Library clips

Greg over at Blogdigger left a comment on one of my posts which answered a few questions…I thought I’d repeat it here in case anyone missed it in the comments as I think it’s of real value.

It was in response to my desire for Bloglines to not only search my subscriptions, but search portions or selections of my subscriptions..down to searching just one feed.

"You can create a Blogdigger Group of you subscriptions (import via OPML from Bloglines) and then search either the whole group, or filter on individual blogs to search across.

Our main search page also uses the blogID: and site: fields to search just a single feed or site."

This also helps with RSS overload…I mentioned previously that sometimes a feed only gives you relevant posts once in a while but you don’t want to get rid of it, well via Blogdigger groups you can firstly load in your subscriptions via OPML, then filter your subscriptions to just one blog (turn off all the others), do a search and subscribe to the search query feed in your  news reader, and delete the original or native feed.

Greg also mentioned:

"You can also put a search box on your own site to provide site search, details here: http://www.blogdigger.com/blog/2005/01/18/1106028202000.html"

Blogdigger is such a versatile tool…it always seems to be the answer to what I want to do…they are so responsive to questions….good onya

I wonder if you can get a search box for a Blogdigger group..then you can search your subscriptions from your blog…Bloglines should be able to provide one of these..this search box also does the job. The Blogdigger  groups search box could have a drop down menu so you can filter which feeds you’d like to search…it would be like your own mini blog engine (although citations searching wouldn’t be exhaustive).

ADDITIONAL: you would have to keep your OPML up-to-date, ie. everytime you add or delete a feed in your subscriptions you’d have to refresh the Blogdigger group by loading in a new version of your OPML…in this case searching your subscriptions via Bloglines would automatically be updated.

PurpleSlurple is a great tool for blog posts…you can point to a part in a previous post you made to the exact line (granular level)…especially good if you are pointing to the permalink of a long post - this directs people straight to it.

Although it’s not working for me on this blog..the error reads: PurpleSlurple only supports http-based urls

Commoncraft posted something similar not long ago.

continue…

Tags seem to scale better when the domain of the social bookmarking network is within a more controlled environment…ie. you can moreso trust the contributors tagging style to conform with the domains style..and overtime eventually create a communal controlled monitored vocabularly (I alluded to this here.) CiteULike groups seem to have developed a space where people can join groups to add and share tags within the network.

I’d like to see an analysis of this over time to see how better (I’m assuming there would be more consistency) the vocabularly scales when the community is limited to people with similar interests over an open version of share all tag all community.

Would these groups have meetings on the quality of current tags used and clean up the tag index (list of tags) every month or so..?

Over time would these tags still be listed alphabetical or would they be organised differently?… (taxonomy maybe…even if this were so, the important element here is that the tags are still user defined.

A CiteULike post on their blog points to a searchable journal collection of not only journals that have rss feeds, but also shows the latest TOC in the search result…similar to the presentation that University of Saskatchewan have within their library website.

The quality (filtering) of information needs to be managed which in turn will help quantity and time…but how to do this?

Will we always have information anxiety?…in relation to current awareness (too much coming too quickly to process, organise and comment on).

A lot of talk about RSS overload has begun to re-emerge of  late (and later..again). I remember the excitement of finding how an RSS reader relieved my inbox of clutter and the whole subscribing/unsubscribing process. With the incredible popularity of RSS feeds (not just blogs anymore), now my news reader is full of clutter (though in a more organised way) and my email inbox is experiencing loneliness…now I need a solution for my rss reader, everyday I’m subscribing to new feeds (not just blogs and social bookmarks) and making search feeds…besides the overload I’m also experiencing a lots of duplication. RSS readers organise information coming in much better than email, and the subscription and reading process is much more convenient and efficient, but the problem still lies in time to read everything. I’ve made a lot of watchlist’s via Technorati of late and have found this does a great job, but it could never replace reading actual blogs - not that it trys to, ( I don’t want to miss anything) although when I research a topic a watchlist helps me identify new blogs of interest..they are a good starter…it also depends on your information needs.

Instead of cutting back or refining the load some solutions of late are converting your readers content into audio (so you can read on the go)…here is a good summary of the tools out there…this is the cutest.

This is a bit technical for me (although it would be cool listening to mp3 versions of your feeds in the car), I’m more interested, at the moment, in sampling a few other readers on the market that alleviate or streamline the reading process,  and recommend relevant information…at the moment I’m still totally hooked on Bloglines (good features like, clip blog, converting email to RSS, search query RSS feed, etc..)

…my thoughts on Bloglines

Readers to the rescue

Feedmarker

Handy tool that combines an RSS reader with a social bookmark manager. This difference with this tool is that you can manage your list of feeds by tag instead of a folder. The advantage of this is that a feed can be essentially tagged or put into multiple folders…and each of these tags have a feed of their own (it also has the "marked read" functionality and mark item to go into your bookmarks manager (this of course has a feed for every tag) so if you mark a post as read in your reader but want really want to read it later you can just put it in the bookmark manager and give it a tag like "to read".

So when you open up your RSS reader you can choose a tag, and read the feeds categorised within that tag…a good extension of this would be to join tags, so you can say I want to read all feeds with the tags: blog, rss, library, search, wiki, etc..Another great thing about this is that the public can subscribe to not just your OPML (your RSS reader account - all your feeds) but just portions of your account. I guess putting a tag on a bunch of feeds which generates it’s own feed is a bit of what Blogdigger does in making compilation feeds…but you can’t read a stream of posts you still have to open each feed to read the posts.

Awasu

Channel filters, channels hooks, search agents (similar to Bloglines search query feeds of your own subscriptions)

Findory

Recommends posts according to your reading history. (the content is delivered by the system not according to a list of your feeds)

Chameleon (plug-in for Bloglines)

Exerpt from the home page:

"Keeps track of which feeds you read, how often, and when
Figures out which feeds are your favorites, using a few novel algorithms, and highlights them
Identifies the top links in your feeds — like Blogdex, but for your feeds only
Shows you your usage scores based on the algorithms"

Urchin (web-based)

Exerpt from the home page:

"Urchin is a Web based, customisable, RSS aggregator and filter. It’s primary purpose is to allow the generation of new RSS feeds by running queries against the collection of items in the Urchin database."

Ego clip

Exerpt from the home page:

"Takes all the items from all your subscriptions.
Prioritizes them by how important it thinks each item is to you.
Saving you from wasting time digging through the many items that you don’t care about.
You don’t have to grade items manually. egoClip just notices when you show interest in an item and learns from that."

You: subscribe

Incorporates into Outlook…

Ultra gleeper (plug-in)

A recommendation engine based on your subscription list and more…

Exerpt from the home page:

"The Ultra Gleeper takes your weblog subscription list and starts from there. It crawls the web for things you haven’t seen and shows you the pages it thinks you’ll like. Your feedback improves its ability to give accurate ratings. With the Ultra Gleeper can find new pages and new weblogs to read. And if you have your own weblog or use del.icio.us, the links you post there will be automatically turned into ratings.

The Ultra Gleeper solves or avoids the problems that give recommendation engines a bad reputation. It won’t give you a lot of links you’ve already seen, because it knows about your subscriptions and what they’ve posted. It won’t just recap the most popular links of the day, because its indie rock algorithm distrusts excessive popularity. It won’t ask you for a lot of calibration ratings up front: you already gave those ratings by telling it what you subscribe to and pointing it to your weblog and/or bookmark page."

Lektora

Exerpt from the home page:

"Lektora understands that your time is valuable. A news item that was not read yesterday is not likely to interest you today. Each time you refresh your feeds, Lektora presents you with only the latest news items, conveniently saving older news items in a daily archive that can be consulted at a later time."

Conclusion

In the end I feel machines can only do so much in organising, filtering, de-duping, and recommending information…you can’t know everything (unless you don’t sleep). We come to a stage when we have to be selective with our information…what’s the ROI, ie. how much of what I track will I blog or clip….if certain feeds are not contributing to ideas or content to your blog or clipping account, then we must let go of them. The next question is what about the feeds that only offer a good post once every couple of weeks (well hopefully another feed you read will also syndicate this information or you can try a search query feed within this web-site and subscribe to that instead).

Apart from more intelligent RSS readers…we need to select others who are already scouting areas of interest, we need to form an informal syndicate or (trust?) network, where people cover different areas - there are just too many blogs and search query feeds to cover everything for one person. eg. why would I subscribe to numerous feeds about Skype, when I can just read Unbound Spiral (of course it depends on the intent of your information needs, if your researching for a dissertation, the more the better, if it’s just current awareness then reading a blog that does all the ground work and presents the findings is enough).

In relation to filtering information for quality, it would be ideal if all blogs had category feeds, and further more, as described here, if every category RSS feed had a check box, in this way you can subscribe to your own mix of categories…also throw in a search query feed (sometimes called a smart list)to the mix or even more, refining a search query feed within a category. A way around this is if you duplicate your blog posts in del.icio.us or Furl(in the case of Furl, you would need a tool like Blogdigger to do the mixing), then people can subscribe to a mix of categories, and so forth (but this is a 3rd party solution).

NOTE: Does anybody know of an RSS reader that can mix feeds, a la Blogdigger?…and even generate search query feeds (within a group or just one feed), and also shut off certain feeds in your list of feeds?

I think if blogs (or more probable RSS readers) eventually are of this functionality it will help with the current problem of filtering for quality, inturn helping quantity and time…I guess it is an innate quality to perpetually push the envelope of "information managment" - at least for librarians and the like.

Here is a comment (by ade) on this post that adds some realism to this issue.

Bloglines has augmented the way I work, I visit it multiple times a day…it has become an extension…

Since I’ve been using it for a long while and lived to see all the new features, I would like to comment some feedback on its functionality.

Channels

Even though I’m managing my feeds in folders the list is getting rather long (I don’t think sub-folders is necessarily the solution)..instead of opening multiple accounts (one for personal, one for work-related stuff, or by subject) I would like to see channels within the same account. I find this works really well with the web monitoring agent, WatchThatPage when I clip an item a prompt box asks which channel to file this in, then later you can organise it within a folder. Then when I’m reading my account I can specify a channel, avoiding scrolling past irrelevant folders. I wonder if this would mean you would have multiple OPML’s, one for each channel, and a total one.

Search

Apart from searching my subscriptions could it search selected feeds from my subscriptions, that is could all my feeds have a check box where I could select the feeds to search…good to search a single feed from a blog that doesn’t have a search box of it’s own (I know I can use Google site search for this but if I do it via Bloglines I can then subscribe to a search query RSS feed for a single blog - this feature would be great for Blogs where I only like some of the content…it could reduce the number of irrelevant posts.

Does anybody know if you can site search on Technorati, Feedster, Blogdigger or Pubsub and subscribe to the feed of the results?

Citation feed

Read my post in the Bloglines forum…apparently this is coming soon.

Admin

I find that I can’t rename a folder, and that I have to make a new folder then transfer all the feeds across…maybe I’m missing something..

Duplicates

Is there a way to filter the same posts that re-occur in many of my feeds…it seems hard as some posts are virtually the same but just re-edited a tiny bit…I suppose I mean can my reader know that a permalink has already been shown in my reader (not a link within a post but the URL of the actual post)…if I want to see it again I will search for it in my subscriptions, but I don’t want to see it in new content…this happens a lot especially when you subscribe to Furl or del.icio.us feeds.

It’s much easier to understand the empirical impact of folksonomies when looking at it from a visual point of view. I find this blog post by Vanderwal.net spot on in my grasp of this topic…I like this proviso relating to the prevention of skewed results:

"It is important to note that folksonomies work best when the tags used to describe objects are in the common vocabularly and not what the person perceives others will call it…"

Based on he’s sample we can see graphically the most used common tags for an item…"that could be used (to)extract a contolled vocabularly…" …I was alluding to this point in an earlier post.

Been doing a bit of research lately on some of the larger full-text databases that include LIS info.

On three of the one’s I regularly use I found it hard to search just in the LIS subject area…but did eventually have some success (I don’t have a subscription to any of these so I’m not sure of any added features).

Ingenta allowed searching in a particular journal or across its entire collection…it allows browsing within "Library Science", but doesn’t search within just this area.

Emerald doesn’t allow you to search within a subject area but in  its "Advanced search" it does allow you to check journals that you want to search, which I think is a great feature.

Elsevier has so many entry points to its collection that I get confused (Elsevier, Scirus, ScienceDirect). I found the best way was to go via the Elsevier homepage, choose the "Social Sciences" subject area and choose "Library & Information Science" (It’s great to see that LIS is within Social Sciences, and not Computers, Business, IT, or Reference). On the right is a Scirus search box where you can search in a single journal or within the LIS subject area. If you want to do another search you have to do it from the original site and not Scirus (as Scirus won’t limit to LIS subject area).

Now that I’ve graduated from my LIS studies I lack the access to full-text of these databases, but I’m ever thankful they allow free peeks at their abstracts.

Just to recap and add some regarding a previous post…I just noticed, (thanks to a great document on Furl - via weblogg-ed), that Furl  has a feature called Furl mates, a recommendations feature of links based on sites I have furled. There is a plug-in for del.icio.us for a similar tool called gre.gario.us, although this works on a different recommendation, it finds people who have bookmarked the same links as you. (I guess I wouldn’t call this a recommendation for web-sites rather a recommendation for people’s accounts, as you visit them to browse for what you like - interesting, tagging isn’t required for this feature to work).

Also I noticed that prior to what I’ve mentioned before I can’t seem to search by tag in Furl. Well I can browse the tags in the "popular headlines" page, but I can’t browse other tags not shown in this page. (Last time I mentioned changing the tag name after the "=" sign in the address bar, but this doesn’t seem to work as the URL is for popular sites not all sites (unlike del.icio.us which allows you to alter the term in the address bar and try your luck).

This gets me back to searching by tag without having to use a 3rd party interface such as Technorati, which I do use if I want an overall picture of the blogosphere and tagosphere (whatever you call it) and it’s conversation tracking features.

I’d like to search for a tag (results shown in an A-Z list, which is also good for browsing - I assume this may be too cumbersome, and a "no results shown" may have to do) or you may be recommended another tag. Then once I find my tag I click on it to search in a tag, and it’s business as usual.

I think at the moment del.icio.us searches in both the description and extended field in the one and only search. If del.icio.us ever does fielded searching, eg. by title (description field), by full-text or abstract (extended field), by tag, in all accounts, just your account…then it would be great having some code for a search box to put in your blog, just like Technorati’s searchlet.

The other day I posted about the lack of repositories being grouped together by subject…well I talked a little too soon. Reading my CAUL update today revealed a little suprise, well a big surprise…

…DOAR (Directory of Open Access Repositories) developed by University of Nottingham, UK and University of Lund, Sweden. Here is an excerpt:

"A new service is starting development to support the rapidly emerging movement towards Open Access to research information. The new service, called DOAR - the Directory of Open Access Repositories - will categorise and list the wide variety of Open Access research archives that have grown up around the world. Such repositories have mushroomed over the last 2 years in response to calls by scholars and researchers worldwide to provide open access to research information.

DOAR will provide a comprehensive and authoritative list of institutional and subject-based repositories, as well as archives set up by funding agencies - like the National Institutes for Health in the USA or the Wellcome Trust in the UK and Europe. Users of the service will be able to analyse repositories by location, type, the material they hold and other measures. This will be of use both to users wishing to find original research papers and for third-party "service providers", like search engines or alert services, which need easy to use tools for developing tailored search services to suit specific user communities."