oasis-library-review — archive

[Robin Cover]

On Mon, 31 Mar 2008, Neil Schelly wrote: > On Monday 31 March 2008 10:41, Robin Cover wrote: >> Sorry, I should have documented the statement >> >>> the Naming Guidelines place >>> no constraints upon TC members as to the use or non-use of the substring >>> 'spec' in a filename, whether that file(name) is associated with a >>> designated "spec-track" document or not. >> >> Here's the principal section: note, e.g., the absence of any directive >> about (non-) use of the substring 'spec' (or 'specs') in any filename > > I'm not looking to have that discussion over email. I've read the docs. I > know that the word spec is not some reserved keyword. I just setup to > demonstrate naming guidelines functionality by making a rule that's easy to > see and play with. This is a proof of concept and nothing more. OK, thanks for clarification. Makes sense. In terms of a broad heuristic (and in reference to what you wrote) about a prototype POC: > You can't publish a non-spec document that has the word spec > in the name. You can upload it, but you cannot publish it > until you rename it. I don't recall any actual "word"-based constraints of any kind, at the level of filename component (other than general commensurability and the filename "extensions" [which are not words]). But I could be forgetting something... As a test, your proof-of-concept would presumably show substring regex parsing for filename (substituting '--' for 'spec') -- that a user could be allowed to upload a file with filename "rights--rule05.html" but (because of naming rule violation) the system would not allow publication of the resource matching this filename until the filename error condition was addressed. Ironically (maybe) the simplest part of the library application is what people frequently talk about as the "upload" tool(s), under the rubric "file upload." While the system needs to manage files at some level, the real intelligence of the system (and the challenges to use of shrink-wrap boxed software) is the need to manage information at levels higher than "file", where the key semantics of the application space are defined. That's what David Krauth worked on a good bit -- because it's way harder (and more important) than "file" management. More on this anon... - Robin > > -- > Regards, > Neil Schelly > > W: 978-667-5115 x213 > M: 508-410-4776 > > Senior Systems Administrator > OASIS http://www.oasis-open.org > "Advancing open standards for the information society." >