- From: Pat Hayes <phayes@ihmc.us>
- Date: Sat, 6 Oct 2007 09:32:27 -0500
- To: noah_mendelsohn@us.ibm.com
- Cc: Tim Berners-Lee <timbl@w3.org>, Susie Stephens <susie.stephens@gmail.com>, Technical Architecture Group WG <www-tag@w3.org>
>Tim Berners-Lee wrote: > >> I agree that that is a terrible word. It is an abuse of the english >> word. (Its use is is historical: when Universal Resource Idemtofiers >> and Resource Description Format were dreamed up, the communities were >> very much thinking of Information resources, such as educational >> resources available on the WWW). >> >> So, Pat, what would be a better word which we should use instead? >> The class of all ____ s? The class of which all classes are >> subclasses is the class of ____ s? > >Tim: I'm a little unclear in how far you would propose to go in changing >the terminology? Would you have it in mind to go over all the >Recommendations, Notes, supporting documentation and Web sites that the >W3C has created and rewrite them to use whatever word is selected as the >best alternative to "resource"? Until quite recently, and everywhere except for a few of the more recherche TAG writings, "resource" in most of this literature in fact means "information resource". And keeping "resource" as an abbreviation for IR seems quite sensible, and conforms to natural English usage and to historical precedents. >Would we promote use of some new >initialism such as "UI" (which I think you suggested at one point) as an >alternative to "URI"? URI could be glossed as Uniform Rigid Identifier, which in fact is quite a good description of its intended use. The middle word is then an adjective describing the nature of the identification rather than the type of thing identified. But in any case, acronyms (especially if easily sayable) often outlive their generating abbreviations. How many people now think of 'laser' as a pun on the acronym 'MASER'? >I'm not necessarily saying that doing all this is a bad idea, but the cost >of doing it now will be very high in my opinion. The alternative seems to >be having normative materials that use both terms. > >Out there in the world are books and other educational materials, APIs and >their supporting documentation, etc. all of which, if they've followed the >path we've suggested in recent RFCs and AWWW carefully refer to "resource" >in this somewhat odd sense, and encourage its use as for the referents of >URIs. I'd be interested to see how many of them actually do this coherently. I get the overwhelming impression that httpRange14 has generated more confusion than accurate documentation. > Earlier this month I taught a course to 150 IBMers, carefully >explaining how the word resource is used in the context of the Web. There >will be some cost and some significant confusion in going back to such >communities and explaining that we've changed our minds. With respect to IBM, 150 IBMers is a pretty small group compared to the entire rest of the planet. Its easy to forget how very small our technical community is. And just think how much LESS care you would have to use next time, and how much time you will save by not having to force people to learn to speak W3C-babble instead of English, and how much time they will all save when trying to talk to everyone else. >So, in addition to the important and amusing game of picking the best >alternative word, I think when we then have to think very hard about >whether promoting use of that new word in place of "resource" will on >balance be a service to the community. It will be of incomparable service to the community, if we include the future in that community. > FWIW, just at the moment, I'm not >convinced that a change is the right thing to do, though I would easily be >convinced that words like "thing" would have been preferable if chosen >from the start. For now, my leaning would be to explain that for >historical reasons we're stuck with the word "resource", and apologize for >the fact that it is now applied for some purposes that are quite a stretch >of its conventional meaning. They are not just a stretch. They are actively misleading, and cause people to misunderstand things. > It's far from the first time this has >happened in the computer business. Indeed, the word "computer" itself is >a bit of a stretch of its initial use, which was to refer to people who >did computation (actually, I think there was a phase in which one referred >specifically to "electronic computers", which I suppose would be analagous >to referring to "Web resources"). Yes, I actually remember "electronic computer" as the normal term, before the PC made them into consumer items. But this is a poor analogy: modern computers do in fact compute, just as people once did. The meaning of the word hasn't fundamentally changed, but the world has changed. But to call everything under the sun and moon a 'resource' is just, well, silly. Most of them aren't. I don't think this change has to be a great revolution and upheaval. It can be done easily and gradually. Just start saying that the meaning of the word 'resource' has been stretched rather a lot, and it might be a good idea to call non-information resources - which can be just about anything - simply 'things', and then do that in the rest of the document, and keep "resource" for information resources. Im sure that this will be seized upon with immense relief by the community and become normal practice in a matter of months, with no effort at all. Pat > >Noah > >-------------------------------------- >Noah Mendelsohn >IBM Corporation >One Rogers Street >Cambridge, MA 02142 >1-617-693-4036 >-------------------------------------- -- --------------------------------------------------------------------- IHMC (850)434 8903 or (650)494 3973 home 40 South Alcaniz St. (850)202 4416 office Pensacola (850)202 4440 fax FL 32502 (850)291 0667 cell phayesAT-SIGNihmc.us http://www.ihmc.us/users/phayes
Received on Saturday, 6 October 2007 14:32:53 UTC