From clemc at ccc.com Sat Jan 1 05:34:07 2022 From: clemc at ccc.com (Clem Cole) Date: Fri, 31 Dec 2021 14:34:07 -0500 Subject: [COFF] on progress -- was [TUHS] moving directories in svr2 Message-ID: Moving to COFF since while this is a UNIX issue its really attitude, experience and perspective. On Thu, Dec 30, 2021 at 8:01 PM Rob Pike wrote: > Grumpy hat on. > > Sometimes the Unix community suffers from the twin attitudes of a) > believing if it can't be done perfectly, any improvement shouldn't be > attempted at all and b) it's already done as well as is possible anyway. > > I disagree with both of these positions, obviously, but have given up > pushing against them. > > We're in the 6th decade of Unix and we still suffer from unintended, > fixable consequences of decisions made long long ago. > > Grumpy hat off. > > -rob > While I often agree with you and am a huge fan of your work both written and programming, I am going to take a different position: I am very much into researching different solutions and love exploring them and seeing how to apply the lessons, but *just because we can change a change*, *does not always mean we should*. IMOI: *Economics has to play into equation*. I offer the IPv4 to IPv6 fiasco as an example of a change because we could (and we thought it would help - hey I did in the early 1990s), but it failed for economic reasons. In the end, any real change has to take into account some level of economics. The examples of the differences in the shell is actually a different issue -- that was territorial and not economics -- each vendor adding stuff that helped them (and drove IVS/end users of multiple platforms crazy). The reality with SunOS sh vs Ultrix sh vs HP-UX sh vs System V (att sh) was yet another similar but different -- every manufacturer messed with a V7 derivative sh was a little different -- including AT&T, Korn et al. For that matter you (Rob) created a new syntax command with Plan9 [although you did not try to be and never claimed to be V7 compatible -- to your point you break things where you thought it matters and as a researcher I accept that]. But because all the manufacturers were a little different, it was exactly why IEEE said -- wait a minute -- let's define a base syntax which will work everywhere and it is something we can all agree and if we all support it -- great. We did that, and we call that POSIX (and because it was designed by compromise and committee - like a camel it has some humps). *But that does mean compromise -- some agreed 'sh' basics needs to keep the base level.* The problem Ted and Larry describes is real ... research *vs.* production. So it begs the question, at what time does it make it sensible/ (worth it/economically viable) to move on? Apple famously breaks things and it drives me bonkers because many (most I would suggest) of those changes are hardly worth it -- be it my iPhone or my Mac. I just want to use the darned thing BTW: Last week, the clowns at Telsa just rolled out a new UI for my Model S --- ugh -- because they could (now I'm fumbling trying deal with the climate system or the radio -- it would not do bad if they had rolled out a the new UI on a simulator for my iPad so I could at least get used to it -- but I'm having to learn it live -- what a PITA -- that really makes me grumpy). What I ask for this august body to consider is that before we start looking at these changes is to ask what we are really getting in return when a new implementation breaks something that worked before. *e.g.* I did not think systemd bought end users much value able, must like IPv6 in practice, it was thought to solve many problems, but did not buy that much and has caused (continues to cause) many more. In biolog every so often we have an "ice age" and kill a few things off and get to start over. That rarely happens in technology, except when a real Christianen style disruption takes place -- which is based on economics -- a new market values the new idea and the old market dies off. I believe that from the batch/mainframe 1960s/early 70s world, Unix was just that -- but we got to start over because the economics of 'open systems' and the >>IP<< being 'freely available' [which compared to VMS and other really proprietary systems] did kill them off. I also think that the economics of completely free (Linux) ended up killing the custom Unix diversions. Frankly, if (at the beginning) Plan9 has been a tad easier/cheaper/more economical for >>everyone<< in the community obtain (unlike original Unix release time, Plan9 was not the same rules because AT&T was under different rules and HW cost rules had changed things), it >>might<< have been the strong strain that killed off the old. If IPv6 has been (in practice) cheaper to use than IPv4 [which is what I personally thought the ISP would do with it - since it had been designed to help them] and not made as a premium feature (i.e they had made it economically to change), it might have killed of IPv4. Look at 7 decades of Programming Language design, just being 'better' is not good enough. As I have said here and many other places, the reality is that Fortran still pays the salary of people like me in the HPC area [and I don't see Julia or for that matter, my own company's pretty flower - Data Parallel C++ making inroads soon]. It's possible that Rust as a system programming language >>might<< prove economical to replace C. I personally hope Go makes the inroads to replace C++ in user space. But for either to do that, there has to be an economical reason - no brainer style for management. What got us here was a discussion of the original implementation of directory files, WRT links and how paths are traversed. The basic argument comes from issues with how and when objects are named. Rob, I agree with you, that just because UNIX (or any other system) used a scheme previously does not make the end-all. And I do believe that rethinking some of the choices made 5-6 decades ago is in order. But I ask the analysis of the new verse the old takes into account, how to mitigate the damage done. If its economics prove valuable, the evolution to using it will allow a stronger strain to take over, but just because something new vs. the old, does not make it valuable. Respectfully .... Happy new year everyone and hopefully 2022 proves a positive time for all of you. Clem -------------- next part -------------- An HTML attachment was scrubbed... URL: From paul at rileyriot.com Mon Jan 10 15:55:36 2022 From: paul at rileyriot.com (Paul Riley) Date: Mon, 10 Jan 2022 16:55:36 +1100 Subject: [COFF] Bootstrap for Sykes 7000 Twin Floppy Drive Unit Message-ID: I think I've posted this question before, perhaps on TUHS, but I'll ask again. I have a PDP-11/03 with a Sykes Twin 8" Floppy Drive unit. It has it's own controller card, so I'm not sure if it's RX01/RX02 compatible. Problem is, I need a bootstrap program for it. I can't find a technical manual for it, so I'm stuck. I see the source for LSX has a driver for the Sykes, so I may be able to install and mount it on MX, which I'm preparing with Noel's help for my machine, without booting from it. I'm hoping that Heinz, or someone who had a Sykes drive in that era still has the bootstrap code. Paul *Paul Riley* -------------- next part -------------- An HTML attachment was scrubbed... URL: From athornton at gmail.com Tue Jan 11 13:22:49 2022 From: athornton at gmail.com (Adam Thornton) Date: Mon, 10 Jan 2022 20:22:49 -0700 Subject: [COFF] [TUHS] Demise of TeX and groff (was: roff(7)) In-Reply-To: <4409b91cd794867d@orthanc.ca> References: <20211231234039.GU31637@mcvoy.com> <20220101005605.GL75481@eureka.lemis.com> <20220101031511.GB8135@mcvoy.com> <20220111015901.GE25103@eureka.lemis.com> <4409b91cd794867d@orthanc.ca> Message-ID: <8BCC2106-67EA-40D7-9C32-523FD7D7A3C1@gmail.com> Taking this to COFF... > On Jan 10, 2022, at 7:13 PM, Lyndon Nerenberg (VE7TFX/VE6BBM) wrote: > > Greg 'groggy' Lehey writes: > >> As long as man pages are formatted with ?roff, I don't see it going >> away. I don't suppose many people use troff any more, but there are >> enough of us, and as long as man pages stay the way they are, I don't >> think we're in any danger. > > Well there is mandoc(1). But as time goes by they just seem to be > re-implementing nroff. Of course that *must* be easier than just > learning n/troff in the first place :-P As someone who did a lot of a Ph.D. in the history of computing, and then went into IT because he liked eating protein sometimes: The great secret is that NO ONE EVER READS THE LITERATURE. We have now made all the mistakes at least four times: Once for each of mainframes, minis, micros, and mobile. You can be a rock star at any development or operations job, even if you are, like me, a Bear Of Little Brain, simply by having some idea of what was tried already to solve a problem like this, and why it didn't work. Which you can get by actually stopping to read up about your problem before diving headfirst into coding up a solution for it. If you happen to get stinking rich from this advice, you can buy me a bottle of whiskey sometime. Adam From athornton at gmail.com Thu Jan 13 11:34:09 2022 From: athornton at gmail.com (Adam Thornton) Date: Wed, 12 Jan 2022 18:34:09 -0700 Subject: [COFF] [TUHS] Coastal cultures, collaboration, creativity and Sun vs DEC. In-Reply-To: <20220112231509.GJ61872@eureka.lemis.com> References: <20220112231509.GJ61872@eureka.lemis.com> Message-ID: Moving to COFF, but Brian Dear's "The Friendly Orange Glow", about Plato, talks a lot about some of the cool stuff happening in the middle of the country. https://www.amazon.com/Friendly-Orange-Glow-Untold-Cyberculture/dp/1101973633/ref=tmm_pap_swatch_0?_encoding=UTF8&qid=&sr= And later, of course, NCSA Mosaic. On Wed, Jan 12, 2022 at 4:15 PM Greg 'groggy' Lehey wrote: > On Tuesday, 11 January 2022 at 14:34:16 -0500, John Cowan wrote: > > On Tue, Jan 11, 2022 at 1:37 PM Dan Cross wrote: > >> It seems like Unix is largely a child of the coasts. > > > > We can add the eastern coast of Australia, where the original > > Wollongong group made the first V6 port to the Interdata 7/32 (not > > to be confused with the Labs port to the 8/32). > > To be fair, in the case of Australia almost everybody is on the east > coast, though we have had our share of FreeBSD core team members from > the "west coast" (which is really only Perth). > > Greg > -- > Sent from my desktop computer. > Finger grog at lemis.com for PGP public key. > See complete headers for address and phone numbers. > This message is digitally signed. If your Microsoft mail program > reports problems, please read http://lemis.com/broken-MUA.php > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joshnatis0 at gmail.com Thu Jan 13 12:04:32 2022 From: joshnatis0 at gmail.com (josh) Date: Wed, 12 Jan 2022 21:04:32 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix Message-ID: Hi all, Given the recent (awesome) discussions about the history of *roff and TeX, I thought I'd ask about where Brian Reid's Scribe system fits in with all this. His thesis is available online here: http://reports-archive.adm.cs.cmu.edu/anon/scan/CMU-CS-81-100.pdf, and in my opinion is very interesting (also cites papers on roff and TeX). Does anybody know if Scribe was ever used on Unix systems? Does it exist at all today? Thanks :) Josh From grog at lemis.com Thu Jan 13 12:03:14 2022 From: grog at lemis.com (Greg 'groggy' Lehey) Date: Thu, 13 Jan 2022 13:03:14 +1100 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <20220113020314.GN61872@eureka.lemis.com> On Wednesday, 12 January 2022 at 21:04:32 -0500, josh wrote: > Hi all, > > Given the recent (awesome) discussions about the history of *roff and TeX, I > thought I'd ask about where Brian Reid's Scribe system fits in with all this. > His thesis is available online here: > http://reports-archive.adm.cs.cmu.edu/anon/scan/CMU-CS-81-100.pdf, and in my > opinion is very interesting (also cites papers on roff and TeX). Does anybody > know if Scribe was ever used on Unix systems? Does it exist at all > today? That brings back memories, not of Scribe, but of Scribble by (I think) Craig Finseth of Mark of the Unicorn, who also wrote the MINCE (MINCE Is Not Complete Emacs) editor. This would have been round 1980. Scribble was the first serious text formatting program that I used, and I quite liked it. I can't recall how it compares to more mainstream systems. Greg -- Sent from my desktop computer. Finger grog at lemis.com for PGP public key. See complete headers for address and phone numbers. This message is digitally signed. If your Microsoft mail program reports problems, please read http://lemis.com/broken-MUA.php -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 163 bytes Desc: not available URL: From treese at acm.org Thu Jan 13 13:24:45 2022 From: treese at acm.org (Win Treese) Date: Wed, 12 Jan 2022 22:24:45 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: > On Jan 12, 2022, at 9:04 PM, josh wrote: > > Hi all, > > Given the recent (awesome) discussions about the history of *roff and TeX, I > thought I'd ask about where Brian Reid's Scribe system fits in with all this. > His thesis is available online here: > http://reports-archive.adm.cs.cmu.edu/anon/scan/CMU-CS-81-100.pdf, and in my > opinion is very interesting (also cites papers on roff and TeX). Does anybody > know if Scribe was ever used on Unix systems? Does it exist at all today? Scribe was in the mix after troff and TeX. Brian Reid’s observation was that writers should be writing, and publishing professionals should design how documents look. Writers can describe that they want emphasized words, chapters, sections, quotations, and the appearance of those is designed by people who know about how to do that. The software was commercially available on UNIX and other systems for some time from a company called Unilogic, later Scribe Systems, but it didn’t survive in the marketplace. Wordstar and Microsoft Word came along on the desktop, and academics didn’t like paying for it. Reid’s idea of how the work should be distributed was swept aside by publishing tools that writers could use to do passable documents but not beautiful ones. In real publishing, the division still exists: lots of writing in, say, Microsoft Word that is reworked in publishing software like Quark or InDesign for actual printing. But one could argue that Scribe sort of exists in a way, in LaTeX. My understanding is that Leslie Lamport started LaTex as exactly a way to bring Scribe’s ideas to producing TeX documents, and the basic LaTeX structure looks a lot like Scribe. Because you can dive into TeX to tweak all the tiny details, and because LaTeX packages work at all different levels of abstraction, it’s sometimes hard to see the separation there, especially when you’re fighting with LaTeX to submit a paper. But it’s the Scribe idea at the core. - Win From tytso at mit.edu Thu Jan 13 13:26:10 2022 From: tytso at mit.edu (Theodore Ts'o) Date: Wed, 12 Jan 2022 22:26:10 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Wed, Jan 12, 2022 at 09:04:32PM -0500, josh wrote: > Hi all, > > Given the recent (awesome) discussions about the history of *roff and TeX, I > thought I'd ask about where Brian Reid's Scribe system fits in with all this. > His thesis is available online here: > http://reports-archive.adm.cs.cmu.edu/anon/scan/CMU-CS-81-100.pdf, and in my > opinion is very interesting (also cites papers on roff and TeX). Does anybody > know if Scribe was ever used on Unix systems? Does it exist at all today? Scribe was used at Project Athena at MIT, where it was running on BSD 4.3+ on Vax/750's. So it was definitely used on Unix systems. There were thesis templates for Undergraduates using both Scribe and LaTeX. LaTeX was pretty painfully slow on 1 MIPS machines, but it was better at typesetting complex math equations, which gave it the edge for people majoring in Math, Compter Science, and Engineering degrees. My impression was that Scribe was a bit more popular for people majoring in Humanities (at MIT, Theater, Music, Social Studies, Foreign Languages, etc., were all collapsed into a single department, aka Course 21 --- and there *were* some people who ended up graduating with an undergraduate degree in Course 21, with a concentration in, say, Theater or Music). Speaking of typesetting equations, how would people compare eqn versus LaTeX? I used nroff for man pages, but I never did learn how to use eqn for nroff. - Ted From ats at offog.org Thu Jan 13 23:54:11 2022 From: ats at offog.org (Adam Sampson) Date: Thu, 13 Jan 2022 13:54:11 +0000 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: (josh's message of "Wed, 12 Jan 2022 21:04:32 -0500") References: Message-ID: josh writes: > [...] Brian Reid's Scribe system [...] Does it exist at all today? SAIL's copy of the Scribe source code is in the SAILDART archive: https://saildart.org/[SCR,SYS]/ It's written in BLISS, and appears to have support for both BLISS-10 and BLISS-11. -- Adam Sampson From clemc at ccc.com Fri Jan 14 00:56:48 2022 From: clemc at ccc.com (Clem Cole) Date: Thu, 13 Jan 2022 09:56:48 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: below.... [excellent job -- you wrote what I had planned -- few tweeks/color added below]. On Wed, Jan 12, 2022 at 10:31 PM Win Treese wrote: > Scribe was in the mix after troff and TeX. Fun fact, I believe that bwk was Brian's thesis committee > Brian Reid’s observation was that writers should be writing, and > publishing professionals should design how documents look. It was even more important than that. Brian Reid observed that *most documents* looked like their brethren. For instance, the Unix doc looks like BTL TM format. Each university tends to have a 'style' for their thesis [-me 'knows' how to generate UCB thesis format]. Office correspondence (letters, memo for file) have a certain look. So he wants to separate the description of the output (*i.e.* a template) for a document and not make the person that did want (need too) think about the output, just do their thing (as you said -- let writer's write). > Writers can describe that they want emphasized words, > chapters, sections, quotations, and the appearance of those is designed by > people who know about how to do that. > > The software was commercially available on UNIX and other systems for > some time from a company called Unilogic, later Scribe Systems, There is a long (and somewhat nasty history history here). Mike Shamus was a one-time CMU CS prof. He created Unilogic and arranged for the IP to be transferred to him. They had a transpiler that converted 'compatible BLISS' (*a.k.a*. Vax Bliss) to other languages. I believe that a flavor of Pascal, Ada and C were possible outputs. I'm not sure of the provenance of the transpiler. Contemporary with Scribe was the PQCC project - Product Quality Compiler Compiler, that Wulf and his students were working. That work forked Tartan labs around the same time as Unilogic and I know transpilers had been part of the original work, but I never closed the loop. To be honest they had to have been related in some manner, but that all happened after I had left. At that point, Scribe had been converted from the original PDP-10 version to compatible BLISS on VAX/VMS. Unilogic sold versions of Scribe for TOPS, Vax both VMS and Unix, Sun, Apollo and I believe a number of other systems (maybe HP and AIX). The sad part is that history seems to have lost both Scribe and the associated transpiler (if anyone knows otherwise, I'd love to hear something). > but it didn’t survive in the marketplace. Wordstar and Microsoft Word came > along on the desktop, and academics didn’t like paying for it. > +1 and for many documents (like business letters), I always found Scribe easier - but maybe that's because I grew up the idea of an editor and *document compiler* (roff and friends).. > > Reid’s idea of how the work should be distributed was swept aside by > publishing tools that writers could use to do passable documents but > not beautiful ones. Amen In real publishing, the division still exists: lots of > writing in, say, Microsoft Word that is reworked in publishing software > like Quark or InDesign for actual printing. > To give ex-CMU and UCB grad, Ken Keller credit. He tried to bridge that with his FrameMaker program (which I think Adobe still owns - I have not seen much about it in few years and have lost track of Keller). IIRC Ken's program could take a Scribe/LaTex style sheets also. But FrameMaker (like Scribe) was expensive and originally required a UNIX box with 32-bit linear addressing to compile, so it was fairly late to the PC. I never really learned it although Ken gave me a copy early on to play with. IIRC our doc folks at Stellar used it (whereas the Masscomp/ORA folks of the time were strictly roff as previously discussed). > But one could argue that Scribe sort of exists in a way, in LaTeX. My > understanding is that Leslie Lamport started LaTex as exactly a way > to bring Scribe’s ideas to producing TeX documents, and the basic > LaTeX structure looks a lot like Scribe. > Indeed - that is what I have been told. I am under the impression that early on when Brian started as a Stanford Prof, he had difficulty getting a use license from Unilogic even though he was the original author. That churn supposedly somehow influenced Leslie WRT to the creation of LaTex to make Tex more accessible. > > Because you can dive into TeX to tweak all the tiny details, and > because LaTeX packages work at all different levels of abstraction, > it’s sometimes hard to see the separation there, especially when > you’re fighting with LaTeX to submit a paper. But it’s the Scribe > idea at the core. > +1 FWIW I was back at CMU a couple of winters ago for the annual 'Build-18' maker event. I was chatting with some folks about a few of the cool things we had worked on in the 70's and which ones had lasted, like the PQCC, the speech recognition work, Mach, Andrew *et al*. At that time, I was told then that there was a linux x86 binary for Scribe still floating around and some people still used it for some specific documents. I asked if I could get a copy to play with and they told me they would try to find it, but that was right before Covidtide. It's been crickets since I inquired. I'll see if I can find out more. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jpl.jpl at gmail.com Fri Jan 14 01:08:24 2022 From: jpl.jpl at gmail.com (John P. Linderman) Date: Thu, 13 Jan 2022 10:08:24 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 9:57 AM Clem Cole wrote: > below.... [excellent job -- you wrote what I had planned -- few > tweeks/color added below]. > > On Wed, Jan 12, 2022 at 10:31 PM Win Treese wrote: > >> Scribe was in the mix after troff and TeX. > > Fun fact, I believe that bwk was Brian's thesis committee > > >> Brian Reid’s observation was that writers should be writing, and >> publishing professionals should design how documents look. > > >> Many of us who wrote articles for the Bell System Technical Journal would disagree. The BSTJ publishers could transform something that made sense when viewed as troff output into unintelligible gibberish. You cannot split a UNIX command line into multiple lines just because it "looks better". Sometimes format really matters. -------------- next part -------------- An HTML attachment was scrubbed... URL: From crossd at gmail.com Fri Jan 14 01:35:46 2022 From: crossd at gmail.com (Dan Cross) Date: Thu, 13 Jan 2022 10:35:46 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 9:57 AM Clem Cole wrote: > On Wed, Jan 12, 2022 at 10:31 PM Win Treese wrote: > >> [snip] > > The software was commercially available on UNIX and other systems for >> some time from a company called Unilogic, later Scribe Systems, > > There is a long (and somewhat nasty history history here). Mike Shamus > was a one-time CMU CS prof. He created Unilogic and arranged for the IP > to be transferred to him. They had a transpiler that converted > 'compatible BLISS' (*a.k.a*. Vax Bliss) to other languages. I believe > that a flavor of Pascal, Ada and C were possible outputs. I'm not sure of > the provenance of the transpiler. Contemporary with Scribe was the PQCC > project - Product Quality Compiler Compiler, that Wulf and his students > were working. That work forked Tartan labs around the same time as > Unilogic and I know transpilers had been part of the original work, but I > never closed the loop. To be honest they had to have been related in > some manner, but that all happened after I had left. > > At that point, Scribe had been converted from the original PDP-10 version > to compatible BLISS on VAX/VMS. Unilogic sold versions of Scribe for > TOPS, Vax both VMS and Unix, Sun, Apollo and I believe a number of other > systems (maybe HP and AIX). The sad part is that history seems to have > lost both Scribe and the associated transpiler (if anyone knows otherwise, > I'd love to hear something). > I understand that Stallman was deeply affected by both the closed nature of and "time bombs" in Scribe. > [snip] > > To give ex-CMU and UCB grad, Ken Keller credit. He tried to bridge that > with his FrameMaker program (which I think Adobe still owns - I have not > seen much about it in few years and have lost track of Keller). IIRC Ken's > program could take a Scribe/LaTex style sheets also. But FrameMaker (like > Scribe) was expensive and originally required a UNIX box with 32-bit linear > addressing to compile, so it was fairly late to the PC. I never really > learned it although Ken gave me a copy early on to play with. IIRC our > doc folks at Stellar used it (whereas the Masscomp/ORA folks of the time > were strictly roff as previously discussed). > Does anyone have any experience with Interleaf? That was another in the lineage of document processors that seems to have fallen into history. But one could argue that Scribe sort of exists in a way, in LaTeX. My >> understanding is that Leslie Lamport started LaTex as exactly a way >> to bring Scribe’s ideas to producing TeX documents, and the basic >> LaTeX structure looks a lot like Scribe. >> > Indeed - that is what I have been told. I am under the impression that > early on when Brian started as a Stanford Prof, he had difficulty getting a > use license from Unilogic even though he was the original author. > That churn supposedly somehow influenced Leslie WRT to the creation of > LaTex to make Tex more accessible. > Texinfo was supposedly developed as an alternative to Scribe specifically; I know Arnold has said he really likes it for writing books. I wonder what the connection between texinfo and latex is, if any at all. To bring it back to Unix, troff et al are obvious examples of the Unix philosophy applied to document preparation, while TeX and its progeny have always felt very foreign to me. They work, of course, but in a way that feels discordant with respect to the aesthetic of the system. Of course, TeX originated on the SAIL system, so that makes sense: the PDP-10 world had different sensibilities than the Unix world. One wonders whether, if Knuth had been working on a Unix machine instead of SAIL, whether TeX would have been as chatty as it is; I suspect not. - Dan C. -------------- next part -------------- An HTML attachment was scrubbed... URL: From imp at bsdimp.com Fri Jan 14 02:02:52 2022 From: imp at bsdimp.com (Warner Losh) Date: Thu, 13 Jan 2022 09:02:52 -0700 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 8:36 AM Dan Cross wrote: > On Thu, Jan 13, 2022 at 9:57 AM Clem Cole wrote: > >> On Wed, Jan 12, 2022 at 10:31 PM Win Treese wrote: >> >>> [snip] >> >> The software was commercially available on UNIX and other systems for >>> some time from a company called Unilogic, later Scribe Systems, >> >> There is a long (and somewhat nasty history history here). Mike Shamus >> was a one-time CMU CS prof. He created Unilogic and arranged for the IP >> to be transferred to him. They had a transpiler that converted >> 'compatible BLISS' (*a.k.a*. Vax Bliss) to other languages. I believe >> that a flavor of Pascal, Ada and C were possible outputs. I'm not sure of >> the provenance of the transpiler. Contemporary with Scribe was the PQCC >> project - Product Quality Compiler Compiler, that Wulf and his students >> were working. That work forked Tartan labs around the same time as >> Unilogic and I know transpilers had been part of the original work, but I >> never closed the loop. To be honest they had to have been related in >> some manner, but that all happened after I had left. >> >> At that point, Scribe had been converted from the original PDP-10 version >> to compatible BLISS on VAX/VMS. Unilogic sold versions of Scribe for >> TOPS, Vax both VMS and Unix, Sun, Apollo and I believe a number of other >> systems (maybe HP and AIX). The sad part is that history seems to have >> lost both Scribe and the associated transpiler (if anyone knows otherwise, >> I'd love to hear something). >> > > I understand that Stallman was deeply affected by both the closed nature > of and "time bombs" in Scribe. > > >> [snip] >> >> To give ex-CMU and UCB grad, Ken Keller credit. He tried to bridge that >> with his FrameMaker program (which I think Adobe still owns - I have not >> seen much about it in few years and have lost track of Keller). IIRC Ken's >> program could take a Scribe/LaTex style sheets also. But FrameMaker (like >> Scribe) was expensive and originally required a UNIX box with 32-bit linear >> addressing to compile, so it was fairly late to the PC. I never really >> learned it although Ken gave me a copy early on to play with. IIRC our >> doc folks at Stellar used it (whereas the Masscomp/ORA folks of the time >> were strictly roff as previously discussed). >> > > Does anyone have any experience with Interleaf? That was another in the > lineage of document processors that seems to have fallen into history. > > But one could argue that Scribe sort of exists in a way, in LaTeX. My >>> understanding is that Leslie Lamport started LaTex as exactly a way >>> to bring Scribe’s ideas to producing TeX documents, and the basic >>> LaTeX structure looks a lot like Scribe. >>> >> Indeed - that is what I have been told. I am under the impression that >> early on when Brian started as a Stanford Prof, he had difficulty getting a >> use license from Unilogic even though he was the original author. >> That churn supposedly somehow influenced Leslie WRT to the creation of >> LaTex to make Tex more accessible. >> > > Texinfo was supposedly developed as an alternative to Scribe specifically; > I know Arnold has said he really likes it for writing books. I wonder what > the connection between texinfo and latex is, if any at all. > You can best view them as -ms vs -me. Two different sets of macros to markup the text with semantic information that's then turned into useful rendering by a variety of ways. texinfo and latex are completely unrelated at a code level. LaTeX predates texinfo by some time (I've not looked it up, but I encountered LaTeX years before texinfo, though it's possible I just ignored it when working on bringing up GNU Emacs on VMS 5.mumble back in the day). It was always my impression that texinfo came more from the ITS info file world and that the TeX bits were initially just a hack because it was also on those machines... It would be interesting to hear from people that were there. > To bring it back to Unix, troff et al are obvious examples of the Unix > philosophy applied to document preparation, while TeX and its progeny have > always felt very foreign to me. They work, of course, but in a way that > feels discordant with respect to the aesthetic of the system. Of course, > TeX originated on the SAIL system, so that makes sense: the PDP-10 world > had different sensibilities than the Unix world. One wonders whether, if > Knuth had been working on a Unix machine instead of SAIL, whether TeX would > have been as chatty as it is; I suspect not. > Likely not. It was only slightly odd to me because our school moved from TOPS-20 to SunOS and 4.{2,3}BSD (maybe others, don't know when the VAX was delivered: it was just there when I arrived with a boatload of HP terminals attached to it which I thought odd). It's quite TOPS-20-y in a lot of what it does. That seemed perfectly natural to me when I started using it. Warner -------------- next part -------------- An HTML attachment was scrubbed... URL: From clemc at ccc.com Fri Jan 14 02:06:11 2022 From: clemc at ccc.com (Clem Cole) Date: Thu, 13 Jan 2022 11:06:11 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 10:08 AM John P. Linderman wrote: > Many of us who wrote articles for the Bell System Technical Journal would > disagree. The BSTJ publishers could transform something that made sense > when viewed as troff output into unintelligible gibberish. You cannot split > a UNIX command line into multiple lines just because it "looks better". > Sometimes format really matters. > I think that is true for any scheme -- professionals and editors need to work together. That's what Jon was suggesting. When they don't have shared vocabulary/goals - bad things can happen. FWIW: I can not speak for him directly as I never had this conversation with him (Win might have), but from what I knew/know of Brian Ried I think he might agree with what I'm suggesting. IMO, *there will always be cases like the one that you described*. This is not particular to any document compiler system. The question is how to bring the two sides together and who has the high order bit? My complaint with Word and the like, is that the 'control' is hidden. It's $%^& magic -- why is it indenting here? Hey I did not tell it to make it go italics ... Like Jon and Larry, I'm a big roff fan and still use it. But to give Brian his due, his style sheets were in ASCII and what was happening on the page was fairly easy to deduce. That said, I never used Scribe for anything large (like a book), which I can say I have done with troff. In the late 1970s, I did use Scribe for some papers and found it quite easy to use. Since that time, as a co-author I've also tried the same with LaTex and/or Word and found both difficult. When I have the lead and if I can, I'll use troff -ms with a few extra Masscomp macros (that ORA used to pass on too -- the Steve Talbot extensions for lists in particular). So from my professional experience, it has been mostly with troff, my use of Scribe was short lived. I'm pretty sure tht Keller tried to make creating books easier in FrameMaker, as that was one of his target users. But again I only played with it, never really had to rely on it for anything. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer at technologists.com Fri Jan 14 02:13:59 2022 From: sauer at technologists.com (Charles H Sauer) Date: Thu, 13 Jan 2022 10:13:59 -0600 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <8ac8106b-9f66-f7bc-ef17-a0b4650179af@technologists.com> On 1/13/2022 9:35 AM, Dan Cross wrote: > Does anyone have any experience with Interleaf? That was another in the > lineage of document processors that seems to have fallen into history. Interleaf was a favorite on AIX, starting with the initial RT release. At Dell, we used the DOS version of Interleaf extensively internally, and I used it for charts for classroom presentations when I was an adjunct at UTCS. I still have that version and can view those charts. Some descendant of Interleaf seems to exist as BroadVision Quicksilver, but I haven't used Interleaf since leaving Dell. -- voice: +1.512.784.7526 e-mail: sauer at technologists.com fax: +1.512.346.5240 Web: https://technologists.com/sauer/ Facebook/Google/Twitter: CharlesHSauer From clemc at ccc.com Fri Jan 14 02:20:12 2022 From: clemc at ccc.com (Clem Cole) Date: Thu, 13 Jan 2022 11:20:12 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 11:03 AM Warner Losh wrote: > You can best view them as -ms vs -me. Two different sets of macros to > markup the text with semantic information that's then turned into useful > rendering by a variety of ways. texinfo and latex are completely unrelated > at a code level. LaTeX predates texinfo by some time (I've not looked it > up, but I encountered LaTeX years before texinfo, though it's possible I > just ignored it when working on bringing up GNU Emacs on VMS 5.mumble back > in the day). It was always my impression that texinfo came more from the > ITS info file world and that the TeX bits were initially just a hack > because it was also on those machines... It would be interesting to hear > from people that were there. > That was always my impression. Stallman hated troff (and man) because it was not integrated into EMACS (his operating system). I always got the impression that texinfo was more of a shot against man pages and trying to push the purity of the 'ITS-way' to Unix. And of course the problem became the more he did that, the less use texinfo became except for anyone that followed his gospel. As Unix became the mainstream, it meant information for any gnu tools was (is) disjoint. If rms had been willing to just accept the man command itself, I suspect that would not have been. To be fair, its not an unusual behavior. Folks coming from VMS (or windows) try to make Unix look like that. And what did we do with VMS, we added the cshell and Unix tools so people like me could type on it. I'm currently being driven nuts by the simh tool which has a very TOPS/VMS style feel. Everything is help commands. The documents are quite sparse. To me the documents/book (or man pages) should primarily ' go to' and if you want something like 'help' then create it from the documents. But I can not really complain. It's a wonderful tool and its author comes from that heritage not Unix. The problem I remind him is that he will get silly questions from people like me, because we can not find things in his help system - its just not how a Unix program tends to work [I want to look in the index of the document, and find the section myself]. -------------- next part -------------- An HTML attachment was scrubbed... URL: From imp at bsdimp.com Fri Jan 14 02:24:56 2022 From: imp at bsdimp.com (Warner Losh) Date: Thu, 13 Jan 2022 09:24:56 -0700 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 9:06 AM Clem Cole wrote: > > > On Thu, Jan 13, 2022 at 10:08 AM John P. Linderman > wrote: > >> Many of us who wrote articles for the Bell System Technical Journal would >> disagree. The BSTJ publishers could transform something that made sense >> when viewed as troff output into unintelligible gibberish. You cannot split >> a UNIX command line into multiple lines just because it "looks better". >> Sometimes format really matters. >> > I think that is true for any scheme -- professionals and editors need to > work together. That's what Jon was suggesting. When they don't have > shared vocabulary/goals - bad things can happen. FWIW: I can not speak > for him directly as I never had this conversation with him (Win might > have), but from what I knew/know of Brian Ried I think he might agree with > what I'm suggesting. IMO, *there will always be cases like the one that > you described*. This is not particular to any document compiler system. > The question is how to bring the two sides together and who has the high > order bit? > > My complaint with Word and the like, is that the 'control' is hidden. > It's $%^& magic -- why is it indenting here? Hey I did not tell it to make > it go italics ... > Yea. There's a balance here: the number of people that tweak things because they can is quite large. and often the tweaks need to be undone because they look like @#^@^ to the professional typesetter (I guess they'd call this the publisher these days). There also needs to be some way to flag the legit "your defaults got this so wrong my readers will trip over this" bits. That's lacking in Word, for example. I've seen other systems cope with this to varying degrees of success. I've used LaTeX for all my professional papers. With the proper style guides, I can easily transport the words from one style requirement to another. However, I run into issues all the time when I go from conference A that has a single column to conference B that has the dual columns of IEEE. Where diagrams fit and are pleasing to the eye in one, they look awkward and out of place in the other. Etc. So this ideal one can approach, but there will always be bits of bricabrack that can't be easily handled by the automation. While most of the issues can be delegated to the macros, some manual tweaking is necessary because there are many works that are more than just a big bag of words with semantic metadata attached. I never got into troff. It always seemed lower level than LaTeX to me when I was learning things, and I didn't want to be bothered with those details. I can read and use it today, but it's not my primary choice unless I'm tweaking a work already in troff. Warner -------------- next part -------------- An HTML attachment was scrubbed... URL: From lm at mcvoy.com Fri Jan 14 02:25:21 2022 From: lm at mcvoy.com (Larry McVoy) Date: Thu, 13 Jan 2022 08:25:21 -0800 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <20220113162521.GC7175@mcvoy.com> On Wed, Jan 12, 2022 at 10:26:10PM -0500, Theodore Ts'o wrote: > Speaking of typesetting equations, how would people compare eqn versus > LaTeX? I used nroff for man pages, but I never did learn how to use > eqn for nroff. I love the tbl|eqn|pic|grap (though that is in theory, I wrote my own) preprocessors. I've done a ton of tbl & pic, there are some historical bugs that I'd like to get fixed, but for the most part they work great. eqn I've used and made pretty math stuff but I haven't used the LaTex version. Somewhere I heard someone grabbed the eqn source and massaged it for LaTex but I have no idea if that was true. I'm especially fond of pic, you can write pic and "see" the picture in your head if you do it right. I'm pretty sure I know why it works like that, I bet it was slow and expensive to get a print out so the more you could get it to be correct in your head, the better. From imp at bsdimp.com Fri Jan 14 02:32:14 2022 From: imp at bsdimp.com (Warner Losh) Date: Thu, 13 Jan 2022 09:32:14 -0700 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 9:20 AM Clem Cole wrote: > To be fair, its not an unusual behavior. Folks coming from VMS (or > windows) try to make Unix look like that. And what did we do with VMS, we > added the cshell and Unix tools so people like me could type on it. I'm > currently being driven nuts by the simh tool which has a very TOPS/VMS > style feel. Everything is help commands. The documents are quite sparse. > To me the documents/book (or man pages) should primarily ' go to' and if > you want something like 'help' then create it from the documents. But I > can not really complain. It's a wonderful tool and its author comes from > that heritage not Unix. The problem I remind him is that he will get silly > questions from people like me, because we can not find things in his help > system - its just not how a Unix program tends to work [I want to look in > the index of the document, and find the section myself]. > The different worldviews are instructive. c-kermit had a tops-20 cmd JSYS built into it and all the escape completion just worked. The help was quite good, but as with most help systems it was a zoomed in on some hyperspecific topic. For that it was usually good (across all the TOPS-20 and VMS and that ilk), but sometimes you got things like "/GERBILS specifies the gerbils to use" without really telling you what a gerbil was in this context. what you didn't get was how things held together (eg /HAMSTERS and /GERBILS worked hand in hand to control parameters to the rodent models used to generate the ecosystem) or some of the higher level concepts (like what an ecosystem was). I used to read the raw VMS help files. They were a verbose version of the Unix manuals in some ways, though less crisp in others. It was helpful to read it all to understand, but even so some important concepts were omitted, or discussed elsewhere w/o a proper cross reference. Warner -------------- next part -------------- An HTML attachment was scrubbed... URL: From crossd at gmail.com Fri Jan 14 02:33:27 2022 From: crossd at gmail.com (Dan Cross) Date: Thu, 13 Jan 2022 11:33:27 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Wed, Jan 12, 2022 at 10:32 PM Theodore Ts'o wrote: > Speaking of typesetting equations, how would people compare eqn versus > LaTeX? I used nroff for man pages, but I never did learn how to use > eqn for nroff. > I hate to be the one who says this, but when it comes to typesetting non-trivial mathematics, there is no competition: LaTeX beats eqn hands down. eqn is fine up to a point (and the neqn thing is kinda nifty for simple things on the terminal; you can kinda sorta get a rendered sigma for a summation, for example) but it breaks down pretty quickly. - Dan C. -------------- next part -------------- An HTML attachment was scrubbed... URL: From skogtun at gmail.com Fri Jan 14 02:39:54 2022 From: skogtun at gmail.com (Harald Arnesen) Date: Thu, 13 Jan 2022 17:39:54 +0100 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <67ee7f15-f5f3-a508-30aa-edb1dee1ab26@gmail.com> Clem Cole [13/01/2022 17.06]: > My complaint with Word and the like, is that the 'control' is hidden. > It's $%^& magic -- why is it indenting here?  Hey I did not tell it to > make it go italics ... That's the main reason I used WordPerfect instead, when I had to deliver my work in Microsoft Word format (about 1990-2000). WordPerfect has a "reveal codes" command, which makes it possible to see why the text suddenly is in italics... And it could/can save in Word format, which my editors demanded. -- Hilsen Harald From lars at nocrew.org Fri Jan 14 02:42:53 2022 From: lars at nocrew.org (Lars Brinkhoff) Date: Thu, 13 Jan 2022 16:42:53 +0000 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: (Clem Cole's message of "Thu, 13 Jan 2022 11:20:12 -0500") References: Message-ID: <7w4k67a8oi.fsf@junk.nocrew.org> Clem Cole wrote: > I always got the impression that texinfo was more of a shot against > man pages and trying to push the purity of the 'ITS-way' to Unix. ITS had a hypertext documentation system, so my assumtion would be that RMS wanted to bring along that to the GNU vision. I don't see that's it would be abount purity, whatever that would mean for documentation. From lm at mcvoy.com Fri Jan 14 02:52:07 2022 From: lm at mcvoy.com (Larry McVoy) Date: Thu, 13 Jan 2022 08:52:07 -0800 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: <7w4k67a8oi.fsf@junk.nocrew.org> References: <7w4k67a8oi.fsf@junk.nocrew.org> Message-ID: <20220113165207.GD7175@mcvoy.com> On Thu, Jan 13, 2022 at 04:42:53PM +0000, Lars Brinkhoff wrote: > Clem Cole wrote: > > I always got the impression that texinfo was more of a shot against > > man pages and trying to push the purity of the 'ITS-way' to Unix. > > ITS had a hypertext documentation system, so my assumtion would be that > RMS wanted to bring along that to the GNU vision. I don't see that's it > would be abount purity, whatever that would mean for documentation. So you know how when you go into someone else's program to fix a bug and they have a hideous coding style? Have you ever had someone else fix a bug in your code and they reformat everything so git blame looks like they wrote the whole thing? That's rude, right? If you were fixing the bug in some crappy coding style, you fix it in that crappy coding style, it's not your style but it is the polite thing to do. If we agree on that then we can move on to RMS and texinfo. Providing texinfo docs for Unix commands is like reformatting the code. It's rude. The Unix way is man pages for basic usage and a user guide, usually in -ms. Not doing it that way is trying to change the way the system works and it's just rude. If I were working on ITS and techinfo is how they do their docs, that's how I'd do docs there, it would be rude to force man pages on system that doesn't work that way. From clemc at ccc.com Fri Jan 14 02:54:51 2022 From: clemc at ccc.com (Clem Cole) Date: Thu, 13 Jan 2022 11:54:51 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: <20220113165207.GD7175@mcvoy.com> References: <7w4k67a8oi.fsf@junk.nocrew.org> <20220113165207.GD7175@mcvoy.com> Message-ID: On Thu, Jan 13, 2022 at 11:52 AM Larry McVoy wrote: > On Thu, Jan 13, 2022 at 04:42:53PM +0000, Lars Brinkhoff wrote: > > Clem Cole wrote: > > > I always got the impression that texinfo was more of a shot against > > > man pages and trying to push the purity of the 'ITS-way' to Unix. > > > > ITS had a hypertext documentation system, so my assumtion would be that > > RMS wanted to bring along that to the GNU vision. I don't see that's it > > would be abount purity, whatever that would mean for documentation. > > So you know how when you go into someone else's program to fix a bug and > they have a hideous coding style? Have you ever had someone else fix a > bug in your code and they reformat everything so git blame looks like > they wrote the whole thing? That's rude, right? If you were fixing > the bug in some crappy coding style, you fix it in that crappy coding > style, it's not your style but it is the polite thing to do. > > If we agree on that then we can move on to RMS and texinfo. Providing > texinfo docs for Unix commands is like reformatting the code. It's > rude. The Unix way is man pages for basic usage and a user guide, > usually in -ms. Not doing it that way is trying to change the way > the system works and it's just rude. > > If I were working on ITS and techinfo is how they do their docs, that's > how I'd do docs there, it would be rude to force man pages on system > that doesn't work that way. > Amen bro ... Amen -------------- next part -------------- An HTML attachment was scrubbed... URL: From lars at nocrew.org Fri Jan 14 02:37:50 2022 From: lars at nocrew.org (Lars Brinkhoff) Date: Thu, 13 Jan 2022 16:37:50 +0000 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: (Adam Sampson's message of "Thu, 13 Jan 2022 13:54:11 +0000") References: Message-ID: <7w8rvja8wx.fsf@junk.nocrew.org> Adam Sampson wrote: > josh wrote: >> [...] Brian Reid's Scribe system [...] Does it exist at all today? > SAIL's copy of the Scribe source code is in the SAILDART archive: > https://saildart.org/[SCR,SYS]/ There is also a copy on MIT backup tapes. A while back I asked Reid's opinion about putting the sources online. He remarked he doesn't really have a say in that since he sold Scribe to a company, but that he wouldn't mind. From clemc at ccc.com Fri Jan 14 03:06:42 2022 From: clemc at ccc.com (Clem Cole) Date: Thu, 13 Jan 2022 12:06:42 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: <20220113165207.GD7175@mcvoy.com> References: <7w4k67a8oi.fsf@junk.nocrew.org> <20220113165207.GD7175@mcvoy.com> Message-ID: On Thu, Jan 13, 2022 at 11:52 AM Larry McVoy wrote: > If you were fixing the bug in some crappy coding style, you fix it in > that crappy coding style, it's not your style but it is the polite thing > to do. > It's like moving to foreign country and not learning the local language and customs. I understand using a crutch for the occasional user, which is what I used when I had to come back to VMS -- I knew enough to be dangerous, but I made fewer errors if I personally have a cshell not Dave's Command Language. Or for windows, I dislike, but I can use if I have too, but I'm happy l to have WSL when I need it. Clem -------------- next part -------------- An HTML attachment was scrubbed... URL: From bakul at iitbombay.org Fri Jan 14 04:00:15 2022 From: bakul at iitbombay.org (Bakul Shah) Date: Thu, 13 Jan 2022 10:00:15 -0800 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <386AE80E-872F-46D4-B441-97C1A8123176@iitbombay.org> > On Jan 13, 2022, at 8:06 AM, Clem Cole wrote: > > > > On Thu, Jan 13, 2022 at 10:08 AM John P. Linderman > wrote: > Many of us who wrote articles for the Bell System Technical Journal would disagree. The BSTJ publishers could transform something that made sense when viewed as troff output into unintelligible gibberish. You cannot split a UNIX command line into multiple lines just because it "looks better". Sometimes format really matters. > I think that is true for any scheme -- professionals and editors need to work together. That's what Jon was suggesting. When they don't have shared vocabulary/goals - bad things can happen. FWIW: I can not speak for him directly as I never had this conversation with him (Win might have), but from what I knew/know of Brian Ried I think he might agree with what I'm suggesting. IMO, there will always be cases like the one that you described. This is not particular to any document compiler system. The question is how to bring the two sides together and who has the high order bit? In one of his blogposts Douglas Crockford has suggested that Scribe would have made a better declarative markup language than SGML! If you read Bibtex's manual they talk about being put in for SCRIBE compatibility. Even the bibliographic reference form looks SCRIBE-like. For example, @Article{Arrabito:EPODD-1-2-117, author = "R. Arrabito and H. J{\"{u}}rgensen", title = "Computerized {Braille} Typesetting: Another View of Mark-up Standards", journal = j-EPODD, year = "1988", volume = "1", number = "2", pages = "117--132", month = sep, } But in a sense the Scribe markup is at a different level than troff or TeX. LaTeX is more declarative, which is one of the reasons for people preferring it to pure TeX. > My complaint with Word and the like, is that the 'control' is hidden. It's $%^& magic -- why is it indenting here? Hey I did not tell it to make it go italics ... Word is lower level than TeX/troff. -------------- next part -------------- An HTML attachment was scrubbed... URL: From crossd at gmail.com Fri Jan 14 04:16:42 2022 From: crossd at gmail.com (Dan Cross) Date: Thu, 13 Jan 2022 13:16:42 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: On Thu, Jan 13, 2022 at 11:03 AM Warner Losh wrote: > On Thu, Jan 13, 2022 at 8:36 AM Dan Cross wrote: > >> [snip] >> Texinfo was supposedly developed as an alternative to Scribe >> specifically; I know Arnold has said he really likes it for writing books. >> I wonder what the connection between texinfo and latex is, if any at all. >> > > You can best view them as -ms vs -me. Two different sets of macros to > markup the text with semantic information that's then turned into useful > rendering by a variety of ways. texinfo and latex are completely unrelated > at a code level. > Oh sure, but I didn't mean in the sense of code, but rather, philosophically and design-wise. Both seem to be influenced by Scribe's idea of separation of content and presentation (an idea reinvented a decade later in HTML+CSS). LaTeX predates texinfo by some time (I've not looked it up, but I > encountered LaTeX years before texinfo, though it's possible I just ignored > it when working on bringing up GNU Emacs on VMS 5.mumble back in the day). > There's some documentation available for both: https://www.gnu.org/software/texinfo/manual/texinfo/html_node/History.html Paraphrasing that document, MIT had a thing called Bolio which evolved into BoTeX. Independently, Stallman created the "info" format (for ITS perhaps?) and then BoTeX and Info merged to become texinfo, with the stated goal of producing both online and printed representations from a single source document. The earliest texinfo formatter was written in Emacs Lisp. BoTeX seems to date from late 1984, but this doesn't put a date on the creation of texinfo. It _does_ mention `makeinfo` in "the early 90s", so we may assume sometime after 1984 and before 1992? `texinfo.el` from the Emacs 18.29 distribution has copyright dates from 1985, 1988, but it's hard to make out the actual provenance of the source in those files (ie, was the 1985 date due to that file being copied from an earlier file created in 1985?). It's somewhat harder to nail down the exact history of LaTeX; Lamport has this: http://lamport.azurewebsites.net/pubs/pubs.html#latex which seems to indicate that, while "LaTeX: A Document Preparation System" was published in 1986, he had been working on it for at least two or three years before that. https://lamport.azurewebsites.net/pubs/lamport-latex-interview.pdf mentions that he started using TeX ca 1979, but doesn't mention when he actually created LaTeX. He goes on to say that he was using a macro package by Max Diaz and thought he could do better "when Don was creating TeX80(?)". I'd guess that means this is in the 1980-81 timeframe? It goes on to say that he moved to DEC in 1985 and never used *roff (I assume he at least poked at it and take that to mean he was never a serious user). It was always my impression that texinfo came more from the ITS info file > world and that the TeX bits were initially just a hack because it was also > on those machines... It would be interesting to hear from people that were > there. > Info definitely came from that world. Texinfo as the marriage of BoTeX as a Scribe-a-like and Info as an online help format seem less like a hack and more deliberate. > To bring it back to Unix, troff et al are obvious examples of the Unix >> philosophy applied to document preparation, while TeX and its progeny have >> always felt very foreign to me. They work, of course, but in a way that >> feels discordant with respect to the aesthetic of the system. Of course, >> TeX originated on the SAIL system, so that makes sense: the PDP-10 world >> had different sensibilities than the Unix world. One wonders whether, if >> Knuth had been working on a Unix machine instead of SAIL, whether TeX would >> have been as chatty as it is; I suspect not. >> > > Likely not. It was only slightly odd to me because our school moved from > TOPS-20 to SunOS and 4.{2,3}BSD (maybe others, don't know when the VAX was > delivered: it was just there when I arrived with a boatload of HP terminals > attached to it which I thought odd). > Which part was weird? The HP terminals? It's quite TOPS-20-y in a lot of what it does. That seemed perfectly > natural to me when I started using it. > Or at least SAIL-y, but it seems like the PDP-10 systems had a lot of cross-pollination between them, possibly due to the shared lineage from the PDP-6 monitor and associated DEC tools like DDT? - Dan C. -------------- next part -------------- An HTML attachment was scrubbed... URL: From lm at mcvoy.com Fri Jan 14 06:00:57 2022 From: lm at mcvoy.com (Larry McVoy) Date: Thu, 13 Jan 2022 12:00:57 -0800 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <20220113200057.GI7175@mcvoy.com> On Thu, Jan 13, 2022 at 01:16:42PM -0500, Dan Cross wrote: > Oh sure, but I didn't mean in the sense of code, but rather, > philosophically and design-wise. Both seem to be influenced by Scribe's > idea of separation of content and presentation (an idea reinvented a decade > later in HTML+CSS). Funny, Marc Donner, then at Morgan Stanley, was sort of a mentor of mine. One day I was discussing with him all the stuff I had done with *roff -ms input, I have scripts to make html slides out of that, another that makes a web site with a site map. I told him that I really liked it but wasn't sure why. He said "You like it because that macro packages says what to do without exposing how. That makes it easy to write scripts to spit out in some other format". True dat. From chet.ramey at case.edu Fri Jan 14 06:26:51 2022 From: chet.ramey at case.edu (Chet Ramey) Date: Thu, 13 Jan 2022 15:26:51 -0500 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <9d43b5f9-68cc-711a-a711-e8922b5c5953@case.edu> On 1/13/22 1:16 PM, Dan Cross wrote: > It _does_ mention `makeinfo` in "the early 90s", so we may > assume sometime after 1984 and before 1992? `texinfo.el` from the Emacs > 18.29 distribution has copyright dates from 1985, 1988, but it's hard to > make out the actual provenance of the source in those files (ie, was the > 1985 date due to that file being copied from an earlier file created in 1985?). Brian Fox wrote the original version of GNU `info' and the first standalone C version of `makeinfo' in 1987, right after he started working for the FSF. (makeinfo has since evolved into a perl script.) -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ``Ars longa, vita brevis'' - Hippocrates Chet Ramey, UTech, CWRU chet at case.edu http://tiswww.cwru.edu/~chet/ From lars at nocrew.org Fri Jan 14 06:40:32 2022 From: lars at nocrew.org (Lars Brinkhoff) Date: Thu, 13 Jan 2022 20:40:32 +0000 Subject: [COFF] *roff history as told to GNU In-Reply-To: <2DD03634-EEE7-4311-9CFB-1FCBBB3C927A@johnlabovitz.com> (John Labovitz's message of "Thu, 13 Jan 2022 08:47:28 -0500") References: <20220101031511.GB8135@mcvoy.com> <20220111015901.GE25103@eureka.lemis.com> <4409b91cd794867d@orthanc.ca> <20220111024218.GE3441@mcvoy.com> <202201120854.20C8sDRR014233@freefriends.org> <20220112180619.mxzojchd62vpwm2f@localhost.localdomain> <7wiluo9j52.fsf@junk.nocrew.org> <2DD03634-EEE7-4311-9CFB-1FCBBB3C927A@johnlabovitz.com> Message-ID: <7wv8yn8j3z.fsf_-_@junk.nocrew.org> Moving to COFF. John Labovitz wrote: >>> The earliest known text-formatting software, TJ-2, was created by >>> MIT-trained computer scientist Peter Samson in 1963. >> >> I see claimed predecessors are JUSTIFY and TJ-1. How do you feel >> about those? > > I’m sure I looked for TJ-1 when I did this research — an obvious > question, given the ‘2’ suffix — but didn’t find anything then. I’m > not familiar with JUSTIFY. Note, later there was also a TJ6 for the PDP-6, written by Richard Greenblatt. > Do you have links/info for those? This one mentions TJ-1 near the end: https://www.computerhistory.org/pdp-1/_media/pdf/DEC.pdp_1.1972.102650621.pdf The TJ-2 page on Wikipedia mentions JUSTIFY and links here: http://www.ultimate.com/phil/pdp10/ From davida at pobox.com Fri Jan 14 08:19:35 2022 From: davida at pobox.com (David Arnold) Date: Fri, 14 Jan 2022 09:19:35 +1100 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <94F48429-0CD7-4C9B-867F-DC75B797DFA8@pobox.com> > On 14 Jan 2022, at 03:33, Dan Cross wrote: > > On Wed, Jan 12, 2022 at 10:32 PM Theodore Ts'o > wrote: > Speaking of typesetting equations, how would people compare eqn versus > LaTeX? I used nroff for man pages, but I never did learn how to use > eqn for nroff. > > I hate to be the one who says this, but when it comes to typesetting non-trivial mathematics, there is no competition: LaTeX beats eqn hands down. eqn is fine up to a point (and the neqn thing is kinda nifty for simple things on the terminal; you can kinda sorta get a rendered sigma for a summation, for example) but it breaks down pretty quickly. Both FrameMaker and Word have GUI equation editors. They’re pretty capable, but are a separate “world” from the text: they open a new window for editing the equation that floats over the document. I much preferred (I don’t do a lot of equations these days) the inline nature of LaTeX (and (n)eqn)). d -------------- next part -------------- An HTML attachment was scrubbed... URL: From davida at pobox.com Fri Jan 14 08:53:40 2022 From: davida at pobox.com (David Arnold) Date: Fri, 14 Jan 2022 09:53:40 +1100 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: References: Message-ID: <1A1EC0A2-4E57-4810-8765-570D1A5E2C16@pobox.com> > On 14 Jan 2022, at 01:56, Clem Cole wrote: <…> > In real publishing, the division still exists: lots of > writing in, say, Microsoft Word that is reworked in publishing software > like Quark or InDesign for actual printing. > To give ex-CMU and UCB grad, Ken Keller credit. He tried to bridge that with his FrameMaker program (which I think Adobe still owns - I have not seen much about it in few years and have lost track of Keller). IIRC Ken's program could take a Scribe/LaTex style sheets also. But FrameMaker (like Scribe) was expensive and originally required a UNIX box with 32-bit linear addressing to compile, so it was fairly late to the PC. I never really learned it although Ken gave me a copy early on to play with. IIRC our doc folks at Stellar used it (whereas the Masscomp/ORA folks of the time were strictly roff as previously discussed). I used FrameMaker for a while in the early 90’s, (on MIPS Ultrix 4.4, DEC OSF/1 3.2/Digital Unix 4.0/Tru64 5.1, and Solaris 8/9). It was absolutely targeted at professional document production (vs. word processing), and had a good separation between templates (with all the styling and layout) and documents (which added the content). It was also good at merging multiple documents into a book and applying consistent style across the collection. I think the versions I used were still produced by Frame Technology(?) — before it was bought by Adobe anyway. Ironically, around that same time, Word for Windows 2.0 was IMHO the peak of that product’s functionality, in that while it allowed the user to randomly apply styling to the text, it was the last version that made the template facilities an equal first-class citizen in the UI. It was almost as easy to define and use “semantic” styles for formatting as it was to just do inline markup. Version 6 (they went from 2 to 6 in one hop) bent the product firmly towards use by amateurs, with toolbar buttons for bulleted lists, etc, that hid the underlying use of styles, and thus avoided users incrementally learning how to do consistent documents. From my perspective, every release since has made it worse. More recently, I’ve used DocBook for producing manuals, etc. It’s an awful source format, but has the usual advantages of being plain text, generating useful textual diffs, etc. Cobbling together a production process using DocBook, XSLT, and FOP to emit decent PDF is not for the faint-hearted, and (to circle back to relevance) I always end up questioning whether it’d just be better to use *roff or (La)TeX with a suitable macro package instead. d -------------- next part -------------- An HTML attachment was scrubbed... URL: From athornton at gmail.com Fri Jan 14 11:53:14 2022 From: athornton at gmail.com (Adam Thornton) Date: Thu, 13 Jan 2022 18:53:14 -0700 Subject: [COFF] Scribe (Typesetting System) and Unix In-Reply-To: <1A1EC0A2-4E57-4810-8765-570D1A5E2C16@pobox.com> References: <1A1EC0A2-4E57-4810-8765-570D1A5E2C16@pobox.com> Message-ID: On Thu, Jan 13, 2022 at 4:00 PM David Arnold wrote: > > Ironically, around that same time, Word for Windows 2.0 was IMHO the peak > of that product’s functionality, in that while it allowed the user to > randomly apply styling to the text, it was the last version that made the > template facilities an equal first-class citizen in the UI. It was almost > as easy to define and use “semantic” styles for formatting as it was to > just do inline markup. Version 6 (they went from 2 to 6 in one hop) bent > the product firmly towards use by amateurs, with toolbar buttons for > bulleted lists, etc, that hid the underlying use of styles, and thus > avoided users incrementally learning how to do consistent documents. From > my perspective, every release since has made it worse. > > Thank you for saying this. I agree and I thought I was the only person who felt that way. Adam -------------- next part -------------- An HTML attachment was scrubbed... URL: From athornton at gmail.com Thu Jan 27 12:09:12 2022 From: athornton at gmail.com (Adam Thornton) Date: Wed, 26 Jan 2022 19:09:12 -0700 Subject: [COFF] Tonight's epiphany Message-ID: "Reflections on Trusting Trust" plus the fact that no one has designed new real computers at the gate level for at least 30 years, maybe longer--it's done in an HDL of some kind, which is to say, software--means it's already way, way too late. I for one welcome our new non-biological overlords. Adam -------------- next part -------------- An HTML attachment was scrubbed... URL: From bakul at iitbombay.org Thu Jan 27 13:00:52 2022 From: bakul at iitbombay.org (Bakul Shah) Date: Wed, 26 Jan 2022 19:00:52 -0800 Subject: [COFF] Tonight's epiphany In-Reply-To: References: Message-ID: > On Jan 26, 2022, at 6:09 PM, Adam Thornton wrote: > > "Reflections on Trusting Trust" plus the fact that no one has designed new real computers at the gate level for at least 30 years, maybe longer--it's done in an HDL of some kind, which is to say, software--means it's already way, way too late. https://arith-matic.com/notebook/4bit-7400-homebrew-computer-cpu https://eater.net/8bit/ http://cpuville.com/Projects/Original-CPU/Original-CPU-home.html Then there are people like Ken Shiriff who are revese engineering from die photos of a chip! https://www.youtube.com/watch?v=aHx-XUA6f9g Of course, you may need some machine learning to reverse engineer a modern chip with bilions of transistors to see if some surreptitious logic was sneaked in! You may need some way to prove the manufactured chip is exactly what was specified. Probably impossible today. From lm at mcvoy.com Thu Jan 27 13:32:41 2022 From: lm at mcvoy.com (Larry McVoy) Date: Wed, 26 Jan 2022 19:32:41 -0800 Subject: [COFF] Tonight's epiphany In-Reply-To: References: Message-ID: <20220127033241.GA18469@mcvoy.com> I'm still waiting for the self timed logic I was taught in the 1980s. Does anyone know why that is not a thing? It seemed smart back then. On Wed, Jan 26, 2022 at 07:00:52PM -0800, Bakul Shah wrote: > > > > On Jan 26, 2022, at 6:09 PM, Adam Thornton wrote: > > > > "Reflections on Trusting Trust" plus the fact that no one has designed new real computers at the gate level for at least 30 years, maybe longer--it's done in an HDL of some kind, which is to say, software--means it's already way, way too late. > > https://arith-matic.com/notebook/4bit-7400-homebrew-computer-cpu > https://eater.net/8bit/ > http://cpuville.com/Projects/Original-CPU/Original-CPU-home.html > > Then there are people like Ken Shiriff who are revese > engineering from die photos of a chip! > https://www.youtube.com/watch?v=aHx-XUA6f9g > > Of course, you may need some machine learning to reverse > engineer a modern chip with bilions of transistors to see > if some surreptitious logic was sneaked in! > > You may need some way to prove the manufactured chip is > exactly what was specified. Probably impossible today. > > _______________________________________________ > COFF mailing list > COFF at minnie.tuhs.org > https://minnie.tuhs.org/cgi-bin/mailman/listinfo/coff -- --- Larry McVoy lm at mcvoy.com http://www.mcvoy.com/lm From bakul at iitbombay.org Thu Jan 27 14:30:12 2022 From: bakul at iitbombay.org (Bakul Shah) Date: Wed, 26 Jan 2022 20:30:12 -0800 Subject: [COFF] Tonight's epiphany In-Reply-To: <20220127033241.GA18469@mcvoy.com> References: <20220127033241.GA18469@mcvoy.com> Message-ID: <6CA70D67-6DCA-4F8E-A17B-00C394E07D8F@iitbombay.org> Companies like Wave Computing built devices using NCL (null convention logic) which are self-timed. If you vary temp & voltage, results may come slower or faster but always correct. There is even a clockless RISC-V design called Aristotle. But there may be practical issues that could've made bringing this technology to market difficult. scj can tell you more as he was involved in Wave Computing. There may be other companies but I have not kept track. > On Jan 26, 2022, at 7:32 PM, Larry McVoy wrote: > > I'm still waiting for the self timed logic I was taught in the 1980s. > Does anyone know why that is not a thing? It seemed smart back then. > > On Wed, Jan 26, 2022 at 07:00:52PM -0800, Bakul Shah wrote: >> >> >>> On Jan 26, 2022, at 6:09 PM, Adam Thornton wrote: >>> >>> "Reflections on Trusting Trust" plus the fact that no one has designed new real computers at the gate level for at least 30 years, maybe longer--it's done in an HDL of some kind, which is to say, software--means it's already way, way too late. >> >> https://arith-matic.com/notebook/4bit-7400-homebrew-computer-cpu >> https://eater.net/8bit/ >> http://cpuville.com/Projects/Original-CPU/Original-CPU-home.html >> >> Then there are people like Ken Shiriff who are revese >> engineering from die photos of a chip! >> https://www.youtube.com/watch?v=aHx-XUA6f9g >> >> Of course, you may need some machine learning to reverse >> engineer a modern chip with bilions of transistors to see >> if some surreptitious logic was sneaked in! >> >> You may need some way to prove the manufactured chip is >> exactly what was specified. Probably impossible today. >> >> _______________________________________________ >> COFF mailing list >> COFF at minnie.tuhs.org >> https://minnie.tuhs.org/cgi-bin/mailman/listinfo/coff > > -- > --- > Larry McVoy lm at mcvoy.com http://www.mcvoy.com/lm From efton.collins at gmail.com Thu Jan 27 15:33:51 2022 From: efton.collins at gmail.com (Efton Collins) Date: Wed, 26 Jan 2022 23:33:51 -0600 Subject: [COFF] Tonight's epiphany In-Reply-To: References: Message-ID: On 1/26/22, Bakul Shah wrote: > > https://arith-matic.com/notebook/4bit-7400-homebrew-computer-cpu > https://eater.net/8bit/ > http://cpuville.com/Projects/Original-CPU/Original-CPU-home.html > Bill Buzbee's homebrew machine built with 200 74 series chips runs minix: http://www.homebrewcpu.com/ good fun. From peter at rulingia.com Thu Jan 27 17:37:01 2022 From: peter at rulingia.com (Peter Jeremy) Date: Thu, 27 Jan 2022 18:37:01 +1100 Subject: [COFF] Tonight's epiphany In-Reply-To: <20220127033241.GA18469@mcvoy.com> References: <20220127033241.GA18469@mcvoy.com> Message-ID: On 2022-Jan-26 19:32:41 -0800, Larry McVoy wrote: >I'm still waiting for the self timed logic I was taught in the 1980s. >Does anyone know why that is not a thing? It seemed smart back then. See also the GA144 from http://www.greenarraychips.com/ -- Peter Jeremy -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 963 bytes Desc: not available URL: From dfawcus+lists-coff at employees.org Fri Jan 28 06:41:18 2022 From: dfawcus+lists-coff at employees.org (Derek Fawcus) Date: Thu, 27 Jan 2022 20:41:18 +0000 Subject: [COFF] Tonight's epiphany In-Reply-To: <20220127033241.GA18469@mcvoy.com> References: <20220127033241.GA18469@mcvoy.com> Message-ID: On Wed, Jan 26, 2022 at 07:32:41PM -0800, Larry McVoy wrote: > I'm still waiting for the self timed logic I was taught in the 1980s. > Does anyone know why that is not a thing? It seemed smart back then. There were clockless ARM variants, i.e. https://www.edn.com/arm-philips-develop-clockless-processor/ https://www.eetimes.com/arm-offers-first-clockless-processor-core/ also https://en.wikipedia.org/wiki/AMULET_microprocessor DF