-
Website
http://www.scripting.com/ -
Original page
http://www.scripting.com/stories/2009/05/24/googlesIncompleteSupportOf.html -
Subscribe
All Comments -
Community
-
Top Commenters
-
eas
55 comments · 4 points
-
AndrewBurton
134 comments · 10 points
-
Michael Markman (Mickeleh)
154 comments · 16 points
-
Rex Hammock
52 comments · 9 points
-
malatmals
81 comments · 3 points
-
-
Popular Threads
-
How I develop formats and protocols. (Scripting News)
1 day ago · 11 comments
-
Open is in the eye of the beholder. (Scripting News)
3 days ago · 13 comments
-
Store Twitter URLs in earth's oceans? (Scripting News)
5 days ago · 16 comments
-
Why today's Twitter is like Napster in Y2K. (Scripting News)
5 days ago · 15 comments
-
If you wrote the words you own the copyright. (Scripting News)
5 days ago · 7 comments
-
How I develop formats and protocols. (Scripting News)
It's similar to the way I built the backend of podcast.com where users can create folders with rss feeds in them. Also they can create 'shortcuts' to other people folders of feeds (same as reading lists) - so when the folder owner changes the feeds in their folder, my subscription to that folder reflects that too - and then I can see the latest item (in a river feed) from any folder I like - no matter what feeds are in it. - here the opml from my account http://my.podcast.com/kosso/opml/
Since I left there, it's shame they haven't built on the system to support any old rss/atom feed for blog posts/news etc. - all it would take is a different UI. ;)
My family still describes what I do as "Michael has a software company, I still have *no idea what it does*." I take it as a personal failure, but I've tried everything I can think of, every mental model, analogy, way of describing it, and it still does not 'stick'. Even after going through the exercise of building related reading lists for topics people are passionate about, it still doesn't seem to gel.
I still think it's a hugely powerful concept, but I don't think the use case / mental model has yet emerged for non-technical people to 'get it'. Just my humble opinion. :)
read an article about it. I don't think families were ever meant to
undestand what we do Mike. :-)
http://www.google.com/reader/public/subscriptio...
The problem I see (which might be what you mean by "incompatible format") is that that there's no obvious way to subscribe to external reading lists in Reader. It's an OPML producer, but not a consumer.
Importing OPML is a one time thing, subscription is ongoing.
them, I asked -- and he said they are indeed dynamic.
Is Kevin still around...?
Geeez
The HTML, OPML and Atom outputs of the bundle are dynamic, and update on editing, others' subscriptions to them in Reader don't. Your comment system ate my links before, so here are my examples via bit.ly - HTML bundle:
http://bit.ly/comics_html
OPML version of that bundle:
http://bit.ly/comics_opml
Atom version of that bundle:
http://bit.ly/comics_atom
If you subscribe to the Atom feed in Reader (or any other feed reader) the it updates dynamically, but if you add the OPML or the bundle link, you get the individual feeds added.
So Reader has a way of creating reading lists in OPML, but only reading them as a unified feed.
http://www.google.com/reader/bundle/user%2F0347...
And here's the OPML link from that page:
http://www.google.com/reader/public/subscriptio...
Google just shipped an editor that generates OPML in Reader. What I was clumsily saying was that this means people who aren't users of your OPML editor can now generate and edit OPML easily.
Both those links will dynamically update as the bundle is edited (I just added Foxtrot, and they did). What I'm not sure works is that if I subscribe to that bundle in Reader, whether it updates dynamically too. I'll check that with the Reader team.
The dynamic bundle of separate feeds would be really cool but I don't see it in the announcement or anywhere in Google Reader. You say you've seen a bundle dynamically update between producer and consumer of the bundle? The web page will dynamically update when you edit your bundle but the OPML file or bundle shared within Reader isn't dynamic. So I guess the dynamic thing is really no more than publishing a read only web page of feeds rather than a portable set of feeds. What is nice (and maybe this is what you're referring to?), is that as well as a static OPML file, you also get an Atom feed from the bundle, so you can aggregate multiple feeds and re-publish them as a single atom feed for others to subscribe to. Until now, I've been using Yahoo Pipes to pull multiple feeds together and re-publish as a single feed. Using Google Reader seems like a better solution.
That's not the important issue anyway -- I can distill it even further. Here's a reading list:
http://static.newsjunk.com/tech/main.opml
Your mission, should you decide to accept it, is to subscribe to that in Google Reader.
If you can, then this was just a communication problem, if you can't -- then the headline of this piece is correct. You've only got half the implementation of reading lists, and everything I said about creating reading lists with tools other than Google Reader is correct.
To be clear, I'd rather be incorrect in this case. I'm not trying to prove Google is evil, I don't care if Google is evil. I care about compatible products, and not locking me or any other users in. I like my suite of RSS tools, but I don't see why GR users shouldn't be able to benefit from my curatorship, and that of my colleagues. That's the power of open formats.
Dave, I found you through Steve Rubel's ("aggregated", "bundled", "reading list") feed which he calls "thinkers". Note: this is NOT his "shared items" in google reader. It is a share of a single folder/tag which is separate to shared items.
http://www.google.com/reader/public/atom/user/0...
This technique seems to do everything you mentioned up to the point where you started talking about extensions to RSS.
Isn't that what you want?
whole idea of RSS is that users have choice, any feature like this that only
works in one product is a travesty. It's absolutely wrong. Hopefully a
pretty clear statement. This is what I mean about Google doing whatever they
can get away with as opposed to doing what's right and good design and good
for users.
I think for "reading lists" where you are delegating the content selection to a "curator", the output should be a single feed, not opml. Don't you just want the articles/posts? Why do you even care about what individual feeds comprise it?
On that basis I don't agree with you.
http://bit.ly/11jLLw