Library clips

LiveMarks shows you not just recent items bookmarked in del.icio.us, but items bookmarked as they happen in real-time.

The column on the left shows popular bookmarks, and the column on the right shows the live bookmarks…wow, people are adding bookmarks every second.

[via Library Stuff]

web2con wiki by SocialText is re-syndicating the web2con tag from del.icio.us.

It is also re-syndicating the web2con Flickr tag, and also an (ego) Technorati link search to the wiki, although the Technorati Tags web2con, and Technorati search for web2con are missing

…although they don’t promote the XML icon of any of these 3rd party feeds.

I like the idea of making the wiki not only a gateway to other services, but acting as a portal where you can view content from other sources from within the wiki.

By the way here is the website to the Web 2.0 Conference.

ResearchBuzz points to her new keyword search tool, Kebberfegg that returns results in RSS (viewed as HTML) or OPML.

So what is the difference here?

Well it searches multiple sites at once, kind of like Talk Digger meta RSS search does for blogs, but it not only meta-searches blogs, it also covers web engines, news wires, tag sites, etc…this is all available in a drop-down category menu…there is a list of sources included in each category.
(TIP! hold CTRL down to choose multiple categories).

Seems to also accept URL searches.

Only thing is that you can’t choose which services from a category you want to include.

So from a search you have the choice of generating a HTML version which lists each individual feed from each service, or you can generate an OPML version so you can subscribe to the feeds from all of the services in one go, but still as separate feeds (just a more automated way of doing it, rather than importing a feed at a time.

But the only thing missing is a spliced RSS feed, so you only have to whack one feed in your RSS reader, this is where it is mainly different to Talk Digger.

So it is not a re-mixing tool but a meta-search tool that returns feeds, it just alleviates the task of repeating this search at all the different places…the OPML offering tops of what this service is about, convenience!

Also covered at SEW blog.

Top sources of del.icio.us links is a place to see the top sources according to all of your del.icio.us links.

This way you can see the sources that matter the most to you, although as said in a comment to the Nivi post:

“The problem is that Del.icio.us captures what you’re linking to, but not how you found it”

The other issue is that even though you bookmark a lot from one blog (as you very much like that blog), if they don’t post frequently then they won’t rank well.

Anyway the idea is to capture the sources of your “attention stream” so you can cut down to only the feeds that matter to you, relieving a bit of RSS stress.

As the post says:

“Companies like searchfox want to watch your attention stream and use machine intelligence to help you find high quality information.

This idea here applies your own intelligence to help you find high quality information no matter what news reader you use.”

This isn’t the first time this has been thought about…see introsp.icio.us…more.

Here is my top 30:

masternewmedia.org 47
thecommunityengine.com 21
techcrunch.com 20
blog.mathemagenic.com 16
corante.com 16
marketingstudies.net 16
blogs.salon.com 15
theshiftedlibrarian.com 15
librarystuff.net 14
urlgreyhot.com 14
commoncraft.com 12
incsub.org 12
myst-technology.com 11
knowledgeboard.com 11
ariel.emeraldinsight.com 10
rss-specifications.com 10
rssdiary.marketingstudies.net 10
radio.weblogs.com 10
blog.jackvinson.com 10
tametheweb.com 10
globelogger.com 9
denham.typepad.com 9
blog.searchenginewatch.com 8
llrx.com 7
weblogg-ed.com 7
infotoday.com 7
readwriteweb.com 6
pasta.cantbedone.org 6
headshift.com 6
corporateblogging.info 6

So I guess whether a blog from your RSS reader makes it high on this list depends on whether the site is a pointer to stuff, or has interesting stuff of it’s own.

You could really love a blog, as it points to great stuff, but you don’t really bookmark pages from it because the posts don’t add value of their own, the value is in pointing you to great timely stuff…that’s why I wasn’t surprised to see Micro Persuation isn’t in this list even though it is one of my essential blogs that I read daily.

Also as you can see from my list, besides the top 3 or so, I don’t bookmark many pages from blogs that I heavily read…I wonder why, is it that the blogs I read always point to other stuff. So from my 1000 or so bookmarks in del.icio.us the source count of the total links is very high, meaning I bookmark stuff from mostly different sources, I don’t bookmark a lot of stuff from the same website/s.

As mentioned before, more importantly how do we capture how we found most of these resources…well we probably know that, as these feeds are on the top of our RSS readers, we know the feeds we like…but maybe a lot of the stuff you bookmark is also being pointed to you from various feeds in your RSS reader.

I wonder if there is a way to determine which feeds pointed you to those pages you bookmark in del.icio.us, this is hard as you are not neccessarily bookmarking a page from the feed, but a page the feed is linking to (telling you about)…maybe this would be easier to work out if your RSS reader and bookmark manager were in the same system, like Rojo.

Then again does it matter as personalised RSS Readers like searchfox, Chameleon for Bloglines (this ranks feeds not items), Findory, etc…examine your reading behaviour and bring what’s important to you personally to the top of your river of news. This way you don’t have to care about which sources bring you personally relevant stuff, as you have a little agent working for you, so your reading is ranked by personal relevancy at the item level.

But this is ranking your reading behaviour…how do you work out the ranking of sources (feeds in your RSS reader) according to the frequency of bookmarking pages they point you to?

All these 3 ways of analysing your reading, and bookmarking behaviour are all useful to help reduce the overload or increase relevancy (both recall and precision)…ie. they are not eliminating any content, they only float, what they think is relevant to you, to the top.