From alb@sct.gouv.qc.ca Wed Jan 8 15:16:25 1997 Received: from socrate.riq.qc.ca (socrate.riq.qc.ca [199.84.128.1]) by dkuug.dk (8.6.12/8.6.12) with SMTP id PAA15693 for ; Wed, 8 Jan 1997 15:16:04 +0100 Received: from 506.riq.qc.ca (riq-44-150.riq.qc.ca) by socrate.riq.qc.ca (5.x/SMI-SVR4) id AA02866; Wed, 8 Jan 1997 09:14:18 -0500 Date: Wed, 8 Jan 1997 09:14:18 -0500 Message-Id: <9701081414.AA02866@socrate.riq.qc.ca> X-Sender: alb@riq.qc.ca X-Mailer: Eudora Light pour Windows Version 1.5.2 Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit To: trondt@barsek.hsf.no (Trond Trosterud) From: "Alain LaBont/e'/" Subject: Re: keyboard standard (ISO 9022???) Cc: everson@indigo.ie (Michael Everson), sc18wg9@dkuug.dk At 07:54 97-01-08 +0100, Trond Trosterud wrote: >Dear Alain! > >Michael tells me that you are working with keyboard standardization. In the >Sámi computer technology working group we are, among other things, working >on a keyboard standard for our Latin 9. Our intention is to make it >according to three principles, hierarchically ordered: > >1. the (one that we chose among the conflicting) Sámi standard(s) >2. the Scandinavian standard (the æøå-stuff) >3. the international standard for placing the Latin 1 glyphs (we just pick >the remining ones and place them where you tell us) > >Actually, the philosophy is the same as for sorting (I attatch our sorting >suggestion to this letter). > >To this Juhani Lehtiranta (a Finnish system developer) protests. He says: >For mac this is fine. They can take 48 x 4 glyphs + the dead keys. PCs >cannot. First, AltGr and Shift cannot be combined, and second, many >programs use AltGr as a key for commands. Keyboards on PC should confine >themselves to the essential glyphs (i.e. to points 1. and 2. above). >Michael claims that this problem can be overcome (at least for W95 and W >NT), and Juhani, again, maintains that 3.11 is what is found out there, and >that this work should start at the international level and not among the >sámis anyway. > >As you see, I have several questions: > >1. Is there a standard (Michael mentioned some number that I do not >remember), and where can I see it? You should look at ISO/IEC 9995 (8 parts, but particularly parts 3, 1, 2 and 7 in order of priority). To be complemented by a recent Input Method standard, ISO/IEC 14755 (it is in the publishing phase, I am managing this right now with the ISO Central Secretariat). I am the current project editor of all these standards. You can get copies of these from the Norwegian national body member of ISO/IEC/JTC1. >2. What is your opinion on the PC problems? According to ISO/IEC 9995, keyboards are arranged in groups. Normally you have a group 1 which is the "national" group for one given population. This group can typically accommodate 3x48 characters or dead key combining elements (for European keyboards; there is also a French keyboard, called the Neuville keyboard, that could accomodate up to 3x51 keys). You normally arrange your national needs in this group. The only question is: for Sámi needs, you don't have a problem if for everyday's life, you don't need more than 48x3 characters or combining elements. Now for "foreign" characters within the Latin script, you can also use *the* group 2, which is by definition the extra European Latin characters that can't be found on your keyboards (this group is composed of all characters of the ISO/IEC 6937 character repertoire that are not common to all European keyboards). You access this group by a "Select Group 2" function (on the Canadian keyboard I use the key that is used as a "right Control" key on an American PC keyboard (same scan code) is dedicated to this function, and that's with this feature that I can enter the character á of Sámi (I did not cut and paste from your message, it's easier to enter it), for example, as it is not a national character in Canada. Group 2 has 48x2 characters or combining elements (in fact many keys, less than 10 though are dead keys in this group), *in addition* to the 48x3 of the group 1 you normally have in your country. You could also use a COMBINE function (the Neuville keybaord and Sun workstations have this function) to complement your keyboard layout, and beyond, ISO/IEC 14755 potentially lets you access all characters of the world if your needs go much beyond your country. The ISO/IEC 9995 standard also has provisions for group 3, and others (it is unlimited in fact, but groups beyond group 2 are not defined; in Canada we will have to define many of these groups for Amerindian languages, for example). So my opinion is that 48x3 or 48x4 is not a real limitation. But the international standard takes it by chunks of up to 3 elements per key in a given group, not 4. The way that the Mac does it with 4 keys could perfectly well be (and in conformance with ISO/IEC 9995, in my opinion) considered as the use of both groups. It is just that you use an extra (a 4th) key in this case as the group select function key to activate group 2 when you need it. So my opinion is that it is highly conciliable and that has been done in Canada. >3. I have been looking for an evaluation of different keyboard positions >(good/bad: the lead line asdf is better than the zxcv, the edrfujik area is >easier to type than the qwzx area, etc, but do we have research on this?) >but not found any. Do you know of any? I have studies in my archives that show ease of use of all keys on all rows for normal people (and even there it is biased for right-handed people). It is interesting but not really useful. Given that, when you start from scratch, you could assign the easiest keys to the most frequently used characters *in your primary language*. However if you do not start from scratch, this does not make sense as experience has shown. People will never get used to such a dramatic change. Slight rightings can be made that can considerably improve productivity, as Yves Neuville has shown in France (he made a doctorate thesis on this and he built a keyboard that is commercialized in France, adaptable to other countries; his studies are available in French; these are the most serious and complete studies I have seen on the subject). You can contact Yves Neuville by phone (he is the rapporteur of the keyboard group of ISO/IEC JTC1/SC18/WG9) at +33 1 46 33 40 60 or by fax to +33 1 43 25 50 60 (in Paris). You probably know that QWERTY keyboards were made to *slow* typing speed down in English because mechanical typewriter hammers jammed when one typed too fast 100 years ago or so (in fact as such it is slightly more efficient nowadays for French, for example, than for English [it is almost insignificant though], but France has an AZERTY keyboard only because of a patent issue of 100 years ago: they did not want to pay royalties to Remington; in Canada for French we use QWERTY keyboards and in Switzerland they use QWERTZ keyboards). The righting of QWERTY keyboard inefficiency *in English* has led to the DVORAK keyboard, but I know only one person in the USA who uses it, an ergonomist at Apple, whom I have seen to "sin" with a QWERTY keyboard himself years after he told us that he was a DVORAK keyboard user! (: > >4. Do you have other comments to the keyboard issue? I think that what precedes is a good start. I invite the Sámi people, if they can afford it, to attend ISO/IEC SC18/WG9 meetings (the next one is in Sydney, Australia, in beginning of April). Current discussions are on standards for portable computer keyboards, for split keyboards and a technical report on future keyboard and entry devices and methods of the future. > >5. Comments to our sorting string? I will have to decode the BinHex file you sent me afterward, first. I just by chance succeeded to find one, after years of research, for PCs. > >Trond > >___________________________________________________________________ >Trond Trosterud email: trondt@barsek.hsf.no >Barentssekretariatet, P.O.Box 276, work: +47-7899-3758 >N-9901 Kirkenes, Norway fax: +47-7899-3225 >http://www.norut.no/barsek/ip/iphome.html home: +47-7899-2243 >------------------------------------------------------------------- >Character test string, please ignore: ᢤ±³¸º-Á¡£¯²µ¹-⦪®½¿-Â¥¨¬¼¾ >___________________________________________________________________ Regards. Happy new year to you and people in Kirkenes, which I had the chance to see (what a beautiful landscape around midnight in the summer!) in June 1990. Alain LaBonté Québec cc Michael Everson ISO/IEC JTC1/SC18/WG9