RSS Question
I notice that whenever my aggregator (still SharpReader) grabs feeds from some blogs, the Date for each entry is represented as the time the feed was downloaded, while on other feeds the Date for each entry is represented as the time the entry was originally published (or perhaps last modified?….nope, checked my own feed, which falls into the latter category, and it’s definitely time of publication).
What’s causing this? It appears that almost all of the “Date = time the feed was grabbed” offenders (and it is offensive) are weblogs powered by Userland software, so I’m guessing this has something to do with RSS 2.0 (which is prefered by Userland) vs. RSS 1.0 (which is the default on Six Apart’s software, Movable Type and TypePad)? Or maybe it’s a finicky feature of SharpReader? Someone enlighten me.
Whatever the reason, the question remains: why in the world would I want the entry’s time/date stamp to be the time of download? How useless!
There are two possible reasons this can happen:
1. The feed does not contain a pubdate per item
2. The feed does contain a pubdate per item, but the format of this date cannot be parsed by SharpReader.
SharpReader’s date parsing code can handle the date-formats in the RSS spec, plus many other often used date-formats. If you see a feed that specifies a date but you don’t see this show up in SharpReader, please send me an email and I will try to alter the code to also support this new format.
For most feeds, the cause will be the fact that no dates are set in that feed to begin with though… In that case you’ll have to contact the feed-author to change that.
Thanks, Luke. A little investigation taught me several things:
RSS 0.9x feeds (including my own) appear not to have any ITEM-level publication date, so I think I’ll be removing my own RSS 0.91 feed from availability.
Those 0.9x feeds and various RSS 2.0 feeds that, for whatever reason, haven’t implemented an ITEM-level publication date (either the RSS 2.0 <pubDate> or the Dublin Core <dc:date>) are, as you suggested, the source of my problem.
All of the RSS 1.0 feeds I checked out (most of which are probably auto-generated by Movable Type or TypePad users) appear to frequently use the Dublin Core %lt;dc:date> tag instead of the RSS <pubDate>, which I believe was one of the arguments of “funkiness.” But you know what — it doesn’t matter to the well-designed aggregator.
Most importantly, this is the first time I ever really scrutinized the differences between RSS 1.0 and RSS 2.0 feeds. It only cemented in my mind the utter ridiculousness of the “human readability” blather. Without knowing a single damned thing about RDF, I could look at an RSS 1.0 feed and immediately make sense of it.
People, if a reasonably technical English major like myself can quickly and painlessly decipher an RDF-using RSS 1.0 feed, why the hell is it so difficult for someone who’s actually (or supposedly) a developer?