Pagination sans composing room staff

[Mechanic]

I have often been at a loss to give a good description of the tradesmen who practiced their craft in the composing rooms of newspapers in the days of hotmetal. However, Tony Sutton, a man who appears to think he is single handedly dragging the print industry into the 21st century, gives an adequate description of these, often maligned individuals, among whom I count myself.

To quote him out of context:-
"The gangs of two-fingered typesetters along with their thuggish leather-aproned compositor-cum-psychopath pals"

Before that he refers to the composing room staff processing editorial copy as, "..... a gang of sweaty, ham-fisted artisans would miraculously turn it into a newspaper." 

If you have nothing better to do, you might like to read his tongue in cheek article on the perils of pagination using front-end systems. Many of the problems he relates to, sad to say, I have also experienced.

The following is the website.
http://www.newsdesign.net/story3.html

[Dave Hughes]

Nice article, George.

I particularly liked the summing up at the end:

It's ironic that the ultimate price these managers paid for effective pagination was to turn editors into production staff. They didn't get rid of the Linotype operators, they got rid of the editors.

The result? Bad-looking, but well-edited, newspapers have been transformed into good-looking, but vacuous, rags.

But, by God, they're saving money!

[Steve Young]

Takes me back Dave ... I recall the most important thing to sub-editors, editors, etc, when I was installing editorial and pagination systems in the 80s and setting up dictionary exception tables, font conversion tables, H&J tables, kerning tables, etc,  was H&J -- vitally important, vitally important more important even than content sometimes!!! I spent hours setting the tables up, adjusting them, adjusting them again to take note of this and that after being called in to see the editor, sub-editor or whoever had time to complain about various word breaks ..

Vitally, vitally important!!!

Until that is the advent of Quark etc with its incredibly bad justification and odd word breaks.

Suddenly H&J became less important as they did it all themselves!

Jaundiced? Me .... no never.

[Dave Hughes]

Yes, it's surprising what can go by the way-side when saving money gets involved!

[Mechanic]

Editorial pagination is rather simple when compared to the pagination of classified advertisements. The rules for the classified pages of The Sydney Morning Herald were very precise. The classified section carried double column and single column advertisements. Double columns came first, sorted to the first character of the first word and, in order of size, largest advertisements first, then came the single column advertisements sorted, where possible to the first three letters of the first word in the advertisement. Classified display ads were also carried and, if memory serves me, they went at the start of the classified section with DC and SC ads used to fill in vacant space.

The classified pagination compositors, in hotmetal, would sort to the best of their ability ensuring that as little leading as possible was used to fill the page. This continued when we converted to cold type. The computer would sort the advertisements alphabetically DC then SC. and the compositors would cut and paste, as best they could, according the paginations rules.

We had a number of production programmers who were involved with classified pagination. Whenever they installed an update to the classified production system, several of the classified pages would be reset as complete paginated pages. Together with classified advertising management and our system programmers, we would review the results. While I was there we never produced pages that satisfied the requirements of the classified management.

System Integrators Inc., the supplies of both the editorial and advertising systems, said they were working on a new technology of flat screen displays that would allow a screen the size a broadsheet page to be viewed. The idea being that the system would make up pages according to set rules, when the rules needed to be broken the page would be displayed and a operator could decide what to do.

The last hotmetal page for The Sydney Morning Herald went to press in 1984. I retired in 1989. I read that in 1994 the Sydney Morning Herald had ordered a new SII Synthesis 66 Advertising system. I assume it included classified pagination. Whether it was interactive or not I don't know.

[Steve Young]

Absolutely. When DuPont and Fuji acquired Crosfield Electronics (with whom I was employed they having earlier bought Hastech) they brought a previously acquired company (Camex) to the table. Camex made a classified advertising programme that though overly complex and running under Unix fulfilled pretty much the same critieria you indicate here. During 1988-91 I was project manager of the El Correo Espanol and El Diario Vasco installation in north west Spain (Basque country). The project's aim was to bring all publications to total pagination using the Crosfield Page Magician, Editorial Magician, Camex classpage and Camex Starcasters (for output in the SPDL language, not Postscript ... Postscript was deemed too slow at the time though display ads were of course Pscript but "jacketed" in the ODBC).

This was eventually achieved after about a year and a half of intensive work.

The acceptance goal was that we could output 20 pages from scratch to bromide output within 30 minutes -- display ads were made up on Macs under Quark and passed to the Unix Object Database (pre-dating ODBC) and editorial input entered via the Magician terminals -- which we achieved .. a red letter day in my career!

The Page Editors made up the complete pages (under template rules for the most part though they could change these if required) after they had been though the ad page planners the pages were held on the DEC fileservers running the Crosfield system ready for the Page Editors. When all items were completed on the pages they were automatically passed via the ODB (picking up the complete display ads from the graphics sub system) and sent to output. The class ads were passed from the Magician terminals via the Crosfield system to the classpage terminal which was also output via the ODB to the Starcasters.

This installation was not without its problems as though we had the concept up front, a lot of it was "seat of the pants" stuff -- writing macros to run the classified input for instance -- I remember the Friday we first got output via classpage and right through the system without a break -- the newspaper systems manager came to me and told me that they had changed the entire way they wanted to do things. I didn't catch my plane home that Friday night but sat in a meeting till 2 in the morning hammering out workflow. I got home Saturday afternoon but was back in Bilbao early Monday morning. Took another couple of weeks to get back to where we were.

The day of the acceptance tests we had a full team on site -- specialists from every part of the process, from Boston, Madrid, Frankfurt, Sweden as well as the local Spanish agent from DuPont. I had produced an acceptance document over the weeks that had tick boxes for every process that we had to fulfil -- and every one of them came in on time. We were one minute shy of the 30 minutes!!!! The newspaper group took us all to lunch which lasted from 1 pm through to 8 pm ... and I thankfully handed the project over to systems support and went back to my family (until the next project which was in Porto Allegro in Brazil)!!

The Magician editorial and pagination terminals were the "clunky Crossfield terminals" that Tony Sutton talks about. Well they worked very well when properly utilised and were ahead of their time.

I worked on many other installations over the years, but none more satisfying that El Correo and El Diario Vasco -- two separate projects installed in tandem!!!

[Steve Young]

Whoops, brain freeze typing error ... "display ads were of course Pscript but "jacketed" in the ODBC)." I meant ODB -- our acronym for Object Database, not ODBC (object database connectivity) as I mentioned, not yet a standard for us as the database we used was of our own definition, not Oracle, Sybase et al ... that came a lot later! When DuPont rid itself of our system and got out of publishing altogether (selling their Howsan Algraphy arm at the same time) I had an interview with Mike Leigh of SII who was keen on getting the ODB technology (had worked for Mike in his Hendrix/Hastech days). This went nowhere -- understandably as technology was marching on -- and the Camex stuff was dumped and I started up Publishing Partners International (PPI) in the UK but in reality spent most of my time in the US. If Camex had had the foresight to publish SPDL (Supersetting Page Description Language) back in the late 80's it may very well have become a "standard" as it was faster (at the time) and not as cumbersome as Red Book Postscript. But they kept the technology to themselves and they went the way of the dinosaur.