.com/forums.htm > > > > > ---------------------------------------------------------------------- > > For information about J forums see http://www.jsoftware.com/forums.htm > > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
