Many, many, years ago, I had both the leisure and the motivation to produce monthly reports on my activities. Many of these documents have long since vanished, but a few survive. They present a strange world in which some names and terms since very familiar jostle with others long since obscured by the passage of time. In preparing them for the web, I've added minimal formatting and (where I can find them) links to any other documents mentioned or related, but otherwise these are unrevised and unrepentant instances of the written persona I cultivated at that time.
Owing to the less than perfect behaviour of the 2988 this summer, I have had at least three opportunities to tidy my desk since the last report, during one of which I wrote out a list of all the little jobs I should try to complete before the date of this report. I abandoned this quixotic venture with number 13 and threw away the evidence: what follows is what actually did get done.
A paper about the wonders of CAFS/DCI for text handling, counting semicolons in the complete works of Shakespeare, searching Bodley's pre-1920 catalogue etc etc was handed over for publication in the next ICL Technical Journal, only a week late. Paul and I were invited to address (a) the ICL Large Systems User Group (Exeter, October, free dinner and "Late Night Entertainment" sic) and (b) the annual CAFS lunch (Cafe Royal, December, champagne). Work on the ICLCUA CAFS WP Report Mark 2 proceeded, at first intermittently, now hastily and next month no doubt at a rate of knots. A complete draft (more or less) has now been assembled on the 8801 and I shall now start trying to make it look coherent before it is typeset.
Most of the summer seems to have been spent worrying about and even working on the Ashmolean prototype database. Some effort went into various amendments to data definitions and documentation; most of it went into grappling with Application Master. This mental equivalent of sumo wrestling is one of those 4GLs you may have read of in the popular press, which is supposed to improve Programmer Productivity by 99% overnight. I'll let you know.
Two major bugs encountered by QM/CAFS were reported to ICL; reps were supplied eventually and one of them (but not the other) appears to have worked. I was disappointed to discover that the whole filestore couldnt be made CAFS searchable as well as NOT CAFS searchable, if unsurprised.
* Spent about a fortnight on and off getting Bodley's pre-1920 catalogue (S to SHERIDAN) into shape. This eventually produced 8 Mb of searchable file, containing lots of fascinating information, which the Shakespeare mob are now playing with excitedly.
* Spent three days getting the Canterbury Tales into shape and sending it off on a tape to form part of a DCI demonstration package. In return, we should be getting a pre-release of DCI-200 (which can access IDMS databases) later this year.
* Spent a morning persuading :KINLEN to stop mucking about with tapes and use DCI instead. Her eyes lit up, as well they might.
* Spent about five minutes (every day for a week) helping the RSL's wonder boy programmer over various hurdles encountered while getting their serials file into shape.
* Converted :DIGGER from PDS to QM in late July; he hasnt looked back since.
Had a week in Gottingen where I met various German historians and signed an historic international agreement on cooperation between text archives. Was visited by N.Calzolari (Pisa), K.Watts (Kidsgrove), R.Oakman (S.Carolina), N.Ransome (Ohio), M.Larsen (Aalborg) et of course al. Visited OXFAM in search of AM enlightenment and saw a DM1.
Version 2.3 was ready for distribution in time for Jeremy to take a well earned two weeks holiday (most of which he seems to have spent massacring ducks) before being re-incarnated as OCP programmer. I produced a new version of the manual, incorporating all the suggestions made to date, except Susan's which arrived shortly after I'd handed it over to Lynne. I distributed 40 tapes, one of which has been returned with thanks so far. The new version will be handed over to Edinburgh next month.
The usual steady drizzle of network mail, letters, dud tapes, missing cheques, silly questions continued throughout the summer. Notable acqusitions included Dante's Divine Comedy which I eventually translated into a usable format. Couldnt find any jokes in it though.
1. Successfully wrote a Spitbol program for converting Protestant Cemetery Gravestone Inscriptions into not just SIF, not just multiple SIF, but Multiple Sif With Trailers!!
2. Thought about CAFSizing some Greek in response to unprecedented demand (four different people in succession in Advisory) and asked Charles nicely about getting Greek output on the Xerox Printer.
3. Converted two more PDS databases to Querymaster, uncovering another bug in our bootleg PDSTOQM software in the process; also fell foul of a bug in QM, the patch for which Paul found by accident when looking through the KEL for something else.
4. Encouraged :KINLEN (actually Lucy) to move her Scottish Deaths onto the CAFS discs; I think this is now nearly complete, thus reducing the number of tapejobs run on the 2988 by 40% at a stroke.
5. Tried unsuccessfully to get Application Master to produce SIF. The first failure was caused by forgetting that SCL INTs aint the same as proper ones; the second is still sub judice.
6. CEP demonstrated that four concurrent CAFS searches took roughly twice as much elapsed time (each) as one. Successfully cafsized a further major chunk of Bodleian catalogue (SH to TH), once Rob had discovered and fixed bug in OUCS tape handling routines, without my having to get very rude, and I had ditto ditto in my conversion program, without Malcolm's ditto ditto.
7. DCI-150 was installed temporarily, withdrawn in such a way as to stop DCI-100 working, and has now (I hope) been reinstated. Tested one of its new facilities (case-blind searching) but not the other (access to IDMS databases).
8. Drafted (and re-drafted) M-series user guide on Using CAFS, featuring full frontal details of CSO, dark hints about DCI and oblique reference to QM, i.e. exactly the reverse of the official ICL profile for these products.
9. The ICL CAFS Working Party Report was finally published, with jubilation tempered somewhat by the appalling job the printer had done on the paste-up. The annual CAFS lunch, (and its rehearsal at Beaumont) occurred. Paul and I were magnificent.
Final version of Famulus77 Manual went back to Documentation for printing. Another final bug in Famulus emerged. I passed it to Edinburgh! About six million letters for the Text Archive, some containing orders, arrived and I answered some of them. Played a bit with HUMBUL (not a misprint for the Christmas Spirit but the Humanities Bulletin Board at Leicester). Toned down a review I wrote for the new ALLC journal which the editor considered too rude for publication. The uncensored version is still available, if anyone's interested.
[ Documents from this year not otherwise linked include :