Library clips

Adam Green is taking OPML to the limits, check out his lastest OPML Mashup.

What he has done is extended his tech.memeorandum OPML hack to include some extra information about the feeds by cross breeding it with Technorati.

This is the ultimate informed Grazing List…I commend him here, as this is what a Grazing List is about, it’s about shopping for feeds, and what better way than knowing at a glance; the tags, linkbacks, and ranking of a feed, just before you dig into sampling some posts (plus the author name, and a link to the homepage…this is awesome).

This is really getting value from your Reading List, or Grazing List in this case.

Hey Adam, could you still load this into an RSS Reader as a Reading List, considering there are non-feed items in the outline?

I’ve also noticed that fellow librarians are catching on, I don’t doubt a second that OPML won’t be big in libraries, at least in the backend, it’s all about outlining, categorising…just what the librarian ordered.

What I do feel is that Reading Lists will be a major player in the education sector…recommended reading that’s alive, and can be communally edited.
Up until now a teacher created a bibliographic type list of resources (books, journals, etc…) with HTML links…now they can create this list in organised OPML, and if feeds are any of the items, well it is like reading the latest from the book right there in the list.
No longer do you have to login to the school website to access your resource guide, now that it’s packaged in OPML, you can read it in your RSS reader (if it is a Reading List where all items are feeds), or in an OPML Reader (reads any type of item)…or you can add it as an inclusion to your own OPML Directory, etc…OPML frees data, anyone can take an OPML package and run with it.

This has always been possible with Blogdigger Groups, one of the first River of News Reading Lists. A student could keep up to date with the contents of the Reading List by subscribing to the spliced feed, or by visiting the schools version of a Blogdigger group.
Now they can subscribe to the OPML URL of the Reading List, so they don’t have to read a river of news, they can read by feed. Just like the spliced feed, as the feeds change in the OPML the contents will change in your subscription to it in your RSS Reader…so both spliced feeds and Reading Lists are dynamic.
Reading Lists gives you the extra option of reading by feed, and since it is OPML can be read in other services other than an RSS Reader.

[ADDED: this is being called an Annotated Grazing List or Reading List]

ZoomClouds is basically the same as TagCloud, machine tags for your blog based on the latest posts in your blogs feed (in this respect the terms in the tag cloud may be based on your last 10-15 posts).

Anyway it is supposed to work better than TagCloud, and there is a lot of customisation…try the demo.

Here’s a sample…you can even share and discover designs.

Find out more.

Check out this post from nearly a year ago (links don’t work anymore), it describes auto-text analysis tags for your blog post, in which you can also add your own tags, then add a weight to each tag.
In this way you have a tag bar cloud for each post, but the font size of the tag is not based on frequency, it is based on how much you think it describes the “aboutness” of the post, ie. if your post is about “folksonomies”, but you also speak a only a bit about “search”…then the font size of “folksonomies” will be big in relation to “search”, which will be small.

From a glance you will be able to tell not only what a post is about, but the level of aboutness…this is similar to Major, and Minor Descriptors in journal databases…something I wanted from del.icio.us.

Then to take this further, and say show me posts from the feeds in my RSS Reader that are tagged “search” with a weight of “7″…I’ve talked about viewing posts in your RSS Reader aggregated by author tags, this instead is by machine tags (or a combination of both), but with a weight/level of aboutness added to the semantic mix.

So how do you get auto tags for your post…try Tagyu…then you can add your own tags, and/or delete ones you don’t agree with…then you need to add the level of aboutness from a scale of 1 to 10…1=a small part of the post is about this topic…10=most of the post is about this topic.

I’m not sure if there is a tool out there where you can add a weight/level to a tag/s…and for them to appear underneath your posts, linking to local blog pages, or to Technorati, or del.icio.us, etc…

Then you would need to aggregate these on the sidebar, but this is where we encounter a problem
…now the font size would depict how often (frequency) you have used these tags…this way, from a glance, you get an idea of how much a blog devotes to a certain topic.
So now we have lost the level/weightness we assigned to each tag at the post level as now the font size is represented by frequency, and not degree of aboutness.

A similar tool is TagCloud….here are auto-generated tags for your whole blog…the font size denotes frequency…the other thing is that this is not a static set, it is always changing based on your context.

What do we need?

1. Auto-generate tags for a blog post

…we can do this with Tagyu.

2. Add a weight level to these tags

3. Include 1. and 2. as a tagbar cloud underneath each post

4. When you click on these, you have a choice to see the local blog tag page, or launch to Technorati, or del.icio.us, etc…

5. Aggregate auto-tags for the whole blog on the sidebar

6. On the blog sidebar, how will we tell the difference between aboutness, and the difference between frequency?

We could illustrate the level of aboutness for each post by labelling a tag in UPPERCASE for “Major Descriptors” and in lowercase for “Minor Descriptors”…it’s only an option of one or the other, whereas before we were suggesting a level from 1 to 10. This isn’t as visually conclusive, but at least you know by looking at the case of the tag, if the post is a little or a lot about that tag/topic.

So now, underneath each blog post, the tags will not be in a cloud but just listed in upper or lower case, then on the blog sidebar we can have either 2 tag clouds or put them into one tag cloud. If we mix them into one tag cloud, you may see a tag twice once in uppercase, and once in lowercase…and the font size of either of these defines how frequently it has been used in the blog.

At a glance we will not only know what topics the blog covers and how often, but also view posts by level of aboutness.

Useful

In hindsight you may be looking for a post that was a little about the topic of “search”, but the post was more about “folksonomies”. So instead of searching through the hundreds of your “folksonomies” posts, or the hundreds of your “search” posts, you can more quickly find the post by the “search” minor descriptor, as opposed to the “SEARCH” major descriptor.