TeXhax Digest Tuesday, May 24, 1988 Volume 88 : Issue 50 [SCORE.STANFORD.EDU]TEXHAX50.88 Moderator: Malcolm Brown Today's Topics: Include-facility for dvi-files (TeXhax Digest V88 #47) Bibtex 0.99c change file for web2c (TeXhax Digest V88 #47) Let's tolerate famous men Re: Bibtex 0.99c change file for web2c Gargantuan Metafont Re: Lamport's recent answers funnies with \raggedright Placement of page numbers with TEX Standard set of Computer Modern. Lamport's role Check Sums Re: Lamport's recent answers TeXhax 42 & 44 Re: TeXhax Digest V88 #48 (LaTeX notes) A little extra space ? Lamport, gurus, thicker skin Re: \leaders and \hbox question RE: Using Postscript fonts RE: \leaders and \hbox question LaTeX arrowheads and vectors FTP Tenex vs Binary mode between TOPS-20 and Unix ---------------------------------------------------------------------- Date: Thu, 19 May 88 08:58:07 PDT From: mackay@june.cs.washington.edu (Pierre MacKay) Subject: Include-facility for dvi-files (TeXhax Digest V88 #47) It is a super idea, but it will take work. It will be necessary to 1) Associate the postamble font names from the including file with their assigned internal dvi font numbers. 2) Make up a list of the postamble font names from the included files so that the font numbers can be rewritten in them to match the numbers in the including file. The font define strings in the body of the dvi code will have to be checked to remove redundancy, and that will require making sure which is genuinely the first. It could be that in the combined dvi file the font-define in one of the included files was the appropriate one to leave in and the one in the including file must be removed. 3) Assign new font numbers to fonts declared only in included files, and correct all references in the included files to match the new assignments. 4) Normalize dimensions to account for possible magnifications and insure correct positioning. 5) Merge the bodies of the various dvi files, and rewrite the postamble. I think that indicates why it hasn't been done yet, but it could be done, and it would be a very useful utility. If anyone takes it on please do it in WEB with a WEB-to-C change file, rather than in naked C. There are some nice dvi-handling routines that can be extracted wholesale from the text of TEX.WEB to start with. Email: mackay@june.cs.washington.edu Pierre A. MacKay Smail: Northwest Computing Support Group TUG Site Coordinator for Lewis Hall, Mail Stop DW10 Unix-flavored TeX University of Washington Seattle, WA 98195 (206) 543-6259 ------------------------------ Date: Thu, 19 May 88 09:12:35 PDT Subject: Bibtex 0.99c change file for web2c (TeXhax Digest V88 #47) I remember that someone is doing work on this, but I can't lay my hands on the address. Maybe Tim Morgan will remember who. Ethan Munson has added a better path-finding to the BSD change file which is currently resident on ~ftp.pub at june.cs.washington.edu as one of three files named newbibtex* (Get all three.) While on the subject of web2c change files. let me urge again the production of METAFONTware change files. I am particularly interested in pktogf (recently released by Tom Rokicki on SCORE) because it would make it possible to send out fonts all in PK format. They are bursting the seams of the distribution right now. I have a BSD change file for pascal compilation of pktogf that I will happily make available to any interested party. Email: mackay@june.cs.washington.edu Pierre A. MacKay Smail: Northwest Computing Support Group TUG Site Coordinator for Lewis Hall, Mail Stop DW10 Unix-flavored TeX University of Washington Seattle, WA 98195 (206) 543-6259 ------------------------------ Date: Thu, 19 May 88 14:00:43 BST From: R Fairbairns Subject: Let's tolerate famous men While I have some sympathy with David Rogers (TeXhax #47) when he disagrees with Leslie Lamport's assertion that trivia shouldn't be submitted to TeXhax, I (equally) absolutely disagree that there should be constraints on the likes of Dr. Lamport's involvement in TeXhax. Dr. Lamport has performed great services to the community at large in the past, and we should welcome whatever pearls he's willing to let fall from his keyboard now. If he occasionally gets irritated by encountering questions again and again (and again...), well, so do I, and I don't have anything like the track record. I enjoy reading his opinions (and occasional fulminations). Those who don't can always skip 'em. A case in point was when he answered a question about sans-serif setting by saying ``don't use it if you want to remain legible'': we in ISO computer graphics have just had a brush with ISO central secretariat because their preferred sans-serif font makes some graphics standards unintelligible (not just illegible!). But Lamport didn't really answer the question (which is probably unanswerable anyway). Robin Fairbairns Laser Scan Labs Cambridge UK Janet: rf@uk.ac.cam.cl Internet: via uk.ac.ucl.cs.nss ------------------------------ From: ekrell@ulysses.att.com Date: Thu, 19 May 88 13:06:15 EDT Subject: Re: Bibtex 0.99c change file for web2c The latest web2c release from Tim Morgan includes a bibtex/ directory will all you need to compile bibtex using web2c. It's also in a separate file pub/bibtex.tar.Z on ics.uci.edu. ------------------------------ Date: Thu, 19 May 88 12:22:15 CST From: Robert Coleman Subject: Gargantuan Metafont I was relieved to see the change file for a gargantuan TeX in texhax v88i46. Thank you Pierre! I am sure that a similar file for a gargantuan METAFONT would also be generally appreciated. Robert Coleman Coleman@uregina2.bitnet ------------------------------ Date: Thu, 19 May 88 10:53:30 CDT From: William LeFebvre Subject: Re: Lamport's recent answers You seem to have missed the main point of Lamport's statement. He was saying that such novice questions should not be in TeXHax because there should be a local contact at every site that supports LaTeX. The idea being that a novice user can go to that contact first with a question. If the contact cannot answer the question, then they can go to the net with it. Lamport could be answering simple questions (and probably the same ones) for the next twenty years and still more would be coming in. If a user goes to a local contact first with a simple question, the question gets answered faster, less network bandwidth is used, and Lamport has more time to answer the really hard questions (like, "how does the tabular environment *really* work"). Everybody wins. After all, it is called "TeX-HAX" (the HAX is for "hackers"). > However, if he chooses to remain involved, then he must answer each and > every question: simple, stupid, intelligent, or interesting completely, > in detail and politely. Unbelivable! I can't possibly disagree more. Nothing says he "must answer each and every question". What's your problem? Give a guy a break, okay? Are you paying him to answer your questions? No! Is anyne paying him to answer any LaTeX questions? I don't think so! He is doing it out of the kindness of his heart. Since he chooses on his own to answer questions, he still has the right to refrain from answering any questions he does not feel like answering. I for one am glad that he answers the ones he does. If people like you keep badgering him, he may just take your first option and then where would we all be? William LeFebvre Department of Computer Science Rice University ------------------------------ Date: Thu, 19 May 88 11:39 EDT From: Dr Michael Soul Subject: funnies with \raggedright I've been trying to make slides using straight LaTeX (we don't have the fonts to make SliTeX work and anyway I don't fancy sans serif). I have \vspace*\vfill at the top, and \vfill and a \tiny footer line at the bottom so that the main text is vertically centred in the page. If the whole document is made \raggedright, the last line of every page (my footer line) completely vanishes. It's not the \tiny font, and its not the DEC LN03 laser (overflow comes out on a new page as expected. Taking the vfills away brings it back. Putting the \raggedright inside an environment that stops before the \vfill makes it comes out alright. SO... I have now got the output to look like I want it, but I would like to understand why. There seems to be a redefinition of \\, cos the vertical spacing in list environments with items ending in \\ changes with \raggedright. How come a horizontal mode thingy does nasty things to vertical mode? Michael Soul Schlumberger Technologies Instruments DIvision Farnborough UK We have TeX and LaTeX running on Vax/VMS and printing on LN03. ------------------------------ Date: Thu, 19 May 88 17:01:33 EDT From: bronwen Subject: Placement of page numbers with TEX Our thesis requirements state the following for the placement of page numbers: 1) The first page of every chapter or main division must be numbered in the center of the page, 5/8" from the bottom edge of the page. 2) All other pages are numbered in the upper right hand corner, 1/2" from the top of the page and 1/2" from the right-hand edge of the page (no left and right pages). This is on a page with 1-1/2" margins all the way around. Any suggestions as to how I would do this? I would prefer a TEX solution rather than a LATEX one. Thanks, bronwen@sbccvm.bitnet ------------------------------ Date: Thu, 19 May 88 14:47:25 EDT From: "Karl Berry." Subject: Standard set of Computer Modern. As I've mentioned before, John Sauter at DEC has produced Metafont files that can make any pointsize of any Computer Modern font. This means that magnifications can perhaps be done away with altogether. If you are doing reduction of magnified originals you will still want magsteps, but I doubt many sites have appropriate equipment. In fact, I suspect printing shops are about the only ones who do, and probably not all of those. If your operating system doesn't support links (a la ln or ln -s on Unix), then user's programs that ask for cmr10 scaled\magstep1 will fail. But if it does, cmr10.360gf can just be a link to cmr12.300gf. The checksums won't match, but the output will be right. I have a few Unix shell scripts that make creating them a bit easier. John has similar ones for VMS. Please write to me if you want this work. (I can also supply an lfonts.tex and sfonts.tex that uses the new fonts.) Karl. karl@umb.edu ...!harvard!umb!karl ------------------------------ Date: Thu, 19 May 88 14:35 EDT From: don%SIMA%atc.bendix.com@RELAY.CS.NET Subject: Lamport's role ------------------------------ Date: Fri, 20 May 88 10:42:44 GMT From: WSULIVAN%IRLEARN.BITNET@forsythe.stanford.edu Subject: Check Sums With the recent comments about checksums, perhaps someone could enlighten me regarding some not overly important cases I have come across. In the 300 pixel per inch AMS font distribution of July 1987 (PK format, on 5.25 discs) about a dozen of the fonts have checksum errors in a single byte, in a rather random way. Has anyone else noticed this? Apart from this, the fonts seem fine, but of course there could be errors of which one is unaware. The other case concerns the LaTeX line and circle fonts. We have different versions of these with different checksums, although the valid characters and character widths are exactly the same. I suspect this is due to a change in the way that checksums are computed, but I am not sure, as I do not know the algorithm used prior to MF 1.0. ------------------------------ Date: Fri, 20 May 1988 08:50:16 EST From: Bill Bormann Subject: Re: Lamport's recent answers Professor David F. Rogers writes: As the author of LaTeX, Lamport basically has two choices. The first is to, like Knuth, refuse to discuss LaTeX any more because he has better things to do. Fair enough, we all understand this position. However, if he chooses to remain involved, then he must answer each and every question: simple, stupid, intelligent, or interesting completely, in detail and politely. One man's viewpoint---I'd like to suggest another. I strongly disagree that Dr. Lamport has a professional obligation to make such a choice. From my perspective, LaTeX represents a gift to TeXers, and the fact that Dr. Lamport tolerates and responds to questions about LaTeX is really very generous. I think Dr. Lamport's answers to questions are right on target. TeXers who won't read the manual don't deserve answers that read like a dedication to a patron. Bill Bormann bormann@vm.cc.purdue.edu (Internet) bormann@purccvm (BITNET) ------------------------------ Date: Fri, 20 May 88 13:59 EST From: Subject: TeXhax 42 & 44 First, I'd like to thank those of you who responded to my inquiries regarding the [titlepage] option and \tableofcontents command. Prior to my posting the question, I read Section 4.1 of the LaTeX manual several times and perfectly understood how to create a table of contents. I realize that in processing my files, however, I erred. I now see that the \tableofcontents command can be placed anywhere in the text. Regarding my inquiry on the [titlepage] option, perhaps I should have asked "How can I number the footnotes differently?" rather than "correctly." As instructed, I did look in the index of the LaTeX manual under "numbering." However, the sections on numbering did not provide an answer to my question. I was fortunate to locate someone who was able to help me. ***** On another note, does anyone know how I can obtain early issues of the TeXhax Digest? (I'm interested in issues 1-15.) Many thanks, Josie colmenar@fordmurh colmenar@fordmulc P.S. It's "she." ------------------------------ Date: Fri, 20 May 88 18:17:23 PDT From: lamport@decwrl.dec.com (Leslie Lamport) Subject: Re: TeXhax Digest V88 #48 (LaTeX notes) Anita Hoover writes I needed to increase the spacing between lines for a document, so I set \renewcommand{\baselinestretch}{1.5}. I doubt that she reall "needs" to do this, but I will let that pass. She continues Most environments allow me to change back to single spacing by doing \renewcommand{\baselinestretch}{1} or explicitly setting the \baselineskip to something (say 13pt for 11pt fonts). My problem is that the tabbing environment will not change its spacing if I do either of these. First of all, as the manual explains, \renewcommand{\baselinestretch}{1} by itself does nothing. However, following it by something like \large\normalsize will work. Putting this sequence of commands before the tabbing environment will do exactly what she wants. Page 181 explains why putting this sequence of commands inside the tabbing environment does not work. Mark Steinberger writes The \makelabels command for the LaTeX letter style produces mailing labels of an unusual size: two labels across and five down on an 8.5x11 page. While this looks good, and offers some extra space for foreign addresses, the labels are hard to get and are expensive. Has anyone written a modification which will produce output for the standard sheet of labels (three across, and, I believe, eleven down)? It would be greatly appreciated, as these are the only labels which our department keeps in stock. It would be quite easy to modify the parameters in the standard style to produce two columns of labels of any desired size, since the labels are produced as a double-column page. Producing a three-column sheet would require more elaborate hacking. Eduardo Krell inquires about the kludge of using amcsc10 in lfonts.tex. As the latex.ins file explains, installers are expected to modify lfonts.tex to use the fonts available at their site. Now that there is a fairly standard set of TeX fonts, some day I will remove this kludge. A comment on John M. Jowett's style option for displaying cross-references. I suspect that it won't work if a \label command occurs inside the argument of a \caption command. (It may also have other bugs that I didn't notice on my quick inspection.) Leslie Lamport ------------------------------ Date: Fri, 20 May 88 17:06:41 PDT From: darrell%cs@ucsd.edu (Darrell Long) Subject: A little extra space ? Hi. I have the following macro to start a new chapter. I am finding that it puts a little space before the first character of the first line. I thought that perhaps there was a space somewhere causing it, but I don't see it. Any suggestions are welcome. Thanks. --DL \def\chapter#1#2{\hbox{} % this allows a skip at the beginning of the page \bigskip \centerline{{\mc Chapter #1}} \bigskip \centerline{{\bc #2}} \vskip 1.25 true in \par \noindent ------------------------------ Date: Fri, 20 May 88 15:10:08 PDT From: langdon@lll-lcc.llnl.gov (Bruce Langdon) Subject: Lamport, gurus, thicker skin In Texhax #44 says that every LaTeX installation should have a guru who can answer most questions and forward only the real ones (or something like that). There can be a problem with this. Until a few months ago, noone on our Sun network of ~15 people was using LaTeX. A few used TeX. Now we have many new users, and they use LaTeX. It is I who `keeps' TeX/LaTeX, bringing across the latest files from score for example. With the LaTeX manual and my knowledge of TeX, I can help solve most problems. But I am hardly likely to become a latex `guru'. I am paid to do physics and to make our work accessible to others. The other people have job descriptions too. I like powerful tools, so I'll do more for tex/latex than I need for just myself. But, we have no guru yet. And I may soon risk a dumn question of my own about framebox location. If LL, or anyone, helps, I'll appreciate that whether or not LL is curt. In texhax #47, D. Roger's suggests that LL must answer each and every question: simple, stupid, intelligent, or interesting completely, in detail and politely. Or refuse to discuss LaTeX any more because he has better things to do. I certainly hope these will not be the only options! I suggest two more: 1. LL could have someone else answer the easy/dumb questions. (I too can say "Let's you appoint a guru".) 2. Grow thicker skin on texhax readers. Bruce Langdon L-472 langdon@lll-lcc.llnl.gov Physics Department 339650%d@nmfecc.arpa Lawrence Livermore National Laboratory Livermore, CA 94550 (415) 422-5444 UUCP: ..{ihnp4,qantel,ucdavis,pyramid,harvard,topaz}!lll-lcc!langdon ------------------------------ From: Mario Wolczko Date: Sat,21 May 13:27:18 1988 Subject: Re: \leaders and \hbox question Trevor Darrell asks why, given the following macro, \def\entry#1#2#3{\hbox to 3 em {#1\hfill}#2\leaders\hbox{.}\hfill \hbox to 3 em {\hfill #3}} and the following invocations \centerline{{\mc Table of Contents}} \vskip 0.375 true in \rightline{Page} \vskip 0.25 true in \begingroup\obeylines \entry{}{Signature}{iii} % *1* ... \entry{I}{Introduction}{1} % *2* \endgroup that line *1* produces Signature .................................................... iii while line *2* produces I Introduction ................................................. 1 i.e, the "I" on a line by itself. The explanation is simple; within the \obeylines each newline invokes \par, which puts TeX in vertical mode. The first \hbox{} within entry is therefore added to the vertical list, TeX remains in vertical mode, and the text of parameter #2 starts a new paragraph, below the box. The fix is equally simple: place \leavevmode before the first \hbox in \entry. Mario ______ Dept. of Computer Science Internet: mario%ux.cs.man.ac.uk /~ ~\ The University USENET: mcvax!ukc!man.cs.ux!mario ( __ ) Manchester M13 9PL JANET: mario@uk.ac.man.cs.ux `-': :`-' U.K. Tel: +44-61-275 2000 extn 6146 ____; ;_____________the mushroom project____________________________________ ------------------------------ Date: Sat, 21 May 88 15:06 MST From: "That ain't no VAX; that's an 1840" Subject: RE: Using Postscript fonts Alex Woo asks about using Postscript fonts with LaTeX. This is not difficult, but it's hard to get all of the pieces together. We have a VAX/VMS system and Kellerman/Smith TeX. To this, we added the Arbortext (or whatever they're calling themselves this week) DVI-to-PS (and -HP) driver. Here are the issues: 1) Math mode. Some things use math mode. You have to have the right set of fonts IN ADDITION to the Postscript TFM files, because you're going to find people trying to use math mode all the time. 2) Slanted vs. Italic. Postscript fonts are SLANTED, not ITALIC. You need to make all references to italic fonts in the Roman family be changed to slanted. If you think you're going to mix them, forget it, it won't work. 3) PS vs. non-PS. We have both HP-flavored and Postscript-flavored laser printers. You can't mix pages in the same document; the fonts are RADICALLY different. I thought I couldn't tell the difference, but if you lay two pages together, you can easily see there are major differences. Beyond that, people concerned with page limits will love the Postscript fonts; they turn out to generate a few percentage points (5 to 10) less output than their CMR brethern. The votes in our department are Apple Laserwriter is better than anything, Dataproducts 2665 beats out HP Laserjet because of fonts and speed, and HP Laserjet is great, but the fonts aren't as good as either of the above Postscript printers. 4) fixing lfonts. If you use Arbortext, no problem. Copy lfonts somewhere and substitute for all cmr10 something like psmtimr at 10pt. Go through and fix them all, then use the virgin-TeX to generate a new LaTeX. Call it PSLaTeX if you want (that's what we do), and you're off. It took me a month to understand the interactions, but only a couple of hours to make it work once I figured out what was going on. We use a "SETUP TEX" command, in association with "closest printer information" to make the LaTeX command point to either HPLATEX or PSLATEX, along with the appropriate DVItoWhatever information. (Note: if you can't figure out to how fix lfonts, send mail and I can distribute a modified copy.) jms +----------------------------+ BITNET: jms@arizmis.BITNET |Joel M Snyder | Inter: jms@mis.arizona.edu |Univ of Arizona Dep't of MIS| Phone: 602.621.2748 ICBM: 32 13 N / 110 58 W |Tucson, AZ 85721 | Quote: "Design is everything. +----------------------------+ Implementation is trivial." ------------------------------ Date: Sat, 21 May 1988 22:51 EDT From: Jim Walker Subject: RE: \leaders and \hbox question Darrell Long writes: > I'm trying to format the table of contents for my dissertation, and have > written this simple macro: > > \def\entry#1#2#3{\hbox to 3em{#1\hfill}#2\leaders\hbox{.}\hfill > \hbox to 3em {\hfill #3}} > > When I give it my table of contents file, everything works well except > for the case where #1 is nonempty: > > \begingroup\obeylines . . . > \entry{}{Abstract}{x} > \vskip 0 true in > \entry{I}{Introduction}{1} > \endgroup . . . > I can see no reason why I get a newline following the "I" -- it just > doesn't make sense. \obeylines means that each end of line acts like \par. After \par, TeX is in vertical mode, ready for another paragraph. However, an \hbox is not recognized start of a new paragraph, it's just treated as a line by itself. (This happened on all lines, whether or not #1 is empty, but is invisible if #1 is empty.) One solution would be to put \noindent at the very beginning of the definition of \entry. Another would be to do away with \obeylines and use \line: \def\entry{\line{\hbox to 3em{#1\hfil}#2\leaders\hbox{.}\hfil \hbox to 3em{\hfil #3}}} -- Jim Walker, Dept. of Math, USC east ------------------------------ Date: Sat 21 May 88 22:05:10-MDT From: "Nelson H.F. Beebe" Subject: LaTeX arrowheads and vectors I discovered several months ago that the arrowheads in the LaTeX fonts (file lasy.mf) for new Metafont (in Web) have an incorrect origin; the ones computed with old Metafont (in SAIL) were correct. This bug results in incorrect positioning of the arrowhead on vectors produced by LaTeX. I corresponded with Leslie Lamport about this, and we both agreed this was a bug in lasy.mf, but, alas, neither of us has had time to fix it. Any Metafonters out there who volunteer for the job? In my view, The IMproper place to patch this is the LaTeX macros that generate the vectors. ------------------------------ Date: Sat 21 May 88 22:20:12-MDT From: "Nelson H.F. Beebe" Subject: FTP Tenex vs Binary mode between TOPS-20 and Unix In TeXhax 48, Dominick J. Samperi writes >> I just ftp-ed bibtex.web.1 from your system, and >> I don't understand why it is not in a readable format. I >> used tenex (whatever that is?), on the advice of another >> user. Would you please let me know what I did wrong, and >> how to properly transfer files. When FTP'ing between a Unix system and a TOPS-20 system, the rule is to use "ascii" mode for text files, and "tenex" mode for 8-bit binary files (like .dvi, .tfm, and font files). All Unix systems are on machines with 8-bit bytes, and most are also byte addressable (though Crays running UNICOS are not). TOPS-20 runs on a 36-bit word addressable machine, with byte instructions that permit byte sizes of 1 to 36 bits to be accessed with equal ease. Native TOPS-20 binary data naturally uses all 36 bits; FTP "binary" mode will result in the transfer of 2 36-bit words as 9 8-bit bytes, without loss of information. Non-native TOPS-20 binary data from byte addressable machines, or in the format used by such machines, is stored with 32 bits per word, 8 bits per byte; FTP "tenex" mode will handle it correctly. Normal ASCII text on TOPS-20 is stored as 7-bit bits (ASCII is, after all, a 128-character set), 5 per 36-bit word, with the rightmost bit unused. FTP "text" mode will handle it just fine. Between 8-bit byte Unix machines, either "binary" or "tenex" will work just fine for all data transfers. Thus, if you accidentally used "binary" mode to retrieve data from SCORE.STANFORD.EDU, SCIENCE.UTAH.EDU, or any other TOPS-20 machine, you have all the needed bits; you just have to put them back together in the correct order. The two programs bintnx.c and bintnxvms.c in APS: at SCIENCE.UTAH.EDU do the job. And where does "tenex" come from? Well, the original PDP-10 operating system was TOPS-10, and BBN's implementation of virtual memory was called TENEX; it evolved into TOPS-20. ------------------------------ %%% %%% Concerning subscriptions, address changes, unsubscribing: %%% BITNET: send a one-line mail message to LISTSERV@TAMVM1.BITNET: %%% SUBSCRIBE TEX-L % to subscribe %%% %%% All others: send mail to %%% texhax-request@score.stanford.edu %%% please send a valid arpanet address!! %%% %%% %%% All submissions to: texhax@score.stanford.edu %%% %%%\bye %%% ------------------------------ End of TeXhax Digest ************************** -------