Phoenician and software development

[saqqara]

A few (of many possible) points from a software developers perspective on Phoenician proposal.   1. Assignment of Unicode characters is only a small part of working with ancient scripts. For instance markup schemes such as XML to distinguish languages, ideoms, dialects etc. is essential to deal with multilingual texts. All the more so when treating historical variations. Transliteration is often used in the subject and such conventions need to be sympathetically treated at the user level as well as in the data structures used for exchange of information among applications. Nevertheless, plain text representations should not be confusing or unnecessarily misleading and the Phoenician proposal looks helpful in this respect.   2. There is much confusion among users generally about fonts, glyphs and characters. One reason for this is pre-Unicode 8 bit conventions that needed hack solutions to enable character display. Another is the fact that Word processors (and other software) have an obsession with fonts - form over content. To be concrete: MS Word, OpenOffice, WordPerfect et al prefer to show a block (undefined) character rather than display the actual text in an alternate font. This is sometimes the right thing to do but far more often simply confusing or tedious. In time, it is likely that software gets better at dealing with this kind of problem and presents a less confusing approach by taking a better script and language based approach.   3. Escaping the 8-bit world is not easy. As an example, I am currently releasing a new version of the InScribe hieroglyph font as 14 separate font files rather than one (or two) OpenType/Unicode/PUA fonts. The reason is application compatibility - although most new applications can cope with Unicode, many users have older versions of Windows and applications. I did not want to introduce complications for these users in this release. In a similar vein, application support for Unicode Plane 1 (where Phoenician is currently expected to reside) is patchy at best in the real world at this time so we can anticipate short term crossover difficulties in using such standards in practice. This is not a reason to defer the standards process, indeed new scripts in Plane 1 will encourage software developers to support it.    4. Many users of ancient scripts are not specialists in all (or any) of the scripts they want to work with. Software needs to recognise this and provide solutions that do not require complex explanations of why what seems counter-intuitive is really the right way to go unless there are sound and compelling reasons. There are naturally compromises to made in any specific case. For instance my own work InScribe attempts to be accessible to the new student of Ancient Egyptian while at the same time addressing some of the needs of professional Egyptologists. It does not attempt to be applicable for children with a fascination in the subject, nevertheless a constituency with as many rights for consideration in the standards process as the experts.     Unification of the Phoenician script with Hebrew would certainly eliminate some short term problems - the Hebrew script is fairly well supported nowadays among applications and we'd eliminate the Plane 1 issue. Terribly confusing to users however - the majority do not read Hebrew and we'd be back to hacks to prevent modern Hebrew fonts sneaking in. Unicode is not meant to be purely about fixing short term problems, rather a platform for moving forward.   From my perspective as a software developer, a distinct Phoenician script would simplify design and implementation of software and help address usability issues.   Apparently, the majority view here and elsewhere seems to be that Phoenician is a distinctive script family. If so, then the only issues are those factual elements of Michaels proposal and there is no need to continue the discussion here of whether it is needed at all.   If �Phoenician� really is not a distinctive script, contrary to appearances, what we need is a fully reasoned argument for this and a proposal for how Phoenician should be treated as a variant of the Hebrew script in Unicode applications. I for one have little interest in more repetitive argument on this list. Is one of the proponents of this view prepared to produce a document for discussion before June 7th?   Bob Richmond Saqqara Technology