- From: Scott E. Preece <preece@predator.urbana.mcd.mot.com>
- Date: Thu, 17 Oct 1996 10:55:19 -0500
- To: mag@ncsa.uiuc.edu
- CC: www-html@w3.org
<<This is getting a little more marginal for this mailing list, except that it *is* a Web problem -- we all have to have little mapping files that try to guess file type from file names, and we also have the problem of whether links should specify the type of the thing linked to or the thing should assert its own type, which does impinge on HTML>> From: mag@ncsa.uiuc.edu (Tom Magliery) | >Why, oh why, has our industry (with the notable exception of Apple) been | >unable to figure out how to do typed files? | | Maybe it's my own ignorance (remember, I'm a Mac user, so I'm on *your* | side), but what's the difference between storing the information in a | 3-character extension + a mapping somewhere in the OS between extensions | and applications, versus a 4-character "creator" field + a similar | OS-internal mapping? --- Aside from esthetics (making the type visible as part of the name of the file is unattractive), there are a couple of problems: Embedding the type in the name is fragile (it's easier to change the name accidentally than changing an embedded property of the file). Names may not be carried through transport mechanisms (internal types may not be, too, but it's more obvious that they need to be). Names are less controlled. For typing to work there needs to be some kind of control on what specific types mean (else you get the overloading problem I mentioned for your ".doc" example). It's equally possible to control the space of extensions, but it's less likely to work, since users tend to think of that as something they can control. NOTE: all these assume that the name is user constructed; if the extension part of the name is an outward sign of the system's knowledge of the type, that is a different situation entirely. scott -- scott preece motorola/mcg urbana design center 1101 e. university, urbana, il 61801 phone: 217-384-8589 fax: 217-384-8550 internet mail: preece@urbana.mcd.mot.com
Received on Thursday, 17 October 1996 11:55:17 UTC