OUseful.Info, the blog…

As Downes starts to commission his daily/serialised feed platform, here are a few more fragmentary thoughts about feeds and feed items…

If we are to construct uncourses out of separate “blog posts” (that is, out of small chunks of content that can be sensibly represented as independent RSS feed items) then it would be handy to get access to that content in bite size RSS chunks.

At the moment, it’s quite hard to get RSS representations of blog content anywhere except from the original feed. A search on a blog search engine does not turn up all the content of the search result posts in the search engine, just the opening fragment. In a feed reader such as Bloglines or Google Reader, all the blog posts are indexed, can often be saved as favourite and are typically displayed as full text/full post results following a search.

But can I construct a new feed containing some of those separate, search result and favourited feed items in an order I require? I don’t think so…

So here’s what I have in mind – something like the Grazr editor that lets me construct new feeds by dragging and dropping the content of separate feed items from my feed reader and into a “new feed” editor. The editor would let me construct and publish new feeds using items collected from arbitrary feeds, and then maybe drag and drop those items around inside the editor to change the order they are listed in.

The easiest way of doing this would probably be to extend a current feed reader, since feed readers tend to sit on top of platforms that already have a database store of separate feed items.

It would be more satisfying, however, for blogging engines to publish single item RSS versions of each blog post. That is, as well publishing each blog post as a separate HTML page, and as well as adding the content of each post to the blog feed and any corresponding category feeds, the blog engine should also publish a separate RSS feed containing just a single item, made up from the title and body (description) of the corresponding post.

That is, for a blog post published at http://example.com/myblogpost.html, I’d also like to see http://example.com/myblogpost.rss (or whatever…;-) containing a single item RSS feed for that page.

This shouldn’t be too hard to do – as well as the page template for each post, we just need a minimal template to publish the post as a single item RSS formatted document.

WordPress blogs already offer page level feeds, of course, but as far as I know they just syndicate the comments posted to each page?

(As a workaround for OUseful.info, my WordPress single feed item (raw) pipe does some screenscraping to generate a single item RSS feed version of a OUseful.info blog post from it’s HTML page version, and the WordPress single feed item pipe takes the scraped blog post content and adds it to the top of the comment feed that WordPress already publishes for each post…)

Why would I want to do this? Well suppose I want to publish my own crazy learning journey through OpenLearn content – the easiest way would be just just grab the RSS version of each page I wanted from whatever course it happened to be on, and construct my own series of posts/pages as a new feed. (If you’re interested in this, here are some tricks and tools that will let you do it (if they still work and haven’t rotted yet?!):

• Embedding Single OpenLearn Unit Pages in an Arbitrary Blog Post;
• Single Item RSS Feeds from OpenLearn Pages

PS Note to self – check to see if Microsoft pursued web slices in IE?

Make of this what you will… the class is in session, so here’s your weekend reading:

Some ramblings about how blog posts (at least, the ones that get “reused” in an online, better-link-to-that sense) sit in some sort of link context or link neighbourhood, where the links are directed, either going from the post (links out to other resources) and links back in to the post (from tweets, trackbacks, bookmarks etc): Trackbacks, Tweetbacks and the Conversation Graph, Part I.

Some amplification by Patrick Murray-John, who’s not scared by databases and tending-to-big data in the way I am: Linkage Graphs in UMW Blogs. The demo shows the results of link-mining/trackback graphing between some of the UMW blogs. The ability of UMW blogs to republish content through syndication in different parts of the URI-space that UMW blogs covers leads to some interesting observations. Like this one:

[I]n chasing through what links to conversations, it throws in the idea that the same conversation could split into different directions from exactly the same content, but presented in different contexts. This would also show up in tracking retweets.

(UMW blogs, if you don’t know it, is, well, what can I say: tending to awesome? Pushing at the frontiers of blogs in higher-ed? Here are some of the ways users are encouraged to use UMW blogs – though of course they can make up their own ways too: Ten ways to use UMW Blogs. For some techie background, check out Syndicatin’ Welfare: UMW Blogs’ Syndication Framework on the Cheap.)

My own dabblings with the trackback graph in the Digital Worlds uncourse blog suggests that that multiple possible forward paths through blogged uncourse content are constructed by the process of linking back to earlier content from later produced content (got that?) as well as links back to the future from earlier posted content, e.g. by people linking to “future” posts from the comments on older posts (think about it…)

So what? So I have no idea… but building context and support through links to other content is one way of reusing that content (or at least, pulling it into a new context and making it (re)usable in that new context).

So when George Siemens notes:

I haven’t come across research to date that discusses how open educational resources are being used. Yes, we get information like “MIT’s OCW gets X number of million hits per month”.

and then goes on to ask: “I’m interested in whether or not universities are using open resources produced by other universities.” I guess my answer is – if I simply link to a relevant OER from an appropriate context, then I’ve provided the opportunity for that resource to be reused?

So links make for easy reuse, at least in a blogged uncourse sense… which means URIs (URLs, whatever…;-) are important… which brings the idea of Linked Data into play… Why? Here’s one example: Telling (non-linear) stories (which also appears here: Building coherence at bbc.co.uk).

As to why you should go and read that post now? It’s illustrated by this:

Now go do your homework and prepare to make some changes when you get back in to work next week…

The notion of linear courses has just left the building…

… again…

A quickie post, this one, to complement a post from a long time ago where I plotted out – as a network – the links between people who served on the same committee on the Isle of Wight Council (Visualising CoAuthors in Open Repository Online Papers, Part 3, half way through the post).

In this case, I trawled the Isle of Wight Council committees to populate the rows of a spreadsheet with column headings “Committee Name” and “Councillor”.

Pasting the results into Many Eyes gives an IW Council membership dataset that can be easily visualised. So for example, here’s a glanceable treemap showing the membership of each committee:

The search tool adds yet another dimension to the visualisation, in this case allowing us to pick out the various committees the searched for named individual sits on.

Here’s a glanceable treemap showing the committees each councillor is a member of:

It strikes me that if the search tool supported Boolean expressions, such as AND and OR (maybe with each term being realised by a different colour bounding box?), it would be possible to explore the variation – or similarity – in make-up of different committees? On the first tree map, this approach would make it obvious which committees the same groups of people were sitting on?

And why would we want to do this? To identify potential clashes of interest, maybe, or a lack of variation in the composition of different committees that might, ideally, be independent of each other?

PS Hmm, I suppose you could use a similar visualisation to look at the distribution of named directors across FTSE 100 companies and their subsidiaries, suppliers and competitors, for example? ;-) Does anyone have simple lists of such information in a spreadsheet anywhere?;-)

Whenever you write a blog post that contains links to other posts, and is maybe in turn linked to from other blog posts, how can you keep track of where you blog post sits “in the wider scheme of things”?

In Trackforward – Following the Consequences with N’th Order Trackbacks, I showed a technique for tracking the posts that link to a particular URI, the posts that link to those posts and so on, suggesting a way of keeping track of any conversational threads that are started by a particular post. (This is also related to OUseful Info: Trackback Graphs and Blog Categories.)

In this post, I’ll try to generalise that thinking a little more to see if there’s anything we might learn by exploring that part of the “linkgraph” in the immediate vicinity of a particular URI. I’m not sure where this will go, so I’ve built in the possibility of spreading this thought over several posts.

So to begin with, imagine I write a post (POST) that contains links to three other posts (POST1, POST2, POST3). (Graphs are plotted using Ajax/Graphviz.)

In turn, two posts (POSTA, POSTB) might link back to my post:

So by looking at the links from my post to other posts, and looking at trackbacks to my post (or using the link: search limit applied to the URI of my post on a search engine) I can locate my post in its immediate “link neighbourhood”:

Now it might be that I want to track the posts that refer to posts that referred to my post (which is what the trackforward demo explored).

You might also be interested in seeing what else the posts that have referred to my original post have linked to:

Another possibility is tracking posts that refer to posts that I referred to:

It might be that one of those posts also refers to my post:

So what…. so I need to take a break now – more in a later post…

See also: Tweetbacks, a beta service that provides a trackback like service from tweets that reference a particular URL.

PS and also BackType and BackTweets

Chatting to Simon Perry of the Ventnor Blog over a pint at the Yarbridge Inn last week, he mentioned that recording of the Isle of Wight Council Planning Committee were available on the Isle of Wight Council website, albeit in an obfuscated and hard to find way (you know the sort of thing: a full stop or 1×1.gif is used as the link text and the anchor has {text-decoration: none};-).

So I thought it might be interesting to see how easy it would be to plot the recordings on a map, “locating” the recording of each application at the place where the planned changes would actually take place. The idea being, of course, that a map based index makes it easier to find information about planning applications in your own locale.

The recipe I had in mind was something like the following:

- scrape a list of recordings, along with the location each one referred to, and make it available as JSON or RSS feed;
- take the feed, geocode it as required, and hook it into a “geopodcast” map, (that is, a pre-existing application that would take a geocoded podcast feed and display it on a map, letting you click on a marker and play the audio file located at that point).

Easy, right? A half hour job, I thought…

Hmmm…maybe not:

- the only place I could find links to the audio files were in the minutes of the relevant committee meeting, minutes that are only published as PDF documents; and PDF scraping is not something I know how to do (yet…?!);
- a quick search around turned up no obvious geopodcast plotting maps.

(Maybe the 4ip funded AudioBoo project, which lets users users to record and share audio from their mobile phones, will also spin off an easy to use “geopodcast map plotter”…? (Let’s also hope AudioBoo gets more traction than audiotagger did!))

So here’s what I ended up doing instead for a proof of concept. Firstly, from the Planning Committee webpage, I opened one of the minutes PDFs, and cut and pasted the details of a planning decision into a Yahoo pipe to create a test feed. (The title of the application links to the audio recording of it’s consideration – so you can check the veracity of the minutes if you want to…)

Here’s the pipe:

A couple of things to note:
- the planning committee minutes don’t make it easy to get a geocoded position for the application; I ended up using the Post Office postcode finder to get a postcode from the address stated in the application, so that I could get a reasonable fix on the location with the (rather crappy) Yahoo pipes geocoder block.

- the URl of the audio files are very long and truly horrible; when I was testing embed codes for various media players, they would occasionally choke on the URI (possibly because I wasn’t escaping or encoding it when I should have been?); anyway, a simple fix was to just get a minified version of the URI and pass that to the audio player (the is.gd block is one I found that will minify a supplied URI).

So that’s the pipe.

For the map, I pulled out the code from my (now broken…) geotwitterous app, and added the ability to display an embedded audio player in marker pop-up box:

Here’s the demo: GeoAudio demo: IW planning applications.

For the future? An obvious next step would be to just cut details from planning committee minutes and paste them into a Google spreadsheet, then take a CSV output into a pipe, geocode it, and pass it into the map. But that’s for another day…

In the longer term, pulling together all the relevant documents associated with a planning application (maybe using the Plannig Alerts API?) into a single interface would be handy. (I’d also love to see kids in local schools and the local college doing some practical ICT/DT CAD work generating 3D Sketch-up versions of planning applications so they can be viewed in Google Earth;-)

Now the dust has settled from last week’s US Presidential inauguration, I thought I’d have a look around for interactive photo exhibits that recorded the event. (I’ll maintain a list here if and when I find anything else to add.)

So here’s what I found…

Time Lapse photo (Washington Post)

Satellite Image of the National Mall (Washington Post)

A half-metre resolution satellite image over Washington taken around the time of the inauguration.

You can also see this GeoEye image in Google Earth.

Gigapixel Photo of the Inauguration (David Bergman)

Read more about how this photo was taken here: How I Made a 1,474-Megapixel Photo During President Obama’s Inaugural Address.

Interactive Panorama From the Crowds (New York Times)

PhotoSynth collage (CNN)

I suppose the next thing to consider is this: what sort of mashup is possible using these different sources?!;-)

[PS If you find any more interactive photo exhibits with a similar grandeur of scale, please add a link in a comment to this post:-)]

A flurry of posts around the interwebs today (e.g. + A CONCEALED ASSAULT ON PRIVACY +) picked up on some proposed amendments to the Data Protection Act that have found their way into the Coroners and Justice Bill that I posted about last week (Data Sharing is Good, Right? Or is HM Gov Evil?).

It struck me that it would be really handy to have a tool that could alert you to proposed amendments in your favourite Act in whatever Bills happen to be live at the moment.

A quick look at one of the .gov.uk websites provides an advanced search form that lets you search over current bills – UK Parliament Advanced Search:

Running a search on the phrase “the Data Protection Act 1998″ and sorting the results by most recent first gave me a URL I could tinker with…

So here’s a pipe that’ll grab the most recent bills mentioning a particular act:

Clicking on a link should take you to the point in a Bill that mentions the Act you’re interested in:

Being a pipe, I get the RSS/JSON feed for free… which I can now subscribe to and use as an alerting service (for as long as the pipe’s screenscraping part works!) Ideally, of course, the parliamentary search would make results available as RSS…

As ever, this pipe took almost as much time to blog as it took to create…!

So maybe Charles Arthur should rethink If I had one piece of advice to a journalist starting out now, it would be: learn to code and instead focus on Learning to Think Like A Programmer?;-)

PS see also: They Work For You: Free Our Bills and They Work For You: Free Our Bills (Techy Stuff).

I had an email today from OUseful.info reader Stephen Harlow asking a Yahoo Pipes related question (I get this evry 2-3 weeks, at the moment, and try to answer them as best I can): “My problem is that the URL Builder in Pipes seems to muddle the order of query parameters in the URL… Is there anyway of fixing the order within Pipes’ URL builder?”

Hmm… so here’s the problem:

(Is this a reverse alphabetical ordering by the Pipe I wonder? One to explore…)

Now most of the time, the order of the arguments in a URI doesn’t matter, although for some systems, (as seems to be the case for the crappy Library OPAC Stephen was building a pipe for) it does matter.

So here was my suggested fix: use the String Builder to build the URI:

If necessary, you can get the argument values from a user input in the normal way:

I’m not sure if this sorted Stephen’s problem, but it’s another trick to remember… :-)

There’s been quite a few posts around lately commenting on how Google is missing out on real time web traffic (e.g. Sorry Google, You Missed the Real-Time Web! and Why Google Must Worry About Twitter). Stats just out showing search’n'twitter activity during Obama’s inauguration yesterday show how…

First up, Google’s traffic slump:

And then Twitter’s traffic peak:

And how did I find out? Via this:

which led to the posts containing the traffic graphs shown above.

And I got a link to that tweet from Adam Gurri who was responding to a tweet I’d made about a the Google traffic post… (which in turn was a follow-up to a tweet I posted about whether anyone “turned on a TV to watch the US presidential inauguration yesterday?”)

Adam also pointed me to this nice observation:

And here’s how a few people responded to how they watched the event (I watched on the web):

So, web for video, broadcast radio for audio and Twitter for text. And TV for, err… time for change, maybe?

So we all know that the Google web search engine famously (and not just apocryphally) returns different results from it’s different national representations (google.com. google.co.uk, google.cn, etc.)…

…and hopefully we all know that if you are signed in to Google when you run a search, the default settings are such that Google will record your search and search results click-thru behaviour using Google Web History, and then in turn potentially use this intelligence to tweak your personal search results…

…and depending on how much you’ve been paying attention, you may know that Google Search Wiki lets you “customize search by re-ranking, deleting, adding, and commenting on search results. With just a single click you can move the results you like to the top or add a new site. You can also write notes attached to a particular site and remove results that you don’t feel belong. These modifications will be shown to you every time you do the same search in the future.“

Well now it seems that Google is experimenting with Google Preferred Sites, which let selected guinea pigs “set your Google Web Search preferences so that your search results match your unique tastes and needs. Fill in the sites you rely on the most, and results from your preferred sites will show up more often when they’re relevant to your search query” (see the official support page here”: Preferences: Preferred sites).

So the next time you give someone directions to a website using an instruction of the form “just google whatever, and it’ll be the first or second result”, bear in mind that it might not be…

(For what it’s worth, I run a cookie free, never logged in to Google browser to compare the results I get from my logged in’n'personalised Google results page and a raw organic” Google results page.)