Michael, I think you should view the character within the "n" attribute as a part of the text AND a way to override the default behavior. I think that overriding the default with nothing and then putting the quote mark in the text itself is an unfortunate direction and makes the quote mark no more or less a part of the text.
How is encoding <q n='"' sID="uniqueID"/>text<q n='"' eID="uniqueID"/> making the quote marks any less a part of the text as compaired with <q n='' sID="uniqueID"/>"text"<q n='' eID="uniqueID"/>. Facts: 1) The majority of encoders WILL NEITHER encode n='"' NOR include the quote mark in the text itself 2) The encoding of quotes using <q> is strongly encouraged 3) The default case for rendering a <q> element for MOST ENCODINGS AND RENDERING PROCESSES will almost certainly be to add a starting and ending quote mark and the better process will process the "n" attribute (if it is adopted as a standard process) 4) You are more likely to get the results you want by NOT putting the quote mark in the text itself. Todd > -----Original Message----- > From: Michael Paul Johnson [mailto:[EMAIL PROTECTED] > Sent: Thursday, March 18, 2004 5:17 PM > To: Todd Tillinghast; 'Patrick Durusau' > Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] > Subject: RE: [Fwd: [sword-devel] Re: [osis-editors] OSIS 2.0.1 modules > updated] > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > At 09:06 19-03-04, Todd Tillinghast wrote: > >Michael, > > > >I think you will find that in the majority of cases that you will be > >using <q> milestone containers rather than <q> containers because > >quotes > >often cross other hierarchies, so the use of milestone exclusively is > >not much of a loss anyway. > > Agreed. > > >However, I would still recommend AGAINST <q n="" sID="UniqueID" > >who="Jesus"> to cause there to be no quotation marks for red letter > >Bibles. The reasoning being that this IS truly a rendering issue. > >If > >you render the same OSIS document without red letters you will render > >with no quote marks or at least have to check to see if the were left > >off because the who is Jesus. It makes more sense to include the > >quote > >marks as the default and then IF the rendering process uses red > >lettering when it sees who="Jesus" THEN drop the quote marks. The > >point > >being that the same OSIS document should be able to render a red > >letter > >edition as well as a all black letter edition, by using the n="" for > >who="Jesus" cases you are LOOSING information. > > WHY do you think I would be losing information? If I mark the > quotations of Jesus as I said, using something like <q who="Jesus" > n="" sID="uniqueid">"actual quotation here inside of real quotation > marks of the kind appropriate to this language and translation"<q n="" > eID="uniqueid">, then you always have the correct punctuation, with no > loss of information, regardless of if you are creating a "red letter > edition" or a "black letter edition." There seems to be no good reason > for your recommendation. Am I missing something? > > I NEVER want ANY use of <q n=""> elements to result in generation of > punctuation of any sort, including quotation marks. I intend to ALWAYS > include the proper punctuation in the text itself, and always use the > n="" parameter, except in one case (described below). IF you generate > quotation marks from the XML markup that says not to, then YOU > INTRODUCE ERRORS TO THE TEXT. If you honor the n="" attribute as you > described it, then there are NO ERRORS INTRODUCED AND NO LOSS OF > INFORMATION. > > Let me put it another way. You seem to think that it is a good feature > to have XML markup generate quotation marks. I believe such behavior > is very bad most of the time. It is not a feature. It is a bug. I > understand that you have been taught to think this way from your > ethnocentric view of the English-speaking universe, but I see things > much differently from my cross-cultural and multi-lingual experience. > > Am I that unclear in what I am saying? > > I want to totally separate two issues: (1) placement of quotation > marks and what exact characters are in use, and (2) marking of Jesus' > words only for the purpose of giving people a choice of rendering them > differently (maybe with red letters) if they so choose. > > For issue (1), I don't want to be forced to use XML markup to generate > punctuation, especially underspecified XML markup like you propose. > Punctuation is part of a language, just like the letters, syllabary, > or pictograms. Why not just represent it in Unicode? Keep it simple, > sir. The advantages of doing so far outweigh the advantages of > converting to and from XML markup. > > For issue (2), all I want to do is make it easy to give an option to > people to render Jesus' direct quotes in some other way or not, > WITHOUT CONFUSING THIS ISSUE WITH PUNCTUATION GENERATION AS PART OF > THE RENDITION PROCESS. > > >For other cases where the quote marks were left off for other reasons > >then I would use n="". > > I intend to use n="" for every use of <q ...> or <speech ...>, except > one, because I intend to treat all punctuation as part of the text, > not the markup. The only case where I would use <q> or <speech> as you > envision using it would be the case where someone is drafting a new > translation, and they want to use an automated process to generate the > punctuation. HOWEVER, once that punctuation is generated and the > translation is finalized, then all of the markup should be converted > to proper punctuation within the text itself and replacing the n > parameter in q or speech with "". That way, the text and style > rendering information can be kept separate. > > > Rev. Michael Paul Johnson > Servant of Jesus Christ > [EMAIL PROTECTED] > http://eBible.org/mpj/ > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.4 (MingW32) > Comment: http://eBible.org/mpj/gpg.htm > > iD8DBQFAWjv0RI/gxxfXR7sRAmRcAKCDlWmIoXRYZYXyplHXMhaa6Xpn1QCgjThC > 9xk8n/WiENEIeoG0PJMmHAI= > =pLDO > -----END PGP SIGNATURE----- _______________________________________________ sword-devel mailing list [EMAIL PROTECTED] http://www.crosswire.org/mailman/listinfo/sword-devel
