I examined a dozen of Tamil truetype fonts (in Unicode as opposed to TSCII encoding) and none of them has U+0B82. TSCII(glyph encoding widely used even now) doesn't have U+0B82 (see http://www.tamil.net/tscii/faq5.html), which can indicate that U+0B82 is not used very often if used at all. In addition to U+0B82, 11 of 12 TTFs (included in Mandrake Linux 9.1: ftp://sunsite.dk/mirrors/mandrake/9.1/i586/Mandrake/RPMS/fonts-ttf-tamil-1.1-1mdk.noarch.rpm ) don't have Tamil digits and numbers (U+0BE7 - U+0BF2). Some fonts lack only a couple while others don't have any digits/numbers. Perhaps, most Tamil speakers prefers Western-style digits (well, they're originated from India as you know well) to traditional Tamil digits. Given the situation of current fonts in circulation, U+0B82 and [U+0BE7 - U+0BF2] may as well be removed from ta.orth. Otherwise, none of fonts in Mandrake Linux 9.1 (also used by other Linux/Unix users) would be recognized as supporting Tamil. I found this problem after adding 'ta' font preference option to Mozilla (see <http://bugzilla.mozilla.org/show_bug.cgi?id=204039>. The patch uploaded there doesn't have font-pref change, yet.).
I asked a native Tamil speakers (Manmathan Kumarathurai : mannan daimi au dk) about those characters and was told that I'm right. Below is his response to my query: <quote> You are right. These characters af very rarely used thats why the font developers didn't use time to create them. They aren't part of the widely used tscii fonts either. </quote>
Thanks for the updates.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.