> Not safe for what? I've come across six characters thatweren't in
> Unicode at all.
As a side note, for Everson, all of which I've reported to this list before.
Then I'm puzzled as to the purpose of this proposed subset.
Right now, we accept Latin-1, because the guy who created it didn't think about character sets at all, with various markup systems for other Latin characters and transliterated Greek outside of Latin-1, and no way to handle anything outside that range. We have a selector, so if you're working with a standard American keyboard at a library, you can still input the Latin-1 and Greek characters. This also works well if you don't want to mess with your system at home.
I'd like to break down the sections of Unicode into similar size panels (less then 96 characters) that can be swapped in and out. If we decompose characters, Latin-1 and Extended-A can be stuffed in one panel, replacing the Latin-1 panel we currently have. It would be nice to stuff the rest of Latin in one panel.
Books before 1923, especially scholarly books concerned withlanguage or
mathematics and logic, might contain almost any charactercurrently
coded in Unicode as well as characters not currently coded inUnicode
including idiosyncratic characters that will never be encodedin
Unicode. They are also likely to contain characters fromnon-Latin
scripts and many symbols.
Math is moot right now; we're using TeX. Non-Latin is moot, because that's handled by other panels, and with the exception of Greek, is probably going to get passed off to relative experts. Some selection of symbols will find its way into a panel.
Why prescribe a closed subset?
I'm not. I'm creating a useful tool. If you want to open Character Map and insert characters, then you are welcome to do so. But some can't use that tool (i.e. library computers), and many want something more convienant and localized to what they'll encounter. --
___________________________________________________________
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm
