Library clips

Jack Vinson has amplified my post on the searching part of your social filter, the other part I discussed was networking.

This post is a round up of a few ways to search your social filter:

DIY HTML search engine

Create a search engine (searchroll) at Rollyo, this will search the HTML of all the blogs, bookmark users, Twitter users, etc…
You could make 3 Rollyo search engines, one for all your favourite blogs, one for bookmark tags or users, and for for Twitter users. Each of your search engines can be accessible via a fancy bookmarklet or widget.

Google Custom Search like Rollyo lets you enter a number of sites to search and gives you a public page and a widget, make as many as you want, even load sites by OPML (it searches HTML)…see here, and more.

You can also make a custom search engine on-the-fly, by putting some code into your blog, it will pick up all the links on your blog pages and provide a search box to search those links…hmmm something Snap Shots could get into.

[ADDED 12/10/07: Forgot to mention you can give others edit rights]

RSS Reader search

Google Reader allows you to search all your subscriptions, by feed or tag…so there you go, straight off you have a social filter by default…not bad at all…Steve Rubel has really taken advantage of this for data mining purposes.

Communal Search Engine

Swicki is a communal search engine, see more…for our purposes you can enter your favourite websites and search your social filter.

RSS grazing

Put in all your favourite blogs, bookmark users or friends lifestreams and with some Grazr scripting you can search this social filter…see here and more.

Lifestream network

In all the examples above instead of entering the feeds of all your favourite bloggers and bookmarkers, you could add their lifestream feeds, if they have one.

One place you can make a lifestream is at Ziki by adding the feeds from your web 2.0 profiles, then you can search your lifestream, even limit to a feed. Since Ziki is also a social network you can add friends, then you can search your friends lifestreams…nice social filtering. In fact you can search the whole of Ziki, making it a general people search engine. Add a Ziki search engine to your browser.
Ziki also allows you to enter your Reading List, ie. blogs you read, I guess your lifestream is a personal filter of all the blogs you read and other content…but you can’t search your Reading List…yes you can.

So you can search your content (lifestream) and your Ziki friends content (your network), but this is your Ziki social filter, what about searching your Reading List as well…see below for the finale.

DIY Social filter engine

Lijit is similar to Ziki, you add your lifestream (which you can search in, even limit to a feed), I don’t think you can add friends (it seems you can add other users to your network, only I can’t see an add friend button), instead what you can do is add your Blogroll (Reading List), your del.icio.us network, your MyBlogLog community or add any feeds you like creating a true social filter search engine (your social filter)…they have a handy widget.

It pays off blogging about this stuff.

As Jack Vinson alludes to why search the web everytime, when you can search your social filter…you have gathered sources you trust, and sources your sources trust, and so on.
Searching your network is a great quality filter for the web…perhaps the attention web.

[ADDED 12/10/07: Lijit could extend it’s offering to include some lifestream services, this way you could add a friends lifestream…saves you adding their bookmarks, blog individually.

The difference between Ziki and Lijit is Ziki is a social network experience, you can only search you and your Ziki friends. Lijit allows you to add other Lijit users to your network, but it also allows you to add feeds from people that are not registered with Lijit…this makes it a more complete social filter (not limited to a system).

As I mentioned Ziki allows you to add your Reading List, but this is not included in the search, so really they are one step away from including external feeds into your search filter, just like Lijit.

Something I forgot to mention is that you can visit someone’s Ziki or Lijit page and search the world through their eyes.

So not only do you have your social filter, but you can search the social filter of others…by adding them to your network or by visiting their profile.

See a post from Stan of Lijit that has a good visual explanation.]

[ADDED 23/10/07: Blog network as your social filter]

[ADDED 23/10/07: BlogRovr as social filter]

[ADDED 11/11/07: Blogbar : search your outlink sources]

Megite discover which I posted back in May, has added a bookmarking feature.

Megite’s main service is the memetracker, this works on it’s own to bring you the hottest stuff of the day, you can even enter your OPML so it’s the hottest stuff according to your feed set (it also throws in some recommended posts).

They also have a discover engine as a way to find the hottest posts by entering a topic query, or enter a URL and get related links and also topics for that URL.

It has now expanded into bookmarking, when you play with megite discover you can choose to save a link right there within your megite user space. You can also use a bookmarklet to bookmark links you find outside of megite…what we have is a memetracker and discovery engine allowing users to now bookmark links it finds in the engine and outside of the engine.

Hmmm, what about being able to bookmark stuff you find on the main megite memetracker site.

If you find a handy bookmark on the web will you bookmark it in megite or del.icio.us?

If you find a handy bookmark in megite discover will you bookmark it within megite discover or del.icio.us?

Related:
megite : Memetracker for a single blog

Finally you can now search your Google Reader.

Search All Items, Shared Items, Starred Items, a folder, a feed…what about searching Google Blog Search in general.

If you subscribe to the feed of someone’s Shared Items, you can now search this content, you could also subscribe to your own blog, and search in it.

What you can’t do is select multiple feeds or folders, and then search across them.

Results are summaries, you can’t expand an item in the search results, but you can star it.

This kind of helps out that memetracker issue from within your OPML, if I read a new post about a new product that has been released, I can now search All Items, to see which other feeds of mine are talking about it, hmmm…you can’t search just New Items.

So now I can find all posts in my RSS Reader about this new product in an instant, for example I just searched my subscriptions about Google Reader search and no-one has posted it yet, and I subscribe to some hot off the press web 2.0 blogs…but if they did post about it, I would know in one easy click.

Once I have collected posts in my RSS Reader about this new product, which I can now do very quickly, then I would want to know if any feeds I don’t subscribe to have quality posts on this new product, for this I have to leave Google Reader and open a new window and search Google Blog Search.
As I mentioned before, it would help if this was a search choice in the drop down menu…a clustering feature is one better as it’s all done for you.

One more thing, what about smart feeds like Newsgator, Bloglines and others, why can’t a search generate an RSS feed.

Just say, I’m interested in wikis, I would subscribe to blogs that often post about wikis, I might even have them in a folder called wikis.
To find new posts that mention the term wiki, without having to wade through every new post in my RSS Reader, I can just search the term wiki, if this generated a feed, then I could subscribe to it, this saves me doing the search everyday.

NOTE: I noticed Bloglines has had a new release, including a startpage view.

I just found the above link about Bloglines by searching in Google Reader and just found mashable posted about Google Reader search 8 minutes ago.
But I couldn’t expand to read this post from my search results, and I can’t mark it as read.

The Google Reader blog also mentions you can now use the browser buttons to go back and forth, and you can click the subscription pane in and out of view.

I’ve complained a few times about not being able to search in my RSS Reader, I use Google Reader…I know, ironic that it’s the search king.

Hack one yourself using Google Custom Search Engine, which I have posted about before, and add it to your Google Toolbar which I also have posted on.

How ?

Sign up at Google CSE

Create a search engine, just add one site for now

When you are complete go to the Advanced section, here you can add your Google Reader OPML file

Each search engine you create has its own homepage (you can open it up for collaboration, and you can add it as a widget to your startpage or your blog)

From here you can either bookmark this page (here’s mine) to your links bar or favourites or you can add it to your Google Toolbar…do this by a right-click in the search box (don’t forget to rename it).

Since the Google CSE doesn’t accept an OPML URL (it only accepts a file), I will have to import my OPML file every month or so to manually keep it up-to-date…hopefully Google CSE will overwrite each time this is done, as my updated OPML may not only have new sites, but also sites I’ve deleted.

Check the left sidebar of my blog under the heading “Reading List”, you will find:

- Grazr mini RSS-Reader

- Raw OPML of my Reading List

- Search my RSS Reader

- OPML2PDF of the latest posts from the my Top 10 feeds in my RSS Reader…max. 50 posts per feed
(this uses your first 10 feeds in your RSS Reader, but I put these 10 feeds in my Technorati Favourites and used the OPML of my Top-10 tag)

- RSS2PDF of the latest posts from my Top 10 feeds in my RSS Reader
(this is a spliced feed mixed posts river of news, rather than reading posts by feed…this time I used the RSS from my Top-10 tag)

- SYO network where I share my RSS Reader feeds, and who in the network subscribes to my blog feed

- OPML Memetracker of my RSS Reader feeds

Megite is getting into the related/recommendation game with their new discover engine.

As we all know megite is a memetracker, it tracks the hottest news and surrounds it with discussions and related links.

This is different to Technorati link search, which will only list sites that link to the site in question, plus it is up to the user to type in a URL, whereas megite will feed you the hottest URL’s of the day.

Then there is Sphere which allows you to enter a URL and it will display related items, but this again is up to the user to type in the URL.

This is why a memetracker is so great as it will tell you the hottest URL’s and you can even limit the news sources to your OPML subscription list, or even to one feed.

Now megite is offering the best of both worlds as it now allows a user to enter a URL or even a keyword and discover related links.

Here is the popular topic “community“.

Here are related links to my blog, the great thing is that I can limit this list of links to a topic.
So it not only discovers related links, but also topics.

When I entered a blog post URL, it didn’t discover topics (like an auto tag type of thing), only related links…a natural idea is to compare a hot URL on megite and compare it to the same URL in their Discover engine.

eg. Microsoft makes its biggest acquisition ever
megite - http://blogs.zdnet.com/microsoft/?p=450
discover - http://blogs.zdnet.com/microsoft/?p=450

technorati - http://blogs.zdnet.com/microsoft/?p=450

blogpulse - http://blogs.zdnet.com/microsoft/?p=450

sphere - http://blogs.zdnet.com/microsoft/?p=450