The Daily Whim

The Daily Whim

Sun. Dec 22, 2002

Things Reid Doesn't Understand About RSS

Things Reid Doesn’t Understand About RSS – Some relate to the concepts of RSS, and some relate to reading the feeds of others. Most are probably the result of a lack of knowledge, and/or the time to gather it.

First, what I think I do understand about what RSS “is”: it’s another exemplification of the benefit of separating content from presentation. It allows someone to sample the 150 KB full presentation you see on my home page in a 17 KB text-only summary (which looks like this in AmphetaDesk), providing a substantial benefit of bandwidth. It also provides a time benefit (at least it does for me), allowing me to sample over fifty sites from one web page. It is another layer of the concept of ”weblog as web filter”; I trust certain people/sites to point me to interesting links, filtering subjects and sites I don’t have time to visit. Using RSS syndication and a feed reader, I can further filter/summarize 50 or more of those sites onto one page.

But here’s something I don’t understand: What’s the point of including graphic files in your mostly text feed? Now, I’ve thought about converting Pixel Pile and adding a RSS feed of it, but how popular would a 40-50 KB load of four images be as an RSS feed? At least people would know, hey, this is a photo syndication, but lately I’ve encountered imagery in RSS feeds where I wouldn’t expect them.

It’s no big deal to me, since I have a DSL connection, but the latest article from Meg Hourihan made me think it might be counterproductive. She spent a month on vacation in Paris suffering with a very slow connection (poor baby), and had some “Dial-Up Revelations”: “Using an RSS reader to catch up on a variety of weblogs is the difference between reading those entries (or whatever portion I can get through RSS) or not reading them at all.”

The bandwidth and time savings of RSS were a clear benefit to her, but throwing unexpected images into that mix would seem to negate some of that benefit. However, Meg seems to contradict some of that bandwidth savings with one wish: “The third wonder of my dial-up world is NetNewsWire Lite, an RSS reader for the Mac [...] Whatever I can get through NewNewsWire is what I’ll take. In the case of weblogs, those writers that syndicate their entire post, rather than an excerpt, have become my new best friends.”

When I switched from RSS 0.91 to RSS 2.0, I encountered this issue of full content vs. excerpt, in a big way. At the time I did it, I had some lengthy posts on my home page, including a 3,000 word screed. In the RSS, all 3,000 words were there, bloating the feed to about 50 KB.

It seemed contradictory to what I understood as the concept of RSS. One of the big changes from 0.91 to 2.0 is the addition of the [CDATA] section, containing the full content. The “excerpt” is still there as well, so perhaps this is as much an app issue as a feed issue. I’d like to be able to choose if I see the excerpt or the full text, perhaps even on a feed by feed basis. But in Ampetadesk, the default is to show the full content. Not wanting to force people to download the entirety of my occasionally verbose venting, I changed the [CDATA] so that it too includes only the 50 word excerpt.

Now, from what Meg says, that means I’m not one of her “new best friends.” But, dang it, it just doesn’t seem right to me. In fact, in some ways it doesn’t seem right that an RSS feed would include both the excerpt and the full content, as it substantially increases the size of the feed. I know, there are probably simple changes I could make to my feed to eliminate this dilemma (and I welcome any suggestions … is the section even required in RSS 2.0?), but in viewing the feeds of others, it’s clear most people aren’t fiddling too much with the default templates they get with their CMS, be it Moveable Type, Radio, or whatever. In fact, there are a few feeds that regularly show no items, because of some malformed content that causes Amphetadesk to barf on parsing the feed (I used to have this problem, too, until I added …encode_xml=”1”... in the appropriate places in my MT template). For many of us, RSS is a relatively “new” format, and we’re woefully uneducated. My feed validates, but is that enough? Why not treat it like learning HTML, and learn by “viewing source”?

Well, there seems to be little consensus, as people take nearly as many differing approaches with their RSS as they do with their web logs. Of the 53 feeds I currently read, 35 of them are 10 KB or less, 4 of them are over 20 KB, and two are over 40 KB. I’ve seen feeds that are entirely linked text, and I’ve seen one with not a single link, just short text excerpts, forcing you to go to the full HTML page to get anything more than the first ten words or so. Doc Searls syndicates his full entries, but only the latest three. Mark Pilgrim, AKA “The Man” when it comes to RSS, syndicates the full entry, in the same number they appear on his front page. And these 53 feeds come in multiple argumentative flavors, 0.91, 1.0, and 2.0.

Perhaps we need someone like Microsoft to create a feed reader that the masses will adopt and impose some proprietary standards so we can all get on the same page.

I’m joking, I think.

Peanut Gallery

Comments are closed for this article
Contact me to find out more