home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!edcastle!dcs.ed.ac.uk!pwg
- From: pwg@dcs.ed.ac.uk (Paul Goldberg)
- Newsgroups: alt.peeves
- Subject: PAC-Learning Geometrical Figures
- Keywords: complexity, inductive learning, computational geometry
- Message-ID: <39842@skye.dcs.ed.ac.uk>
- Date: 24 Jul 92 16:09:52 GMT
- Sender: nnews@dcs.ed.ac.uk
- Organization: Department of Computer Science, University of Edinburgh
- Lines: 48
-
- !Peeve: This morning, the final version of my thesis rolled off the
- presses. Going against Rick's better judgement, I took it to the
- bindery to inflict savage paper cuts on unsuspecting bookbinders.
-
- Peeve: The Delay Paul's Thesis conspiracy. Two of the laser printers
- didn't have enough toner, so I sent an email to the technical support
- to replace a toner cartridge in one of them. There's another laser
- printer which is the only one which can do double-sided documents, and
- since I have to print it single-sided, I'd beeen reluctant to hog this
- printer. Given the likely delay in getting the toner cartridge
- replaced however, I now tried this third printer. It got as far as
- page 48 before stopping. The job remained in the printer queue, and
- the printer itself displayed no error messages. But - nothing further
- was printed. (It later emerged in an email I got from one of the
- system administrators that the printer had "somehow got confused".) I
- went back to one of the other printers and discovered that when you open
- it up there's a calibrated dial, which a little experimentation
- revealed was equivalent to the lighter/darker control on photocopiers.
- I was in business again. Although my first attempt to print the rest
- of my thesis worked, the pages after 67 had a ~1.5 inch horizontal
- strip about 1/4 of the way down the page where the print was
- noticeably darker, and also contained faint "shadows" of the letters
- contained therein. I managed to get an acceptable output from the
- other printer, albeit with the right-hand side of the page ever so
- slightly fainter than the left. An earlier peeve of mine on the
- subject of the printed output is the butt-ugly fonts for some of the
- mathematical symbols. Until a few weeks ago, the ones that seemed to
- be available for the natural numbers, real numbers, etc symbols were
- just simple, non-nonsense, sans-serif symbols. Then one night they
- suddenly changed these to some oversized, elaborate and hideous
- symbols, which if you weren't careful would end up touching commas or
- other punctuation marks that you put after them. And the two slanting
- lines in the natural number symbol aren't parallel. Be that as it may,
- I eventually caved in and used these fonts. Some people have their own
- fonts available so that their documents are not defaced by a whim of
- whoever manages the document preparation software, but I don't have
- time to work out how to do that stuff.
-
- Peeve: There is no rest for the wicked. In my thesis I stated a
- theorem and gave a sketch of the proof, omitting the full proof "in
- the interest of coherence". So I promised a complete proof in a later
- paper, and am now under a moral obligation to write that up.
-
- Paul.
-
-
- Voltaire to Rousseau, on reading Rousseau's "Ode a la Poste'rite'":
- "Mon ami, voila une lettre qui ne sera jamais recue a son addresse"
-