On 22 Dec 02 at 7:27, [EMAIL PROTECTED] wrote: >In article <[EMAIL PROTECTED]> you write: > >> a problem with all these very nice offline html viewers >> is that they do not care about non-English texts with diacritical >> (8 bit) characters. If you wanted to read a Czech text you will have >> to install a iso8859-2 font on your DOS system... > >From what I can see, most of the authors of these small and minority >interest utils tend to write what they need and I'd agree that does >result in a certain parochialism unless they happen to have >correspondents in non-english locations. How many ever get asked about >other's needs, I wonder ? > >The current Browse has an attempt at handling 8859-1/Win1252 so the >principal of character translation is in (quite good if the default page is >850, not so hot if it's 437, probably rubbish with any other). Of course, >mere translation is a far cry from loading a new font. To be honest, the >only driving force for this sort of thing in the UK is the need for the UKP >currency symbol, although in time the EURO may be another. The real problem >is the small number of program options - if more utils were developed in >non-english areas of the world it would help, or if a greater sense on >internationalism prevailed, but the juggernaut that is english increasingly >sweeps all before it. > >Alex.
Alex, I do not complain at all, about small utilities not reflecting about non-English characters. And the development of code page standards was of course due to a historical constellation, heavily influenced by IBM and Microsoft business strategies. My point is actually this one: the majority of those older programs handling character translation are helpful only with ISO8859-1 (or its DOS equivalents) because the code page tables are hardcoded and the source is not distributed. This was the common way of writing DOS software. However, from the programmer's point of view, it is does not make too much difference if you work out a solution for just character set or for any character set. The unicode tables are freely available on the web. The latest viewers and parsers for RTF or MS-Word files emerging from the Linux world eg. Antiword or CatDoc use to link to those translation tables delivered together with the program. Fortunately some of those younger software authors are still sentimental enough to port their programs to DOS! It is a pity that other DOS authors (eg. MARTHA/ISHTAR or DOX) still improve their programs, but do not include a variable handling of character translation. Christof Lange _______________________________________________ Christof Lange <[EMAIL PROTECTED]> Prokopova 4, 130 00 Praha 3, Czech Republic phone: (+420) 222 78 06 73 / 222 78 20 02 http://www.volny.cz/cce.zizkov
