UKTeX V88 #40 Friday 2 December 1988 AST Turbolaser (Not an ad, honest...) Change files/statistics Re: UKTeX V88 #39 TeX Users' Group membership and orders Advice Needed about DOSTEX Poolfile PostScript metrics Problem with double indices DVItoLN03 V2.1 Bug report (and fix!) uktex sbtex.arc DVIto Centronics PP8 Editor Peter Abbott Archive users will find from time to time files in subgroups not listed in 000aston.readme. These files are software which has been received and made available on an `as is' basis for anyone to collect. As soon as possible the software will be relocated in the correct place in the archive. Plans are being formulated to try and ensure each item is complete. Issues 41 will be prepared on Thursday December 8 and then despatched at 1700 GMT on the Friday. Likewise issue 42 the next week. Issue 43 (the last for 1988) will be sent late on December 22 or early December 23. V89 #01 will be despatched as normal January 6 1989 at 1700 GMT Latest TeXhax in the Archive is #104 (104 IS complete at Aston) Latest TeXmag in the Archive is V2N5 --------------------------------- Date: 25 Nov 88 22:29:10 gmt From: G.Toal @ uk.ac.edinburgh Subject: AST Turbolaser (Not an ad, honest...) To: info-tex@uk.ac.aston.mail Message-ID: <25 Nov 88 22:29:10 gmt 050760@EMAS-A> Well, alright - I lied... But at least its not an ad for my own gain: I've just been down to Morgan's Computer Store in London where they are selling AST Turbolaser/PS's. These are Postscript lasers, 8ppm, 35 Adobe fonts, 250 page paper tray and other goodies. (Serial, parallel and Appletalk interfaces spring to mind). These appear to be brand new unused and with guarantee. I've never seen them on sale anywhere before under \pounds 3200, so Morgan's price of \pounds 2300 seems like a genuine saving of \pounds 900. As these are ideal TeX printers, I thought I'd pass the news on. And no - I'm not after commission. (And yes - I did buy one myself). Graham. --------------------------------- From: Tim Bradshaw Date: Fri, 25 Nov 88 22:47:36 GMT Message-Id: <26051.8811252247@ms.maths.soton.ac.uk> To: uktex@uk.ac.aston Subject: Change files/statistics I have just installed TurboTeX on my machine (Unix V) as a replacement for Common TeX. The default version has the statistics stuff disabled: what do I put in the change file to enable them? Thanks in advance. - --Tim Bradshaw. --------------------------------- Date: Mon, 28 Nov 88 10:37:50 GMT From: Sebastian Rahtz To: ABBOTTP@uk.ac.aston Subject: Re: UKTeX V88 #39 > > At the Exeter conference, I remember hearing someone describe how > they had converted metafont to produce outline fonts. Not having > received a copy of the conference proceedings (nudge, nudge, Malcolm...) he was from Paris, Victor Ostromoukhov Phil Taylor points out that Arbortext have already done what I wanted with graphics languages to a certain degree: > >>> As an alternative approach, ArborText distributes separately an upward > >>> compatible set of extensions to LaTeX where the picture environment > >>> implementation has been changed to draw figures using PostScript via > >>> DVILASER's \special commands described [in this chapter]. Since PostScript > >>> can easily handle arbitrarily shaped figures composed of slanted lines and > >>> circles, as well as vastly more complicated figures, there is much more > >>> flexibility. I was aware of the ArborText stuff, in its PC incarnation a couple of years old; unless they have updated what they provide, I would comment that a) AT's stuff doesn't work for me, except when printing their examples. It would need hacking substantially for other dvi to PS drivers. Maybe others use it sucessfully, I just gave up after many iterations of hacking b) the LaTeX picture macros provide a horrid interface (who wants to specify lines like that?), and are strictly limited, ie they do not allow shading. If AT have added that, I apologize, but as I recall their extensions were fairly trivial. but I would agree that this is the right direction. > > But there, either Sebastian (I love Unix) has failed to read what > was actually written, or (but far less likely) is demonstrating a sad lack > of familiarity with WEB: what was actually requested was not simply a means the former, actually. I grovel. Sebastian Rahtz --------------------------------- Date: 28-NOV-1988 15:48:08 GMT From: CCZDAO@UK.AC.NOTT.VAX To: UKTeX@UK.AC.ASTON Subject: TeX Users' Group membership and orders Message-Id: <22000453_0016272C.0091C8ADE60AA5C0$10_2@UK.AC.NOTT.VAX> Originally-to: JANET"UKTeX@Aston",CCZDAO Originally-from:CCZDAO "David Osborne" Sender: JANET"CCZDAO@UK.AC.NOTT.VAX" X-Organization: Systems Group, Cripps Computing Centre, University of Nottingham X-Postal: University Park, Nottingham NG7 2RD, England X-Phone: (0602) 484848 ext 2064 i recently had a postcard from TUG reminding me to renew my membership and inviting me to pay by Visa or MasterCard. rather than return the postcard (so low-tech, these days!), i sent an e-mail message to Karen Butler with my Access/MasterCard number. that was at 11:59... Karen's reply arrived back at 13:38 to confirm the renewal. that's service! UK TUG members might like to know that you can order TUG publications by e-mail if you have a Visa card or MasterCard. i typed in a simulation of the order form for my last order and sent the mail to Karen. for mail other TUG people, see Barbara Beeton's "News from the Editor" in TUGboat 9 #1 (April 1988), page 7. the relevant addresses are: Karen Butler : membership records, subscriptions, orders; Ray Goucher : TUG management; Alan Wittbecker : TeXnical enquiries; Barbara Beeton : TUGboat, update info for TeX implementors. those mailing from JANET can get mail to AMS's "seed" system by reversing the domains and relaying through the EARN relay. for example, klb%com.ams.seed@uk.ac.earn-relay dave osborne, Cripps Computing Centre, University of Nottingham. (see, a real signature, Barbara!) --------------------------------- Received: from inset by kestrel.Ukc.AC.UK with UUCP id aa01335; 28 Nov 88 22:06 GMT Received: from sisl by inset with v7.12; Mon Nov 28 17:34:45 GMT 1988 Received: by sisl.uucp (smail2.5) id AA14894; 28 Nov 88 11:44:50 GMT (Mon) Organisation: Secure Information Systems Ltd. X-Mailer: Mail User's Shell (Vers 6.0) Sat Apr 2 19:36:07 PST 1988 To: info-tex@uk.ac.aston Subject: Advice Needed about DOSTEX Poolfile Message-Id: <8811281144.AA14892@sisl.uucp> Date: 28 Nov 88 11:44:48 GMT (Mon) From: Gavin Saunders Once more I find myself in need of expert advice. We are just getting started with TeX, and are in the process of porting TeX to our system (Thanks Don Ward for the tape conversion). I thought that in order to get familiar with system that we will end up with, that I would put up Dostex on a PC for people to use while the porting exercise is going on. Humm! I've downloaded all the .BOO from the texserver@spock.aston. Everything unpacks ok with the exception of POOLFILE.BOO. The arc file that I get from this will not unarc properly. Its not a transmission problem from aston, I have downloaded the file twice, and got the same thing both times. I suspect that the deboo step is causing the problem since the arc format keeps checksums. Unfortunately I can't get the deboo.exe downloaded from the aston archives (even with uuencoding etc), and so I have been using another version, which may (for some strange reason) be incompatable with this particular boo file. The only thing that is strange is that the version of deboo that I have managed to deboo all the other files ok!! Its all very frustrating, and I would welcome some help and/or advice. Specifically ( and in order of increasing effort) 1. Has anyone else had trouble with POOLFILE.BOO from the aston archives. Do you think that it is possible that my version of deboo could unpack all the boo files except this one ok? 2. If I need the deboo from aston, how can I get executables out of the archives. I am accessing by mail over uucp. (and yes I have RTFM, I dont seem to have the right voodoo to get the /encode option to work. 3. Is there another archive where I could download these files via kermit or something like that ( the name Lancaster comes up in some documentation) 4. Failing all of the above, is there some kind soul who has all these files who could put them onto a floppy for me. Thanks Gavin Saunders ( gns@sisl.uucp) - ------------------------------------------------------------------------------ Gavin Saunders gns@sisl.uucp Secure Information Systems Ltd +44 252 850011 Fax +44 252 851110 - ------------------------------------------------------------------------------ +++Editor - One of the planned developments is the distribution of requested files via diskettes and/or tapes but this service is not yet in place. I realise that this does not help Gavin but at least the problem is recognised and will be solved in the coming months.+++ --------------------------------- Date: 29-NOV-1988 10:17:10 GMT From: FPS@UK.AC.IMPERIAL.CC.VAXA To: INFO-TEX@UK.AC.ASTON.MAIL VorTeX: there is an article describing some aspects of the current status of VorTeX in `Computer' (the IEEE Computer Society's glossy) for January 1988 - vol 21, no 1. Chelgraph's rip and CM besides Lance Carnes implementation of dvi-to-IBX, there is reputedly a UK version. i have found it singularly difficult to communicate with Chelgraph's agents on this, since they don't know what TeX is. perhaps tackling Chelgraph directly would be a good route? MacBibTeX: i have just sent off for this (as featured last month in TeXhax). i will be happy to copy it for anyone in the UK to save time/effort at this and the US end. TeXhax: i find that a few go astray. is this my dodgy mailer, or is there another reason? malcolm clark +++Editor - It seems to me you have a dodgy mailer - there was a delay between 99 and 100 due to the site change. Also your mailer does not include a subject field (hint, hint) although you are not the only culprit. +++ --------------------------------- Received: Date: Tue, 29 Nov 88 10:51:02 GMT From: Sebastian Rahtz To: info-tex@uk.ac.aston.mail Cc: text@uk.ac.soton.cm Subject: PostScript metrics I have been having a short correspondence with James Clark about his very interesting DVITOPS program, and it caused me to investigate the state of Adobe AFM to TeX TFM conversion. I looked at Palatino Roman, and found the following .pl files on our system and at Aston: - one generated here - one generated here by Clayton Elwell's AFTOPL - Aston [.apple.pl] - Aston [.pslatex] - Aston [.dvi2adobe] (in [.unix]) - James Clark's It will amuse you to know that no two are the same! Not even in the number of lines in the file. The differences are of three types: - rounding procedure (.458 vs .4655) - ligature choices - choice of values not supplied by Adobe The ligatures differ depending on where in a font various characters are assumed to be. Here, for instance, Les Carr keeps the `fi' ligature at position 14 (don't worry if this isn't right!) for the purpose of the TFM file, and rejiggles the font encoding when he outputs PostScript to move the character into that position. James Clark adopts the opposite tack, and leaves the `fi' where it is in Abobe fonts (much later on), and redefines some common TeX macros (like \ae) so that they pick up the correct Adobe character. Which is the `right' approach is open to question (Clark's is `PostScript purist', Carr's is `TeX purist', perhaps), but the important thing is that the two are incompatible for actually printing TeX documents. Thus when I asked Clark for his program when I saw the note on TeXhax, and he asked if I had PostScript .tfm files, my answer of `yes, no trouble' was misleading, as when I compiled his program it generated garbage for me until I built myself a whole new LaTeX with different TFMs. That problem is solvable. Rather more worrying is the variation in some global values in the different versions of Palatino: Design Slant Space Stretch Shrink Xheight Quad Extra Soton 10 0.0 0.3333 0.16666 0.1111 0.468 1.0 0.111 Elwell 10 0.0 0.25 0.3 0.1 0.468 0.694 0.125 .APPLE 10 0.0 0.3333 0.16666 0.1111 0.468 1.0 0.111 .PSLATEX 10 0.0 0.275 0.1375 0.0916 0.4655 1.0 0.0916 Clark 1 0.0 0.25 0.125 0.083 0.468 1.0 0.083 Note that the Elwell was run on a TFM file taken from the PostScript interpreter itself. Do these slight variations matter? yes, they do! I set a test page using the Soton values, and again with Clarke's values, and found that in one case a whole extra long word was fitted on a line by the tighter fitting of the Clarke metrics, and other lines showed similar changes. The differences are quite clear to the naked eye (I wouldn't like to say which was `nicer'!), and disturbing. These remarks are not in the least exhaustive, but I offer the following questions to those with the time, energy and expertise to pursue the matter further: a) Is it possible that the Palatino metrics have changed between releases of PostScript? b) Are the parameters which vary _supposed_ to depend on preference? Is there loose Palatino and tight Palatino? c) What do other people think about how the relationship between an Adobe font layout and a TeX font layout should be managed? d) Why do we have at least 4 sets of metrics for the same font floating around? Do we want to agree on a common set? Sebastian Rahtz PS Just to depress myself further, I spent half an hour setting the opening page of Moby Dick with TeX (with various TFMs), troff, MacWriter and Microsoft Word. Yes, you guessed, no two are the same, despite my setting the same line length and using Palatino 10 pt throughout. Such is life. +++Editor - From my point of view the less duplication the better so if the respective guru's can agree which one to use then I will gladly delete the other files. +++ --------------------------------- Received: from QQ47@UK.AC.LIVERPOOL.IBM by ISMAIL(2.1.18); 1 Dec 1988 16:14:04 GMT Date: Thu, 01 Dec 88 16:10:45 GMT From: David Lloyd Subject: Problem with double indices To: Peter Abbott I found something odd when I tried to use an expression like $$x^{i^{2}}$$ A vertical bar grows out of the top of the 2. Could this be a problem with the font? --------------------------------- Date: 1-DEC-1988 17:41:13 GMT From: RM028A@UK.AC.CRANFIELD.CDVA To: info-tex@UK.AC.ASTON Subject: DVItoLN03 V2.1 Bug report (and fix!) Thanks to Phil Taylor, who managed to create a .DVI file with so many characters that it need to download more than ten LN03 fonts, I have discovered a bug in DVItoLN03, which caused it to try to image the eleventh and successive fonts using whatever was loaded as the tenth font! It must have been a moment of complete mental abberation when I wrote the item of code, because it was so obviously wrong! However, his file has also shown up what appears to be a more serious problem, that the LN03-plus (and presumably the ordinary LN03) don't actually work properly with the 15th (and successive) downloaded fonts (each of up to 188 characters). It is perhaps significant that if one prints a summary sheet after downloading 15 or more fonts, only the first fourteen get listed! Trying to print characters from these fonts results in lots of white space and error 24 on the printer (character not in font). I have accordingly imposed a limitation in the program such that it now refuses to work if the .DVI file calls for more than fourteen fonts to be downloaded --- if you get this message, just process the document in smaller chunks, using the /STARTING_PAGE and /NUMBER_OF_PAGES qualifiers. Until I get around to incorporating the changes (with some enhancements) into V2.2, herewith a change file which should allow you to build a corrected DVItoLN03. - ------------------------Cut here--------------------------------------- VAX/VMS Change file for DVItoLN03. Modification History ~~~~~~~~~~~~~~~~~~~~ 01-DEC-1988 BHK Corrected bug which imaged WRONG font out of those above first ten 01-DEC-1988 BHK Decreased max_lnfonts to guard against LN03 limitations @x <<<<< Added on 01-DEC-1988 by BHK @ RMCS >>>>> @d max_lnfonts=21 @y @d max_lnfonts=13 @z @x <<<<< Added on 1-DEC-1988 by BHK @ RMCS >>>>> esc_out(dcs+'1;'+str_int(10+max_SGR)+'}U0000'+@=dec@>(max_SGR,2,2)+ '002SK00GG'+st) @y esc_out(dcs+'1;'+str_int(10+max_SGR)+'}U0000'+@=dec@>(ln_font,2,2)+ '002SK00GG'+st) @z - ---------------------------and here---------------------------------- If anyone out there is using an ordinary LN03 instead of the LN03-plus, please contact me so that I can send you a .TEX source which exposes the hardware limitation --- also enyone who got an LN03-plus actually built as such from the factory could cooperate on this; all the LN03-pluses I have access to were either field upgraded, or delivered as the ordinary LN03 and the upgrade kit, which amounts to the same thing! Brian HAMILTON KELLY .-----------------------------------------------------------------------------. | Snail Mail: | Telephone: Swindon (0793) 785252 | | School of Electrical | International: +44-793-785252 | | Engineering & Science | | | Royal Military College of Science | e-mail: | | Shrivenham, SWINDON | Janet: rmcs-tex@uk.ac.cranfield.cdvc | | SN6 8LA | | | United Kingdom | | `-----------------------------------------------------------------------------' TIMESTAMP: This message was queued for delivery on 1-DEC-1988 at 17:37 GMT. --------------------------------- Received: from harwell by kestrel.Ukc.AC.UK with UUCP id aa18293; 2 Dec 88 11:16 GMT To: uktex-request@uk.ac.aston.mail Subject: uktex Date: Fri Dec 2 11:14:43 1988 From: judy@uucp.harwell 2. Can you tell me how to get hold of the latest tex distribution suitable for running on a Sun-4 under SunOS 4.0 WITHOUT pascal ? I have Tex 2.1 running on a Sun-3, but my colleagues have not bought their Sun in time to be eligible for the free 4.0 pascal. Thanks, Judy Booth, Computer Science & Systems Division, Harwell Laboratory, Didcot, OXON. --------------------------------- From: Wujastyk (on GEC 4190 Rim-A at UCL) To: abbottp@UK.AC.ASTON Date: Fri, 2 Dec 88 10:33 Subject: sbtex.arc Message-Id: <02 DEC 1988 10:38:55 UCGADKW@UK.AC.UCL.EUCLID> I have repeatedly tried to get [PUBLIC.PC]SBTEX.ARC by FTP (FETCH BINARY) but without success. I have no trouble with other files, including binaries (TFMs). Anything you can suggest? While I'm at it, I also wanted the latest version of PCWRITEX. It's in CLARKSON as a .SHAR file. I have a couple of Unshar programs on my PC but they all failed to unpack it completely. I suspect this is because it has such a lot of 0-32 and 128-255 characters in it, and the PC versions of these Unix utilities are not wholly robust. It seems perverse to distribute a PC utility in Unix format anyway! Could it be unshared by a pukka Unix site, and repacked using ARC, BOO, UUencode, or something like that? Dominik - ----------------------------------------------------------------------------- Dominik Wujastyk, | Janet: wujastyk@uk.ac.ucl.euclid Wellcome Institute for | Bitnet/Earn/Ean/Uucp: wujastyk@euclid.ucl.ac.uk the History of Medicine, | Internet/Arpa/Csnet: dow@wjh12.harvard.edu 183 Euston Road, | London NW1 2BP, England. | Phone: London 387-4477 ext.3013 - ----------------------------------------------------------------------------- +++Editor - The file I collected was transferred using binary in every case. I am sure that someone will repackage it for the archive. The clarkson collection is stored in [public.clarkson] and Michael DeCorte reads this digest so no doubt a copy will be sent shortly. +++ --------------------------------- Date: 2-DEC-1988 11:35:13 GMT From: PHYOKANE@CS8700.UCG.IE To: info-tex@UK.AC.ASTON Subject: DVIto Centronics PP8 sender: hea"phyokane@cs8700.ucg.ie" My department (Experimental Physics) has just purchased a Centronics PP8 laser printer. It's print quality is very good ( compared with a DEC LN03 ) and it comes at a very nice price (#995 with serial interface). The bad news is that the HP Laserjet plus emulation has problems. packages that download entire fonts before printing ( Ventura Publisher ) work fine. Attempting to download a character to the current font, or possibly to any font used on the current page, results in corrupting that font. This means that Beebe's DVIJEP driver does not work. I spent most of this week just finding this out. The PRELIMINARY Emulator interface manual shipped with the printer is incomplete, self contradictory in places and just plain wrong in others. The technical support from the suppliers is zero ( after 10 years experience in both hardware and software I find the suggestion that I have handshake problems in the serial interface less than useful when I ask a specific question like "How does the download character support differ from a genuine HP Laserjet?" ). Has anyone out there got a working DVItoPP8 ? or is trying to make one ? The problem ( and maybe the solution ??) looks like that reported with the Brother HL-8. I am going to try the approach ( like the Flavio Rose LN03 driver) of making two passes over the dvi file and then downloading only the used characters before printing the text. Finally... If I had a MC68000 disassembler that ran on VAX/VMS or a BBC micro or an IBM clone, I might fall to the temptation to reverse engineer an accurate emulation manual from the ROMs in the machine. Does anyone know of a free one ? Peter O'Kane --------------------------------- !! !! Files of interest [public]000aston.readme !! [public]000directory.list !! [public]000directory_dates.list !! [public]000directory.size !! [public]000last30days.files !! !! Editor - I have a tape labelled TeX 2.9 LaTeX 2.09 Metafont 1.3 !! Unix 4.2/3BSD VAX SUN 2/3 Pyramid Sequent SYS V: 3B2 Tar 1600 bpi blocked !! 20 1 file dated 26 may 1988 (from washington.edu). - Later one on its !! way. !! I have the facility to copy this tape for anyone who sends the following !! 1 2400 tape with return labels AND RETURN postage. !! !! Send to !! !! P Abbott !! Computing Service !! Aston University !! Aston Triangle !! Birmingham B4 7ET !! !! A VMS backup of the archive requires 2 (two ) 2400' tapes at 6250bpi. !! Remaining details as above. !! !! Replies/submissions to info-tex@uk.ac.aston please !! distribution changes to info-tex-request@uk.ac.aston please !! !! end of issue