RSS feedCategory Archives for Technology
A while ago, I talked about drip feeding items via a feed, basically providing a subscription date in the RSS URL, so the site knows how and when to drip items down the pipe, such as a daily novel or similar. Sometimes when I have multiple items to post in a day, I’ll put a later publish date on them so they get dripped out over the next few days instead.
Well I’ve just added an RSS feed for my links blog, which if you’ve been reading this main feed for over 3 months, you probably didn’t realise that I even had one.
Anything I see throughout the day that I want to keep a copy of, or which relate to my posts in some way, I drop into my links blog. What the feed does, is generate a single item for each day, and includes all the links that were added that day. In order to stop duplication throughout the current day, the feed only contains links from yesterday and earlier.
Yet another RSS hack. Fun. You’ll find the feed here: http://www.kashum.com/rbf.pl?c=linksrss.xml
Forget bubble wrap, forget popcorn, forget polystyrene. This turned up the other day in a box containing a hard drive mechanism. I’m guessing the point is that you can make it from recycled cardboard, but who knows…
Recent developments in blogging technology have been interesting to
watch, because many of them seem to match with the big content
aggregation picture I’ve been talking about over the last year or two.
CNet’s Newsburst, Amazon’s OpenSearch, and various discussions in the
videoblogging space… Time to start thinking out loud again.
On this site, I have the ability to filter my feed by category,
giving people a way to customise my RSS feed. I also have the abililty
to keyword search, but I haven’t publicised that at all. I should have
though, because now OpenSearch will take the credit for it.
Anyway, another interesting feature I have is consumer* targeting**.
Regular readers know that I’m completely biased in that an RSS feed
should be considered a pull, where the consumer is in control and not
the publisher. But if the publisher is going to have control, then they
may as well take advantage of it…
With my feed, I have an admin page that allows me to select
individual subscribers by IP address, domain name and user agent, and
then assign certain blog posts to only those subscribers which match. In
fact they’re matched with a Perl regular expression (pregex), so it’s a
little more advanced than simple address matching.
For example, if I want to send a message to just those subscribing
via Bloglines, I can assign the post a pregex that only includes it if
the word “bloglines” is in the user agent. Why I’d want to do that I
don’t know, but still, it’s possible to do.
A more useful example, is if I have a subscriber to my feed from a
certain organisation, say through their domain name and IP address. I
don’t have any obvious way of contacting them, because they’re virtually
anonymous, but I can selectively send them a message via my RSS
feed, by simply matching their IP address or domain name.
This opens up a new world of possibilities, such as delivery of email
via a feed, personalised feeds, and of course targeted advertisements. I
bet there’s a market in sending feed reader client advertisements to
people using other readers!
This all comes back to the feed being a generic supplier of content,
whatever it may be. We get confused when we talk about text vs.
enclosures, which is really like comparing paper with
boxes. Text is a generic carrier for language, not a synonym
for news item, and all text in a feed should really be typed in
some way. Thus we could have a distinction between plain news item text,
and a personal message, both delivered via a feed.
The text vs. enclosures stupidity continues once you realise that an
enclosure isn’t (sic). Enclosures are a reference to data which is
stored elsewhere in the network, and not stored within the feed, thus
the word enclosure is a misnomer. You can put text into an enclosure for
example, and you can even put web pages inside an enclosure, which opens
up some interesting business opportunities which I shouldn’t expand upon
here.
So is mail in a feed, first class textual feed data, the same as a
news item? Or is it an enclosure? If it is feed data, then shouldn’t all
enclosures be feed data? Or at least the references to them should
be?
This is yet again why RSS is flawed as a future carrier for content.
Atom seems to have failed at this as well unfortunately. Well, they’ve
failed at a few other things as well, but that’s another story.
Over the next few years, as typed content starts to become prevalent,
and the bandwidth issues disappear, either RSS will change, or some
other more appropriate protocol will step in and take us all to the next
level.
For now, if I see you pop up as a subscriber, at least I can send you
a personal message if I want to. And you thought subscribing to a feed was anonymous.
🙂
* I use the word consumer a lot in my writing,
and I have done for many years. When I use it, I’m referring to someone
who consumes information, usually via a feed of some description, and
not consumer in the traditional and arguably outdated sense, that some A-listers can’t seem to get over.
** I’m calling this feedmail or fmail, because it is the ability to send you mail via a feed, in much the same way that email is mail sent electronically. You’ll find a lot of hits in google for feedmail, due to an unrelated UNIX package of the same name. Mail feed by the way, from news feed, tends to refer to the entire feed, and not an individual item.