Here are my responses: Charles: (cpbs@ukc.ac.uk) A short dash is the same as a hyphen. Check on the greve accent typo. See revised format (included somewhere). No; Cartesian product is an oversized X, where the lines are at 90 degree angles to each other, and the character is usually depicted as being slightly larger than a normal capital letter. For \MATH codes, I'm afraid you'll have to check to make sure the brackets inside match up. But other than that, it can be skipped completely (for now). I've added a bit more to the description that will help to effect this design. Your example for using a forced line break doesn't require the line break. You say to use \| as a forced line break, and you give the following example: \jstl{500 Lime Ave.,\|Winklestreet,\|Codsworth.} But this is exactly the same as \jstl{500 Lime Ave.,} \jstl{Winklestreet,} \jstl{Codsworth.} Isn't it? (To be quite honest, I haven't looked at the format since I left school)... I don't see the need for a line break, since ending the current brace command does that automatically. The special effect \. creates the following situation: \jstl{500 Lime Ave.\.Winklestreet\.Codsworth} And would produce something similar to: 500 Lime Ave. Winklestreet Codsworth I like it. I added the * choice for list characters (see revised format). Who's next? Ah, Tobi... You're right, I missed the e on the ASCII fall back for a umlaut. Fixed. Missed o umlauts? You're right... They've been added. I added the ASCII code for the character, but I don't know any of the other codes... Next! Andrew Forrest... Hmm... Explicit footnoting shtuff. I understand what you mean, but I'm not sure about it. When I designed the format, I was kinda thinking of having the author put in their own footnotes, using the superscripts and such to fill them in themselves, as if they were standard footnotes from a typewriter (or word processor). Hmm. It'll take some thinking. It doesn't sound like a bad idea, and it has some hypertext type qualities to it (click on a footnote number and get a window that displays the footnote only when you need it). That could could be done without going too far out of the boundaries. What I'm thinking is this: Create, perhaps, %fxxxx in the article header (before the entry, and where xxxx is a number), and then have an escape code like \foot{xxxx} within the text to refer to it. Between the two %fxxxx lines would be something like a "mini article", using all the same text commands and such, to display the footnote. Unfortunately, I don't have any access to the net, or I'd be able to chat with you about this within a decent time period. Think about that idea, and get back to me (through Steve), and perhaps we could even clear it up before I get back to school... As for more style commands... I guess it's possible, but I think we should hold off on them for a while. The articles are supposed to be informative and entertaining, and the text should be enough to convey the idea properly. We're not looking to create perfectly formatted documents that would make most advertising agencies cry... Perhaps later? They should be pretty easy to implement after the fact. As for changing the article ID numbers. There are pros and cons to various ways. We can't conceivably go on the straight number system, as if two editors were to process articles at the same time (or nearly the same time), who gets dibs on what number? I think the ID system can stick for a little longer, at least. As for linewrapping, I'm not sure I understand your question. You ask if you should insert a space at the end when linewrapping; I'm not sure why this is necessary? If a word goes over the edge, so to say, take the whole word and put it on the next line. No spaces involved. Okay, who's next... Ah, Stephane Lussier... I remember you... :) Geez, are you always so long-winded? :) Honestly and seriously, though, I see just about all of your points. Yes, the format is in between compositional and render-itional, but I think that's because it needs to cover that kind of stuff. Theoretically, each article should look as the author wants it, and thus the bulk of it is render-itional. Most of the compositional qualities of the format are to allow rendering to happen due to the fact that almost everyone's terminal is going to be different. It's REALLY hard to keep a similar rendering design when you have no idea what kind of environment you'll have to render on. I think the format takes care of that, and I believe that was the initial point of the format. Believe me, putting that format together was a b**** and a half, and I must admit that it is quite convoluted. But I was forced to put together something that encompassed as much as everyone wanted. So everyone got about 50% of what they wanted, instead of 50% of everyone getting everything they wanted. It was a calculated risk, and I don't think we'll ever know if it was the "right" decision. (at least, not until we get someone, hopefully other than me, to draw up a new format all over again...) Andrew Forrest and Stephane Lussier's reply to Andrew Forrest... Okay, let's see, where to begin... We can easily keep the tokens down to four characters; I had this in mind when I wrote the format, and I had originally intended to keep it that way, but if you ask around, you'll find quite a few people who DON'T like the idea of fixed-length tokens. In fact, quite a few of them wanted things like \bold (bad example, but...) \italic, and \underline, because they're easier to write in from memory and bug-check when you're looking at it in an ASCII dump. \{ isn't needed (though I added it to the \MATH command to facilitate that) because { isn't the escape character, \ is. If you're parsing, and you come across a \, you set a bit and check the next character(s) for an escape code and if it requires a { you find it, and everything's hunky-dorey. But if you come across a {, and you're NOT looking for it, then accept it as a normal character. As for the accents applied to letters, instead of maintaining a huge list, you may have a point. Except that it's probably a lot easier to maintain a huge list, since it's easier to cross-reference the character with the appropriate extended character. For instance, in DOS a-umlaut is a few positions different on the ASCII chart than e-umlaut, and is quite a few positions from just plain a. It's a lot easier to have in the program, \ch"a means use character xxx than to program it to realize, okay, \umlt{a}, okay, a is character xxx, so on my list of umlaut combinations... Ah, character xxx. \ch"a takes up a bit more space to program, perhaps, but it saves a step in the translation of characters. As for abstraction, you're completely wrong (IMHO, of course). The Bold IS abstract, though the word isn't. Setting text to bold doesn't mean you have to configure your reader to display it bold, and, in fact, you often won't be able to, depending on your terminal. If you have a monochrome terminal, you might very well be able to get a boldface on it. Perhaps you have a computer that doesn't DO bold, you can change it to a different color, perhaps an intensified version of whatever is normal. So in all actuality, it is kinda abstract. As for article IDs and footnotes; I discussed them a tad earlier in this message. General message to all: If someone comes up with a new & improved and possibly completely different format, PLEASE send it to the net, and send it to ME! I'm not saying my conglomerated format design is the end-all, be-all of formats. But! I think once you get started on working on this, you'll keep finding new things that either don't work, or can't work in the same way, and you'll have a MAJOR headache on your hands. Trust me, I know, I put this beast together... If you can come up with a better format, I'm open to COMPLETE formats. Individual ideas are pretty useless, since, as I believe Andrew (possibly someone else) mentioned, you need a good base to start on, and then you can tack stuff on. Well that may be true, but I'm inclined to say otherwise. Believe it or not, THIS format started out that way, and it got messy, all by itself. ...Paul