Library clips

A while back I posted on the idea of an RSS newspaper, similar to the process of an RSS Reader, only you read the contents as a print newspaper layout instead of the email cloned 2 or 3 pane layout.

So far OPML2PDF is the first step, but this is just a PDF RSS Reader, with the bonus of handy printing out the latest posts.
Then myBroadSheet came up with a working model, but this is a work in progress as Chris Saad is busy on Touchstone…if there are any flash animators interested send Chris an email.

Now there is a new comer RSS Star, to be released soon (inspired by a development competition)…can’t wait, I think this has some potential, especially as a daily newspaper print out for the enterprise (either as an e-book type thing or as an actual print out newspaper style, maybe advertisements and all).

My idea was to load in your OPML, categorise the feeds in folders, then supply some interest keywords…the folders will be the sections of your newspaper, the keywords based on text analysis along with incoming link popularity will decide which posts are the larger editorials including the front page story.

Eg. grab lots of feeds and keep them in folders like Politics, Sport, Finance, Technology, Entertainment, Classifieds (edgeio), Weather (RSS Weather), Events (RSS Calendar), Employment (RSS Jobs), etc…there are feeds for all types of content and data these days.

Then within each folder list some keywords, so if you list the term “folksonomy” in the Technology folder, any posts about folksonomies will get a bigger editorial…but if a post is about “folksonomies” but isn’t lengthy this may prevent it from elevating to a bigger editorial.
Also if there is a post about “Wiki’s” with 30 incoming links leaving the other posts for dead, well then this may be a bigger editorial piece even though it is not one of your keywords.

I’ve posted lots of times on Talk Digger.
Simply type in a URL and it will show posts linking to it from various search engines, even generate an RSS feed, use it as a bookmarklet, put it in your post template…you can select from various engines (great for generating a Bloglines citation feed)…it will even de-dupe links.

You can do link searches at Technorati and other engines, then splice them together or just use Talk Digger.

Fred has noticed that some people have been bookmarking URL searches in del.icio.us…this is going further by bookmarking a webpage and the conversational webpages linking to it. But I suppose this is not to different than bookmarking a Technorati link search, only it is a meta-search.

He is now taking it towards the community and conversational arena, from the post:

“…I intend to create a community infrastructure, in order to:

Help people to easily track conversations.
Help people to find interesting conversations.
Finally, I plan to create an infrastructure that lets people easily enter into a discussion they have found using Talk Digger.”

From his latest post:

“As you probably know, Talk Digger is: a new way to find, follow and join discussions evolving on the Web. So you have three elements: (1) finding discussions, (2) following discussions and (3) joining discussions.

With these new features, Talk Digger will become a social platform that helps people to connect with other people that follow the same stories (the premise here is that people that follow the same discussions will also have some personal and professional interests in common). It will also become a search engine of its own, and not only a meta-one.”

Check out the screenshots:
Result page of conversations
A single conversation page.

It seems we will be able to tag discussions (link searches) within Talk Digger itself, so now others can browse the tags to find various discussions under a tag…our personal endeavours aggregate to social greatness…this is such a great idea, BlogPulse could of done this ages ago with its conversation tracker.

A single conversation page displays the conversation in full, with tags, the users who have bookmarked it, the users interests, and a comments form…you can also give a title and description of your bookmark.

This is very clever, first Talk Digger was just re-displaying various search engine link searches on the one page, but now it is allowing users to organise a search result into a folksonomy.
When users bookmark a search result (a conversation) they will give it a title, description and tags…this data will eventually go from just browsing to searching as well.

In the future I can see searching for conversations with a given tag (browse tags for now), search for terms in the tile and descriptions of bookmarked conversations…this will extend it into a human indexed conversation search engine.

Conversational List folksonomy

Something I’d like to do is manually piece together a conversation, bookmark it, tag it, and point to it
…something I was trying to explain in, Distributed conversations: pinging and tagging.
Kind of like a list folksonomy, like H20, but all you do is enter in a heap of links on a page, then tag that page, this page has a permalink so you can point to it.

The difference here is the you are manually constructing the conversation, whereas in Talk Digger the conversation in based on who’s linking to a post. The good thing about Talk Digger is that the RSS feed of a conversational bookmark will keep you up-to-date with the latest additions to a conversation.
But then again a manually constructed bookmark folksonomy can still have a feed for each conversation, if you want to add a new link to one of your bookmarked discussions, just open up the page and add the link to the bottom, the feed will inform us you have added a new link to the discussion.

In both types of conversation folksonomies you could also follow feeds for tags, if you follow the tag “opml” you will be alerted when someone has added a new conversation to the folksonomy and tagged it “opml”.

Once again, this is a great idea, a conversational bookmark folksonomy…with social networking features (user interests, user account).

Enterprise bookmarking is getting some lift off of late, I wrote about it a while back.

Here are some recent findings:

More Tagging Articles
Even More Tagging Articles
More Tagging Articles - Enterprise Tagging

My collection is here:
Tags_km
Tags
Tag_apps
LIS_tags

I need to use some more user-friendly tags eg. km+tags, actually when I first started tagging I used them as categories and haven’t had time to be make them more specific, although I like that I can point to one URL containing a bucket of links…that’s why I like social bookmarking with both folders and tags.

The other problem is that I could also be more specific and use the term bookmarks, or social bookmarks or folksonomy, tagging when appropriate…also now that my tag bookmarks are growing these could even be more specific like pros, cons, etc…

[ADDED 07/05/06: It’s a tag, tag, tag, World! Next up the enterprise!]

Search Engine Watch post about the launch of the new awaiting blog search engine, Sphere…still is in closed beta (I checked in my backlog of unread emails and found I had a password so I checked it out).

From the article:
“Sphere takes a new approach to blog search, looking at three critical variables to understand both individual blog posts and the nature of the blog they appear on. As with web search, Sphere attempts to understand link structures—who’s linking to whom, and what are the quality of the links. Crucial to this analysis is an attempt to understand who’s starting or leading discussions in contrast to those bloggers who are simply commenting on existing conversations.

And finally, Sphere’s algorithm content does some heavy lifting with semantic analysis of blog postings. “It’s the hard part, and most important piece of the secret sauce,” said Sphere co-founder Steve Nieker. Sphere doesn’t use tags in its ranking of blog posts, even if they exist.”

Very clean interface and lots of targeted searching:
- boolean
- brackets
- phrase
- date
- blog site search
- search in blog post titles

Example…blog:”library clips” title:ziki
Or I could of used…site:libraryclips.blogsome.com title:ziki

Choose between relevance or time for results

On the results sidebar is a list of recommended blogs

Each result has a search feed

Each hit in the results has a brief profile of the blog and a link to the larger profile page
- avg. posts/week
- avg. words/post
- avg. links/post
- last Three links In
- last Three links Out

Bookmarklet to see related posts for the page you are on

Submit a blog

Coming soon
- related posts for each search result

I just noticed Rojo has a search box:

- find & add feeds
- search for stories

We know Rojo has a feed folksonomy and bookmark folksonomy (as well as voting), but now it has included search into the mix.

You can search for feeds and preview them (as usual), now you can also search for stories…so it’s becoming quite the search engine, and what a clean engine as the content is just from Rojo subscriptions.

Still seems in its early stages, they haven’t blogged about it as of yet, but I’d like to limit these 2 searches to my own account, and also within a tag of feeds, or tag of stories:
- search for a feed
- search for a story

Advanced Search Requests
- Stories I’ve Read
- Unread stories
- Unread and Read stories
- Tagged or Saved stories
- Flagged Stories
- Mojo’d stories
- Shared Stories
- Boolean (AND, OR, NOT)
- Phrase searching
- Date ranges
- Search your Contacts stories
- Search your Rojo and your Contacts
- Link searches (eg. Bloglines citations)

They are playing with my head, now the search stories has disappeared…keep a look out it will be coming soon, I must of caught them experimenting with it.

I was just checking out GROU.PS which enables you to create a read/write web2.0 portal…basically it brings together maps, blogs, wiki, bookmarks, etc…in the one service.
As the name implies it is for groups, so the idea is to create a communal portal…we have seen these all-in-one personal information management tools in the past for individuals, well here’s one for groups.

We are seeing groups in social bookmarking services, such as ma.gnolia, and a wiki is kind of a group by default, and communal blogs are also around…well the idea is to combine all these types of modules into one service.

GROU.PS is similar to these services, it’s like Ziki or a startpage, but for groups, and it is more dashboard oriented, with lots more access to various different types of web2.0 services.

Like Ziki, each group is described with tags, this adds to social discovery.

In a nutshell

Join a group and add your blog, you will then see your blog posts in the Blog section…add your del.icio.us links or choose from another service to see your bookmarks at the Link section (even specify to see bookmarks at the tag level), etc…basically a group river of news for blogs, bookmarks, photo’s, etc…with the bonus of wiki, IM, forums, etc…

If you haven’t got a blog, or use a bookmarks service, you can use the inbuilt blog and bookmark service…the wiki is also inbuilt, you can also post to your favourite forum or use the inbuilt forum….check out the rest of the features.

See more on GROU.PS…here are the modules.

Here’s a group I made called OPML group…also see the start page view
…join a group or start your own.

Keep up with their blog.

PIM

I’d like to have a PIM tool that gives me a userspace, where I can create tabs, each tab launches my favourite service, such as Wiki, RSS Reader, Photo’s, Bookmarks, IM, blog, forum, web monitoring, email, calendar, tasks, etc…but these services launch within the PIM tool somehow, and each service can talk to each other…this would be easy enough to do if all the services were inhouse, but since they come from different providers it’s a bit more complicated due to proprietary issues and data formats.

Basically a build it yourself PIM tool by connecting your favourite services.

What about a personal admin user space where you can access and use (publish)
…actually Startpages also enable you to view your bookmark service, gmail, weather, to-do, etc…this is the kind of idea I’m aimed at, accessing your modules from the one service.
Furthering this concept would enable you to write a blog post, write an email, IM chat, etc…from within the one service.

Startpages and people pages enable you to re-syndicate this type of information on the one page, but this is more of a display…GROU.PS also adds the collaborative aspect.
Although they can be interactive, which is what I’m after…some Startpages let you add a post to your bookmark service, and GROU.PS enables you to post to your favourite forum.

Soon I think we will see these Startpages and people pages incorporate more interactive features and group features, than more than just a re-syndication display page.