Archive of UserLand's first discussion group, started October 5, 1998.

Re: ScriptingNews 2.0b1 Comments

Author:Mark Nottingham
Posted:6/24/1999; 9:23:31 PM
Topic:ScriptingNews 2.0b1 Comments
Msg #:7808 (In response to 7613)
Prev/Next:7807 / 7809

*deep breath* OK. *smile* Sorry.

What are the content sources you're anticipating (just original content, or also weblogs?)

Ultimately, RDF is supposed to do this stuff, but IMHO it's too heavy for what SN/RSS et al seem to want to do (but I haven't had a good look at what they're doing, this is just an impression).

I don't know if it's going to be possible to keep things from fragmenting (RSS/ScriptingNews/MoreOvernews/etc), which is a shame. In an ideal world, I'd like to sit a process on my box at home that would suck in news and links from a bunch of servers and display it. If it has to understand too many formats, there isn't much benefit.

On a more practical/constructive note, my wishlist for scriptingnews is: * a publish/add date for each item (HTTP date format? or something else?) * an optional title for items (that can be a link. SN2 can duplicate the positioning of the link, but not the emphasis/context of a title). * optional categorisation per item? * optional publisher and publisher URL per item.

Publish date would allow us to display the last n items, or items since x time, or items in the last y hours. It would also make it possible to keep some client state, so that users don't see any single item more than once.

Categorisation is good if you're looking at a weblog, and you only want a subset of its content; for instance, memepool. It gets interesting when you start to consider whether categories are set by the publisher, or are pre-defined.

Publisher would be nice for the same reason, so you can tell where the link originally came from.

WebLogs add value by pooling together interesting online resources; now we need tools to manipulate what they produce. It's ironic, really; they were designed so that users don't have to go to each individual publisher to get content. Now, instead, you have to go to each WebLog that you're interested in.

There are responses to this message:

This page was archived on 6/13/2001; 4:51:03 PM.

© Copyright 1998-2001 UserLand Software, Inc.