Re: Metaclass & McCLIM revisited
"Tim Bradshaw (as tfb at cley dot com)" <[email protected]>
| Newsgroups | gmane.lisp.lispworks.general |
|---|---|
| Message-ID | <[email protected]> |
After decades of searching I have discovered the golden rule of CLOS. It is this: If you ever define a method which specialises on any class which is not either a structure-class or a built-in-class you have made a mistake. You may think this is contentious or extreme, but I can assure you it is not. It was handed to me by a mysterious figure on the peak of a mountain in a distant land which I had reached only after three years of wandering, accompanied only by my faithful, battered F2 and a small bag of film. The being – cloaked as it was I could not make out its face, if it had a face – said only 'obey this rule and you will come to no harm' before turning away. The rule is written in an ancient language, inscribed in letters of unearthy fire on a slab of polished adamantine. With some difficulty – it is curiously heavy, as if it protruded somehow into some other dimension – I carried it home with me where it sits next to my computer. Sometimes it seems to cause strange distortions in the display. There is often a curious muttering. On the back is inscribed a word: it is badly damaged and partly intentionally obliterated, but seems to read 'Gr..or'. I wonder if it is a signature of a kind: the unutterable name of the entity responsible for the rule? I can reveal no more: I have already said too much. The stone is angry: the chanting now clearly audible, the image my screen deformed beyond legibility. I fear I have awoken it.