Library clips

I’ve posted twice about automatic link searches at the end of your blogs post (via the post template)…instead of launching to each site maybe you could graze the feeds of the link searches from each engine within the blog post itself…I think Grazr is running with this.

Another thing, at the end of each post people usually list several Technorati tags, why do they have to launch to Technorati to see similar posts, why not graze these in the blog post as well.

It’s simple enough to manually add a Grazr box or a link to a Grazr box on your blog post, the hard part is to code this for the post template so it automatically appears when you add the tags.
If you have more than one tag, it has to create an OPML file to store them and then put them in a Grazr URL all automatically.

EdTechPost points to an amazing newsmastering portal called postgenomic.

They seem to use a lot of web 2.o techniques: blogging, feeds, aggregation, mash-ups (connotea, google maps), tagging, voting, microformats, etc…

This is no ordinary newsmastering, even though most of the content is populated by the house blogs, it is not just a river of news, spliced feed type of thing…it’s not even a river of the latest news sorted popularity or relevancy.

You won’t see every post by every blog as a river of news, what you will see are posts by blogs that are linking to the same website…if a lot of the house blogs are posting about a journal article or a blog post, etc…then this site being linked to will be an entry, and the house blog posts will be listed underneath this entry.
This is basically an inbuilt chuquet type engine with more…where selected content in your feed set is displayed if lots of others are blogging about the same website, journal article, blog post in question.

In addition to what sites people are talking about, what about what topics people are talking about…this would require some text analysis…Anjo is working on this.
Something more simple but effective is the popular keywords or word bursts…from the webpage:

“If a word is suddenly seen more frequently than expected then that’s a sign that it might be related to a new, interesting paper or a news story that has just broken. “

There is so much value on the front page: current hot papers, current hot stories, impact factors, top keywords…feeds everywhere.

See the about page for more.

Blogs

Here is a list of blogs that provide content for this portal…search or browse by a category group, such as Bioinformatics.

Where’s the Reading Lists, this portal has everything except for OPML.

Profile

Here’s contributors profile, see HubLog
…it displays recent posts, and papers cited.

Stories

“Popular stories are recent news items or blog posts that have been linked to by multiple blogs”.

As mentioned at the start of this post, what it does is look at all the hyperlinks in each blog post, if several blog posts are linking to the same website then it must be popular so it makes it to this Popular Stories page, and all the house blog posts talking about it are listed below…it has a feed if you want to follow this in your RSS Reader.

eg. 4 blog posts in postgenomic linked to this article, Web 2.0 in Science…this article came from nature.com.

It will show an expert from each blog post, with a link to the native post…it also links to a profile of the blog…at the end it will show a list of possible tags (these tags are only there because people have saved and tagged this article in Connotea, which allows us to combine features of Connotea within this newsmastering service).

One of the tags used for this article in Connotea is “web 2.0“, here you will see the Connotea information represented in postgenomic.

You can also apply filters…the article above came from nature.com, so we can filter to see popular stories from nature.com.

You can also filter by the blog category (all blogs in postgenomic are grouped into categories), here is popular stories within the category bioinformatics (blogs that are about bioinformatics)
…is this starting to sound like Technorati Explore, well it’s not!

NOTE: you can’t use 2 filters, eg. popular stories from nature.com within just bioinformatic blogs.

Technorati Explore will show the latest posts from a group of expert topic blogs, whereas postgenomic is only interested in showing posts from these expert blogs if they are linking to a same website, so for the house blog posts to get featured they need to be talking about the same website.

In essence it will show a list of stories that these blog experts are linking to, and under each story it will list the excerpts of the house blog posts…then you can filter or limit to see if there are any other popular stories from that website.
eg. blog post A, B, and C are pointing to a blog post or webpage or news item from D (this is a blog or website or newspaper, etc…), then you can say are there any other stories from D that blogs in postgenmic (A, B, C and the rest) are pointing to (this part is the filter).

Rojo could display this data the same, they call it “Recommended Links”, SharpReader will do a basic type of threading in the normal river of news, whereas postgenomic has a devoted section.

I’d like to see this in Rojo, but for every post in the normal river of news section (like SharpReader).
If I’m reading a post which points to webpage A, and other read/unread posts in my RSS Reader also point to webpage A, then thread it at the end of the current post. I don’t want to have to keep reading my feeds to find out 5 minutes later another post also points to webpage A, then 10 minutes later of reading I see yet another post points to webpage A…that’s 15 minutes of reading to realise 3 posts in my RSS Reader point to webpage A, when I could have found that out from the start via threaded aggregation.

The problem arises with frequently linked to pages
…if the current post you are on points to a popular page like the del.icio.us home page, then you will have heaps of read/unread posts in your RSS Reader that point to del.icio.us home page threaded at the end of the current post.
What we would need is an expand/collapse link, and also maybe a date range.

Mini-blogosphere

We have established that popular stories aren’t neccessarily popular stories from the blogs within postgenomic, they are popular stories because blogs in postgenomic have linked to them…if lots of postgenomic blogs point to a blog post from another postgenomic blog then this too can be a popular story just like external popular stories.
To see this we can filter popular stories from indexed blogs (maybe this could be called postgenomic blogs), this is a marvellous idea, it shows the blog conversations going on between just the postgenomic blogs…its own mini blogosphere…awesome!

Popular stories get more popular

It also seems that you can vote for stories so they can climb the list…this is popular x2
(also designates if a story has been bookmarked on Connotea)

Papers

The reason I didn’t include journal articles or conference papers in the Popular stories section is because they have their own Popular section called Papers…the sidebar also lists the impact tracker (Which journals are bloggers linking to most frequently?)

Filters are the same as the Popular stories section

…this is great, 2 Popular sections by document format.

Reviewed Papers

This section is aggregated via microformats, similar to blog post aggregation in edgeio
…from the about reviews page:

“If a blog post is about one paper in particular then the post considered a review […] Reviews are identified by special markup in the blog post. The first option is to mark an anchor tag that links to the URL of the paper with a rev=”review” attribute […] Your post will be picked up and should appear in the reviews section the next time that Postgenomic is updated.”
…they also accept hReview microformat.

A different mark up is used for Conferences..check out both these sections: Reviews and Conferences
NOTE : you can’t filter review, by blogs in a category group

Search

See for yourself…search or browse for: Papers, Posts , Blogs, or everything.

Zeitgeist

Keep on aggregating…

Domain tracker
“Which science related domains are bloggers linking to most frequently?”

Impact tracker
“Which journals are bloggers linking to most frequently?”

I Am The Very Model of A Major Model Organism
“Which model organisms are the current batch of cited papers about?”

Wordiest blogs
“Which blogs have the largest average word count?”

Link heavy blogs
“Which blogs have the most number of links per post, on average?”

Current tags
“How are life science bloggers tagging their posts?”
NOTE: with a link to Corante Web Hubs very own Improbulus.

Watchlist

Create search feeds and Connotea feeds, these are kept as a list, also generates a spliced feed…this doesn’t have an RSS Reader built in, it is just a place to create and store your feed list.

Congregate

I’m really happy to see blogs taking off as credible sources of information for the academic community…blog conversations and aggregations enable science to express and discuss ideas and thoughts in an immediate and distributed way. Journal articles are usually researched and take a while before being published, whereas blogs provide another and different avenue to spread and share information…and since they are indexed in the blogosphere it allows people to discover these discussions by chance, and also to discover and be a part of the communities that manifest.
Authority can be managed by portals like these, a credible slice of the blogosphere.

Blogs can draw the world in together, anyone can be heard, and faster, and by lots of people.

Related post:
Carnival of discussion

[ADDED: One thing that is missing is author tags from the native blog posts, like SuprGlu and MySyndicaat…although this might get confused with the Connotea tags.

What I like about this portal is that it has a focus and it does just that, in a simple, clean and informative way. They are not about a river of the latest news they are about most talked about posts and papers, so they stick to this prime directive only.

Gnosh is a great meta search utility service…covers search engines, news, tags, photos, videos, blogs, etc…even a myGnosh.

Looking for something a little more spiced up than the house tagrolls.

Check out the post on abject learning where you can add your del.icio.us front page to your blog, the only thing is that once you click on a bookmark or tag it will launch you to del.icio.us.

NOTE: Your whole del.icio.us account (database) is not part of your blog, it is just the front page…if you want to mirror your del.icio.us account see SlashLinks: del.icio.us without the folksonomy.

I just noticed that an old favourite, BlogStreet has an RSS Publisher…similar to publi.sh and others, where all you have is a feed but no public place to see it (eg. a blog) unless you re-syndicate the feed.

The difference with RSS Publisher is that you have to host the RSS file yourself…they also have a bookmarklet to add the page you are on to your RSS file (kind of like a blog this)…fill in the title, link, and description and hit publish.

NOTE: “Blog this” is different to “Clip this”…with “Blog this” you are posting a some text and a link to what it is you are talking about (I suppose this was the traditional blogs and feeds, more about pointing to a link, than blabbing on)…with “Clip this” you are actually re-publishing what is on screen to your feed, but only if the content you are clipping is in a feed format itself (it’s like re-syndicating RSS-to-RSS).
So “Clip this” won’t work for webpages that don’t have feeds, unless it scrapes the HTML of the page into your feed…see more at RSSify your Daily Catch

…and then what about for non-webpage stuff like MS Word documents, how do you send the whole content of a document to an RSS feed.

What I’d like to do is create an OPML outline with email addresses
…then I would like to convert this OPML outline into an group email address.
That’s probably not the right terminology, but I would like to send a broadcast email from one group, that I have in OPML format.

I know MS Outlook has email groups, but I’m thinking in an open format web 2.0 way, where you could plug in your own OPML email group address into any email client.
Actually maybe this wouldn’t work as each individual email has to be input into the contacts, all the email group does is send a broadcast email, it is not it’s own email address.

Well maybe when you plug in your OPML contacts it automatically adds all your email contact addresses into the email client, and also displays them in a group view…cool, now it does two things, export/import contacts, and broadcast email groups.

NOTE: I’m aware both these things can be done, but I’d like to see the future of the web, where any service will accept your data, plug in ‘n play.

When is gmail going to introduce, category groups for contacts (maybe even tags)…not only could you browse your contacts by tag, but you could send a broadcast email via a tag…not in OPML format I bet.

Related post:
OPML for email groups