From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.io/gmane.comp.tex.context/7557 Path: main.gmane.org!not-for-mail From: Idris Samawi Hamid Newsgroups: gmane.comp.tex.context Subject: Re: ConTeXt and critical editions Date: Sat, 06 Apr 2002 12:13:23 -0700 Sender: owner-ntg-context@let.uu.nl Message-ID: <0553OLKGOKSQB7TRIFSQ96FDHEDTNK.3caf48d3@c1909372-a> References: <5.1.0.14.1.20020330013315.02b5ccb0@server-1> NNTP-Posting-Host: coloc-standby.netfonds.no Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" X-Trace: main.gmane.org 1035398011 20175 80.91.224.250 (23 Oct 2002 18:33:31 GMT) X-Complaints-To: usenet@main.gmane.org NNTP-Posting-Date: Wed, 23 Oct 2002 18:33:31 +0000 (UTC) Cc: ntg-context@ntg.nl Original-To: Hans Hagen In-Reply-To: <5.1.0.14.1.20020330013315.02b5ccb0@server-1> Xref: main.gmane.org gmane.comp.tex.context:7557 X-Report-Spam: http://spam.gmane.org/gmane.comp.tex.context:7557 Dear folks, Here are some configuration options/ideas/dreams for an implementation/extension of EDMAC's critical edition capabilities in ConTeXt. I hope that some readers on the list may add other options/ideas as well. I am assuming the basic configurability offerred by EDMAC and described in Ch. 2 of edmac.dvi is understood and/or implemented; most of what follows is an extension of that. I apologize in advance for any ignorance of ConTeXt on my part; I'm still learning it's philosophy so some of my comments may sound uninformed. >>I'm working on this right now. This is coming a week late, sorry for the delay:-) The following is divided into four areas: lineation, apparatus, footnotes, and marginpar-endnotes-miscellany. ======================================= I. LINEATION 1. Consider an imaginary vertical line at the distance \linenumsep from the main body of text, the \linenumsep _line_. The line numbers are printed along this line. The area between the textblock and the \linenumsep line is the \linenumsep _range_. The default and EDMAC configuration (not configurable at a high level) is for the line numbers to be ragged to the outside of the \linenumsep range. So if the line numbers are to the right of the textblock, the numbers will be set flush left to the imaginary line and ragged right; if they r to the left of the textblock the numbers will be set flush right to the imaginary line and ragged left. I propose a total of three options: a) ragged to the _outside_ of the \linenumsep range as described above; b) ragged to the _intside_ of the \linenumsep range (so the line numbers will be between the textblock and \linenumsep line and flush with the \linenumsep line) c) centered to the \linenumsep line (so a line number like `123' will have the `1' inside the \linenumsep range, the `3' will be outside the \linenumsep range, and the `2' will be centered along the \linenumsep line.) 2. Sublineation in EDMAC takes a form like `3.6', where `3' is the line number and `6' is the subline number. I propose a parameter like \sublinepunctuation to make the period configurable. Some may prefer a comma or slash to a period (Arabic frequently uses a comma for the decimal point). This parameter should take both the desired font and the desired symbol as arguments. ======================================= II. APPARATUS. 1. In a critical footnote, given a footnote series in a multiple footnote layout, there is a 3-part sequence followed by the actual note: SIGN--LEMMA--SEPARATOR. I propose configuration options along the following lines: a) SIGN. The purpose of the sign is to identify the footnote with a string of text in the textblock. In EDMAC the default configuration (not configurable at a high level) is for the sign to be the range of line/subline/page numbers matched by the selected text, so that if the selected text is on lines 18 through 21, the sign will be `18--21'. This should be extended to three cases of signs: i) line/subline/page number range of selected text (already default in EDMAC); ii) markers (like asterisk, dagger, double dagger, section, pilcrow, etc). Given a footnote series, one should be able to define a sequence of markers (as in a \count register) to serve as the sign for that footnote series. Both the desired font and the desired symbol should be configurable; iii) both line/subline/page number range _and_ markers. So a note may begin with a marker, then the respective line/subline/page number range. The punctuation for the range indicator (en-dash, slask, or colon, etc) should be configurable, font and symbol. b) LEMMA. The purpose of the lemma is to abbreviate or otherwise modify the selected text which is passed to the note, as explained in edmac.dvi. If no \lemma{} command is issued, EDMAC just prints the selected text from the textblock. If a \lemma{}-like command is issued in ConTeXt, I propose something along the following lines: Let's divide the lemma into the following elements: # of initial words of selected text---joiner---# of final words of selected text. Given the \lemma of a footnote series, one should be able to configure these three elements, including the font and symbol of the joiner. The default should be first word--ellipsis--final word. That is, given a lemma command with no arguments, what should be passed to the note is the first and last words of the selected text connected by an ellipsis. c) SEPARATOR. The separator marks the end of information passed from the textblock and the beginning of the note proper. Both the desired font and the desired symbol for the separator should be configurable, as well as it's distance from the beginning of the note proper. =========================================== III. FOOTNOTE FORMATTING 1. Although EDMAC has a fixed nunber (5) series of footnotes available at any one time, it is not hard to add more series as John and Dominic point out. Can the number of footnote series be parameterized in the underlying code so that a user can set the number of series needed and have these automatically generated? 2. A very useful feature would be a footnote series identifier. When a footnote series starts on a given page, an identifier should be printed symbolizing the purpose of that series, e.g., `V' for variant. So a the footnote series for variants will start on any given page with `V', followed by the first SIGN--LEMMA--SEPARATOR sequence in that series; 3. It is a common feature of critical editions that if, in a given footnote series, two or more selections of text begin with the same line or have the same line-number-range, then instead of repeating the line number for each text selection in the footnote, a \parallel sign (called \notenumsep on p. 103 of edmac.dvi) is used instead of the line number range for the second and following notes that use the same line or line-number range (take a deep breath now). Rudimentary support for this is illustrated on page 103 of edmac.dvi (not part of EDMAC proper!). I've used this code and it has the following bug: if you have even a single note in a footnote series where the selected text is on the same line as those of _another_ footnote series where \parallel is called, then \parallel will also be called on that note (\linenum gives a workaround). Anyway, this is an important feature which should be supported. Both the desired font and the desired symbol for the \notenumsep should be configurable, of course:-) 4. For multi-column footnotes, an \intercolumnnoteglue analogous to the \interparanoteglue (p. 13 of edmac.dvi) should be implemented. This will control the width of columns and the distance between them. An error message or warning when that glue is too large with a suggestion for the maximum allowable glue would be nice as well (``\interparanoteglue (.4\hsize) too large. Should be .3275\hsize or smaller''). 5. The minimum allowed vskip between the last baseline in the textblock and the footnote rule should be configurable (Arabic needs larger values than Latin, for example). The inter-footer vskip between footnote series should be configurable as well; 6. The footnoterule itself should be configurable, both symbol and font. In fact, there should be two footnote rules: a) textblock to first footer rule; b) inter-footer rule; 7. Some mechanism for modifying the actual formatting of the footnotes should be provided. I am new to the ConTeXt philosophy so I'm not sure what advice to give here. Will the format macros be easliy accessible/readable to the user so he can redefine them as needed (EDMAC's macros are good in this respect)? Or will there be another mechanism? I'm talking about adding such things as hanging indents where e.g., the sign in the sign-lemma-separator sequence mentioned above is hung and so forth. 8. DREAM 1. I don't know if TeX is even capable of the following: Can we give a given footnote series line numbers and have another set of footnote series refer to the given one? Consider a critical edition of a text where one series of footnotes is used for a commentary on the main text. Now what if I want to add notes to that commentary keyed in to markers and/or line numbers in the margins of the commentary? Can it be done (I saw a sample from Yannis Haralambous' work whre he actually achieved such a feat but I think it involved post-dvi manipulation using perl)? Even if only line numbers for the footnotes can be implemented there may be some other workaround using cross-referencing to another regular series of footnotes to the main text. ============================================ MARGINPARS, ENDNOTES, AND MISCELLANY 1. Implementing critical apparatus configuration (as mentioned above) in the marginal notes would be quite useful. Again, I'm ignorant of much of Context's philosophy but I would like to be able to do things like configure marginpars to always be on the opposite side of the page as the line numbers and/or appear e.g., on the right side in odd pages and the even side in even pages. The following three options for marginpars would be nice (and I suspect they already exist in some form in ConTeXt): a) bottom line of the marginpar on same baseline as the line from which the note is called; b) bottom line of the marginpar on same baseline as the line from which the note is called; c)bottom and top lines of the marginpar equidistant to the same baseline as the line from which the note is called, with the middle line of the marginpar flush with the line from which the note is called if the number of lines is odd. Marginal notes work really nice in an Arabic context where word stretching allows marginpars to be fully justified. 2. The ConTeXt endnote mechanism should be extended a la section 2.9 of edmac.dvi (p. 18). 3. Does ConTeXt have a \centerline command that takes crops into account? Sometimes (e.g. chapter title pages) I want to center material with respect to, not \hsize of the textblock, but the width of the entire page. I defined a \cropcenterline for my purposes. Is there already a mechanism for this? 4. DREAM 2. Can ConTeXt do dual-language balancing for translations, etc.? ============================================== Well, that's it for now; I hope this is what u were looking for. Perhaps more suggestions from others on this list will follow. Please let me know if you need any clarifications, etc. I eagerly look forward to seeing how this all turns out. Thnx again 4 your kind and serious interest in all this! Best wishes Idris ========================== Dr. Idris Samawi Hamid Depatment of Philosophy Colorado State University Fort Collins, CO 80526