An ‘expert talk’ by Joss and some pitches for the project at the JISCRI conference: http://devcsi.ukoln.ac.uk/demonstrator/tag/jiscpress/

An interview with Alex on the dev8D blog: http://dev8d.jiscinvolve.org/2010/02/24/interview-alex-bilbie/

We were also featured in the ‘Tool Shed’, a newspaper for the jiscri programme. I don’t think it’s online, but it’s a fine publication!

Questions

1. Can you tell us a bit about CommentPress and why the move to digress.it? (00:10)
2. What design decisions have you made for digress.it? Is there anything that other developers should be aware of? (06:00)
3. What single area of work on the JISCPress Project has been the most time-consuming (and therefore expensive)? (10:45)
4. What’s been the biggest challenge for you on the JISCPress Project? (20:40)
5. Paragraph-level trackbacks and remote embedding of paragraphs which also provided a trackback, were two requirements we kept pushing for. What problems still remain with these features? (23:50)
6. What software tools or productivity methods do you use and how do you use them? (34:40)
7. What was the most important thing that brought value to your work? (52:40)
8. What’s the future of digress.it? How will it be sustained now the JISCPress project has finished? (55:16)
9. Any more plans for digress.it? (01:02:15)
10. You’ve started writing a digress.it server, right? (01:04:56)

Eddie Tejeda talks about digress.it and JISCPress from University of Lincoln on Vimeo.

In our original bid, we proposed a ‘prototype demonstrator platform’ for JISC’s Funding Calls and Final Project Reports. We outlined 11 deliverables:

1. A WordPress Multi-User based platform for authoring and publishing JISC funding calls in a form that allows paragraph-level comment and discussion either locally or remotely.
2. A meta-site that aggregates all document data into a single site for search, navigation by categories and tags and can syndicate searches, tags and categories.
3. Develop CommentPress to meet WCAG 2.0 accessibility guidelines, meeting public sector requirements.
4. Evaluation and integration of “related content” utilities to dynamically link related project calls and reports based on content and/or semantic analysis.
5. Evaluation and possible integration of remote, realtime messaging services such as Twitter and XMPP integration.
6. Evaluation and possible integration of enterprise authentication services such as LDAP and Shibboleth.
7. Evaluation and possible integration of OpenCalais, a semantic tagging service.
8. Documentation on how to exploit the benefits of AWS and clone the project instance for other uses.
9. A documented suggested workflow for document authors
10. Documented examples of how to fully exploit the platform for data extraction and syndication.
11. Documented ‘user stories’ for the JISC funding call process. Note that we do not guarantee fulfillment of all user stories.

I’ll go through each of these one by one with illustrations where relevant. A more informal reflection is also available (Thoughts on JISCPress ):

Paragraph level commenting and discussion of JISC funding calls

This was achieved through the development of the digress.it plugin. digress.it is a rewrite of the original CommentPress WordPress theme which we used on WriteToReply (which JISCPress is based on). I’ve posted a video interview with Eddie Tejeda, developer of the original CommentPress and digress.it, where he discusses the move from CommentPress to digress.it. In terms of local and remote paragraph commenting, the same feature set found in CommentPress has been retained. Remote, document section level comments are possible through the use of trackbacks.

We spent quite some time looking at remote paragraph level remote commenting and Eddie expects to support this with digress.it in the near future. We discovered that the use of trackbacks and pingbacks is an unreliable method of guaranteeing ‘comments’ from remote websites. It depends on the CMS being used and the settings of both the remote and local site. Sometimes, test comments we made never arrived, other times they did. So for example, whilst internal links within a Wordpress domain may be recognised by other sites on the same platform, links from posts on other blogging platforms may not be. Link tracking using third party services (e.g. Google, Google blogsearch, BackType) rely on links being hardcoded in third party web pages (rather than being added dynamically to a page via Javascript, or within an embed object) and even then are not detected reliably (it depends on the crawler). Commercial tracking/monitoring services  were not explored.

WordPress provides a robust commenting system, with excellent spam filtering and comment moderation features. digress.it leverages this locally to allow commenters to respond at the paragraph, rather than the section (i.e. blog post) level. For more on digress.it, Eddie talks in length about his work in a previously posted video interview.

An aggregated meta-site

Alex has been working on this, which can be seen in the screenshot below (and until we take the server down, can be browsed at http://jiscpress.org )

What you see here are a number of ways of finding documents on the site. The large tag cloud uses tags generated from Alex’s Open Calais/Yahoo Term Extractor related tags plugin. This plugin uses both these third-party APIs to tag each document and then create intelligent relationships between documents on the site. More on that later. The tags are held in a separate database table to the human created, native WordPress tags, but are equally a source of information that can be used by theme designers and plugin authors. Here the tags are simply being used to display a cloud, similar to the one on http://en.wordpress.com/tags/ and marked up with the rel=”tag” microformat.

Clicking a tag lists the documents by title

As you can see, each result for a tag has an RSS feed, which can be found at the top right of the results. So if you’re interested in watching for key words in JISC documents, this would be a useful way of doing that. RSS feds can be monitored from feed readers such as Google Reader, or via web desktops, such as Netvibes (e.g. An Example Netvibes Dashboard). You don’t have to use the tag cloud to do that. You can construct your own and wait for the results to come in. i.e. http://jiscpress.org/?jiscpress_tag=MY_KEYWORD&feed

(Note that Wordpress also offers the option of subscribing to a free text search using the default Wordpress search utility, e.g. http://jiscpress.blogs.lincoln.ac.uk/feed/?s=jiscpress )

Similarly, you can do this with ‘Topics’ (otherwise known as WordPress categories), which are aggregated from across all documents and displayed on the right side of the home page. For JISC’s purposes, there is a controlled list of about 40 Topics that are used by the organisation. Our example shows the use of a few of those. Again, if you’re interested in Funding Calls for Data & Text Mining, then you can subscribe to a feed for that Topic.

The main thing to point out about the use of WordPress categories on the Home Page, is that it assumes a controlled list and not a publishing environment where authors make up their own taxonomy. The list would get very very long and unmanageable. It need not use JISC’s Topics. Their Themes and Programmes would also work.

You’ll see that it displays the document title and author’s name. The use of author’s names is worth considering too. While WordPress is a multi-user CMS, it may be that each Programme decides to publish under their Programme’s name rather than the individual’s name. This is just a matter of changing the settings in WordPress, so that all the IE Team publish under their Programme’s name. The choice is up to JISC.

Above the ‘Topics’, is a list of the latest funding calls. It’s just a text box which some HTML links pointing to the latest documents. Nothing fancy nor difficult to maintain either.

In the backend, Alex has provided some options for the Tag Cloud (otherwise known as the JISCPress Browser widget). You can see that we have the option of using User, Open Calais and Yahoo tags, as well as blacklisting the display of certain tags, too. You can also decide how many tags you want to display.

The Topics are listed using a simple WPMU Site-wide categories widget that Alex wrote. It looks at the categories across all documents/blogs and displays them in alphabetical order.

For searching, we’re using the built in BuddyPress search (did I mention we’re using BuddyPress?). It simply allows you to search document titles from the front page or, on the Authors page, you can search authors by name, too.

Each Author has a profile page.

We looked at full-text search and it’s quite possible using this lucene-based plugin, which indexes all document text and all comments, too. It could be integrated into the theme to allow site-wide full text search but we chose not to because of time constraints and are simply using the built-in BuddyPress search. If JISC would like full-text search, it’s something that Alex could do. Of course, full-text search is possible on each document site and constructing searches derived from the Open Calais and Yahoo Term Extraction plugin is also possible as I’ve shown above. Full text search could end up returning more results than are useful. I’ve no strong opinions about this either way.

Finally, you can see in the menu bar on the front page that there’s an ‘About JISCPress’ page and a JISCPress blog. Nothing fancy going on there. We’re just using basic BuddyPress features.

Accessibility

Well, we haven’t ignored accessibility requirements but then again, they haven’t been a driving factor in the development of JISCPress either. I think we’ve improved on CommentPress and have had useful feedback from the British Computer Association for the Blind on the accessibility of digress.it. At one point Eddie included the jQuery.accessible plugin in digress.it but it’s not currently being used. Along similar lines, much time was spent on IE6 compatibility, which has been achieved at some cost to the project. The user experience in IE6 is not as nice as using Chrome, for example, and I wonder whether it was worth the effort in a project such as this that was about producing a ‘prototype demonstrator’. Nevertheless, because digress.it is now in widespread use elsewhere, IE 6 & 7 compatibility was one of the first requests that came through on the mailing list. We did a quick survey of browser use in HEIs and Andy Powell followed this up with something more detailed and wide ranging. What both show is IE 6 & 7 can’t be ignored

I feel that we didn’t manage to do as much toward accessibility as I originally hoped but it is something that can be worked on with digress.it over time and I know it is something that interests Eddie a great deal. Hopefully as he does work for more organisations, like Cornell and the New York Public Library, their accessibility requirements will filter through into the core code. In addition, Eddie has recently been able to employ a designer to work with him on digress.it so the theme should get more close attention over the next few releases.

Related content utilities

Alex did a lot of work on this and has released his wpmu-related-blogs-and-posts plugin on the official WordPress plugin repository.

Here’s an overview:

The WPMU Site Admin options look like this (click the image to see it full size):

You can see that both or either the Open Calais and Yahoo Term Extraction APIs can be used. The plugin provides a background service which runs via a cron job, which can be set to daily, twice daily or hourly. The cron job can be started manually and the entire platform can be re-tagged at any time. The relationships between document sections and documents can be re-established at any time, too.

Both the relevance of the tags (using features of the APIs) and the relevancy of the posts (when showing related document sections), can be adjusted.

Finally, you can opt to ignore certain blogs/documents, so test sites and the main site don’t mess up the weighting of the relationships made.

As we’ve seen with the JISCPress Browser widget, those tags can be used to provide a way to navigate the entire platform. However, the principle intended use of the Open Calais/Yahoo services is to display related document sections or, optionally, related documents while reading any given document. One potential issue with the auto-tagging services relates to the variable quality and usefulness of the tags they return. One possible way of addressing this would be to limit the range of tags used on the site by filtering the automatically generated tags via a whitelist, blacklist, or based on semantics (e.g. ignore placenames).

Here’s the widget options:

Here’s the widget as displayed to the reader. In principle, it works and should work better as the number of documents on the site grows.

Related documents/blogs:

Related document sections/blog posts:

Realtime messaging

This deliverable has been tackled in two ways. Eddie developed realtime alerts for digress.it so that if someone comments on the site while you’re reading it, the comment will ‘pulse’ in the Comment Box. For remote realtime messaging, things have been taken care for us. As realtime on the web is becoming very much mainstream, it’s getting easier to push content out through a variety of ways. For example, since the start of our project, WordPress now has RSSCloud and Web Hooks plugins. The latter provides a relatively simple framework for developers to push any notification from WordPress to external services, as I’ve discussed on my blog. There are also SUP (FriendFeed) and PubSubHubbub plugins, too. In addition, Google is now indexing in realtime and Google Alerts are showing up via RSS almost immediately, too. XMPP PubSub is also in use on WordPress.com and that work is due to be released as  plugin for WordPress once it has been well tested. Likewise, their work on a Twitter compatible API for WordPress will also be released. Given all of this, we didn’t think we needed to reinvent the realtime wheel for WordPress.

Authentication Services

I can confirm that LDAP works well with WPMU. I know this because I run WPMU with this plugin at the University of Lincoln. It is a feature rich plugin and well supported. If anyone wants to integrate LDAP with WPMU and is running into problems, please get in touch. Note that there is no reason why much of JISCPress couldn’t be used as a private platform for document discussion and annotation, internal to an organisation.

Shibboleth support was not tested because we don’t use it at Lincoln and I never got around to trying to convince someone to help me test it. However, I have been told by the plugin developer that it is in use at other universities and it too, is well supported by Will Norris, the developer of the WordPress Open ID plugin. I would be very grateful if someone running a Shibboleth service at their institution would help me test the plugin. I’m pretty confident that it will work.

Open Calais

See above!

Amazon Web Services

I documented our use of Amazon Web Services on our Google Code site. I had intended to leave an image of the JISCPress server on AWS so that anyone could clone it for use or play. However, unless asked, I’m not going to do that now. If you want to use such a system, you can use http://digress.it or http://writetoreply.org or and if you’re skilled enough to work on AWS, then you’re skilled enough to install WPMU and our plugins on your own server for testing purposes. I’d be happy to advise anyone wanting to create their own version of JISCPress and even work on your server if you want me to. Leaving an Amazon Machine Image lying around for testing means that it will soon go out of date as new versions of WordPress are released and need rebuilding anyway.

Workflow for authors

I have begun writing documentation for authors on our Google Code site. I’d be grateful for feedback. This documentation is also used on the http://digress.it site. I also intend to provide documentation for Administrators, too. Although if you are running WPMU, then you will be familiar with much of what you need to know. One of the great things about this project is that we’re using one of the most popular pieces of publishing software on the web and there is already a lot of familiarity with it.

Data extraction and syndication

A lot of work went into developing digress.it with this in mind and Alex has also developed a plugin that posts RDF triples to the Talis Platform from WPMU. I’ll say more about that below, but here’s a post I wrote recently about JISCPress and Open Data. I hope you’ll agree that we’ve made good progress in this area.

A demonstration of how paragraph level content from JISCPress can be transcluded (that is, embedded) on a third party site is available here: Paragraph Embedding from JISCPress. The post includes a Javascript snippet that can pull in a random paragraph from a contiguous set of paragraphs on a single JISCPress page. Such a utility might be used to rotate through headline pargraphs on a JISCPress document fron page, for example.

On the subject of web analytics, we reviewed how to collect JICWEBS recommended metrics using Google Analytics (Measuring Website Usage With Google Analytics, Part I , how to monitor traffic coming in from University networks, and how to exploit Google Analytics campaign tracking codes from shared links and via Feedburner. We did intend to look at Piwik, too, as we’ve tinkered with it on WriteToReply, but I’m afraid it didn’t happen. I do want to highlight Piwik to anyone interested in open source analytics software. It allows you to expose the analytics for any given site to the public and provides a number of ways to access the data, via CSV, JSON, XML, RSS. Have a play around on the WriteToReply site to see what I mean. Naturally, there is a WordPress plugin for Piwik, too. Google Analytics has also recently opened up an API that would support the development of custom reporting dashboards. Steph Gray at Helpful Technology has recently described an “ultimate dashboard” built around free tracking/monitoring services – Minding the shop – and is willing to share the code. Such a dashboard could be used to provide a useful overview of document related discussion on the wider web.

Documented user stories

This is an area where I wonder how we might have improved things. As I’ve written about before, the project team has worked virtually, hardly ever meeting and we worked very much in public, with digress.it having a fairly broad range of users mid-way into the project. For some of our work, we were not short of user feedback but for other areas, we received hardly any. I set up a UserVoice site early on and we’ve asked people to use it and linked to it from our blog, but it’s never seen any use. Feedback on the project in general has been quite low, although we receive informal feedback on how WriteToReply might be improved quite regularly and similarly, there has been a lot of feedback on digress.it over Twitter.

I have tried to document some of the uses of JISCPress that I could think of. They’re not really User Stories or even complete scenarios, but it does sketch out some of the uses that I could think of and I’ll add to them as I think of them. It will be interesting to demonstrate JISCPress to JISC and be able to respond to their feedback now we have a working platform. JISC staff have started using digress.it on WriteToReply (the #jisclms Call was re-published by two JISC staff) and I anticipate there being more opportunities for us to work with JISC staff on using some of the work we’ve done in the near future.

Other work

I mentioned the work that Alex has done on connecting WPMU to the Talis Platform. Using Triplify, he’s developed a plugin that runs a service which posts RDF Linked Data to Talis. It can be downloaded from the WordPress repository.

The plugin has some global options, available to the platform admin, which are pretty self-explanatory (click to enlarge)

Each site owner also has the option to participate, too. First you agree:

and then you set your license:

The RDF data is available at both the http://example.jiscpress.org/triplify endpoint and for query on Talis. This is what you get back, for example:

The configuration file for WPMU is available here. Be sure to check it over to ensure that it exposes the data you want it to expose and no more! Also, do read the Triplify documentation. You’ll see that it does more than produce RDF. There’s JSON output, too.

Final reflections

These final reflections are from Joss.

I’m pretty pleased with the way the project has gone. It’s been a real pleasure to work on the project and to work with Tony, Eddie and Alex. Not once have I felt it a chore even at some of the ridiculous hours of the night that we’ve worked. I’m also very grateful to JISC for thinking that it was a project worth funding and the support from JISC in terms of the JISCRI conference/IE Demonstrator and general lack of interference, but steady encouragement, has been really welcome, too.

I think we’ve delivered enough of the original objectives to have made it all worthwhile. There are areas, such as accessibility and more user testing, which I wish we’d been able to do better, but it wasn’t for want of trying. I really wish we’d cracked simple embedding of paragraphs and paragraph-level trackbacks, too. We’re close, but not quite there, yet.

On the other hand, we have produced some pretty nice plugins for WordPress. For WordPress developers, digress.it provides some significant innovations around document publishing and Open Data in WordPress and the code is free to be built on and improved. Likewise, we’ve also produced the first plugin that allows WPMU admins to run a background service for Open Calais and Yahoo Term Extraction across all blogs/documents. Another ‘first’ for the project, is that we’ve joined WPMU to Triplify and the Talis Platform. Huge amounts of data are generated by WPMU installations and now that can be Linked Data hosted on a well known Data Store. It would not be difficult to bring both plugins together so that the Open Calais semantic tags are included as data that is posted to the Talis Platform.

The legacy of the project will hopefully be similar to what we’re trying to achieve with WriteToReply. Not necessarily a wholesale conversion of organisations to using the platform with all its features that I’ve described above, but for people to pick and choose what works best for them. It may be that the project just inspires something better or different on another platform, like Drupal, and that’s good, too. It’s also about trying to demonstrate how publishing and engagement with public documents on the web can still, very much, be improved. In that sense, there is probably as much value in the blog posts that our work has generated as the software itself. We’ll continue to write about this on the WriteToReply blog.

All the code is open source and available under the GPL (digress.it) or Modified BSD license (Calais and Talis plugins). Eddie continues to maintain digress.it and I know that Alex is keen to ensure that any issues identified with his code are dealt with too. If you use the code and improve it in anyway, do tell us as we’re keen to include contributions from other developers.

]]> http://jiscpress.blogs.lincoln.ac.uk/2009/12/16/the-jiscpress-prototype-demonstrator-platform/feed/ 4 Paragraph level comments JISCPress Home Page Clicking on a tag JISCPress Browser widget options Author profiles WPMU Related posts admin options WPMU Related Posts widget Related Documents Related Sections Talis Global Options Talis T&C Talis site options Talis Platform query result http://jiscpress.blogs.lincoln.ac.uk/2009/12/07/following-on-from-my-notes-on-performanc/ http://jiscpress.blogs.lincoln.ac.uk/2009/12/07/following-on-from-my-notes-on-performanc/#comments Mon, 07 Dec 2009 09:49:40 +0000 Joss Winn http://jiscpress.blogs.lincoln.ac.uk/2009/12/07/following-on-from-my-notes-on-performanc/ Following on from my notes on performance, here’s a brief video showing the speed of browsing a WriteToReply consultation document using digress.it. digress.it performance

JISCPress: A prototype demonstrator publishing platform for JISC funding calls and project reports.

Screenshots or diagram of prototype

Description of Prototype

JISCPress allows communities to comment on, discuss, annotate and review documents in considerable detail. As a platform, JISCPress discovers relationships between hosted documents and provides a variety of ways to make discovery easy and useful.

Link to working prototype

http://jiscpress.org <– temporary until Feb 2010 (see also http://writetoreply.org)

Link to end user documentation

http://code.google.com/p/jiscpress/wiki/Documentation

Link to code repository or API

http://code.google.com/p/jiscpress/source/

http://code.google.com/p/digressit/source/

Link to technical documentation

http://code.google.com/p/jiscpress/wiki/Documentation

Date prototype was launched

17/12/09

Project Team Names, Emails and Organisations

Joss Winn, Alex Bilbie (University of Lincoln)

Tony Hirst (Open University)

Eddie Tejeda (Freelance developer, Visudo)

Project Website

http://jiscpress.blogs.lincoln.ac.uk

PIMS entry

https://pims.jisc.ac.uk/projects/view/1348

Table of Contents for Project Posts

Please use the tag cloud in the right-hand sidebar.

The four members of the JISCPress team are:

Joss Winn (Staff, University of Lincoln)

Tony Hirst (Staff, Open University)

Alex Bilbie (Student, University of Lincoln)

Eddie Tejeda (Freelance developer, Visudo, San Francisco)

Project Methodology?

I wrote a bit about project methodology at the start of the project. The emphasis from JISC on using an ‘agile methodology’ just didn’t seem to fit the structure of our Team. All the agile methodologies I knew of emphasised the importance of working closely, sometimes even in pairs, and seemed designed for tightly focused, customer orientated projects with regular iterations around sprints of development. There was no way that I was going to be able to organise Tony, Eddie and Alex around regular sprints and had I tried to crack the whip, I think I would have received far less co-operation from them than I did. As it turned out, the team never actually met in the same room (err, pub…) until the JISC CETIS Conference earlier this month. On a couple of occasions during the summer, Alex worked in my office, but it wasn’t especially productive. It was clear that he was much more comfortable working at home. While Eddie was over for the CETIS conference (he paid his own way here, I want to add!), he came to Lincoln and worked for a day in the office with me. It was a productive day and I could see how teams working together like that could be highly productive both in the development of ideas and code but it wasn’t unlike how Eddie and I would work together virtually.

The other significant contributing factor to the way we worked was the type of development we were doing. Again, agile methodologies seem designed around creating a user-led product. There’s an expectation that the development team are working for a client who is embedded in the decision making process for the project. The client ‘owns’ the project and they ultimately own the code, too.

JISCPress didn’t have this type of client and we weren’t solely focused on the writing of new code either. The project was as much about taking existing code and piecing it together in a fruitful way, developing what we’d already pieced together with WriteToReply. We started with the WordPress Multi-User platform and CommentPress. We looked at some other existing WordPress plugins for working with Open Calais and creating relationships between content and also created a configuration file for Triplify and a plugin to publish data to the Talis Platform. We bootstrapped a complete rewrite of CommentPress, called digress.it and Alex wrote plugins for making WPMU work with OpenCalais and Triplify.

IPR

The University of Lincoln has no interest in ‘owning’ the code (no-one here would maintain it), JISC have no interest in owning the code (no-one there would maintain it) and the code is completely reliant on other open source software and APIs which we don’t maintain either. I was advised by OSS Watch to ensure that the code we funded was, wherever possible, copyright University of Lincoln, but I’ve ignored that advice. I just don’t see who it would benefit. In order to make the work interesting and sustainable, we needed Eddie and Alex to have a vested interest in the code they were writing. Eddie’s work on CommentPress was already credited to him and Jesse Wilbur and licensed under the GPL3. There was no advantage in the University of Lincoln claiming copyright in the digress.it code we were funding and I’m sure that had we insisted on this, Eddie wouldn’t have worked for us. We could have claimed copyright for the code that Alex has written, but as I said, no-one here will maintain it, so I think it’s better to encourage Alex to remain the copyright holder in the hope that he might maintain it for a while. There is no ££ value in WordPress plugin code. There’s only value in providing services around the code and Alex is the person best suited to do that, not the University.

How we actually worked together

My original intention was that we’d work together in public on IRC. Tony and I had started doing this on WriteToReply and thought we’d just continue the weekly meetings we were having. This happened once at the start of the project, but after that, we all just used Google Talk. The reason for this is simple: we all use it anyway. Eddie and I work with GMail open and rather than firing up another application just to ask a quick question, we used GTalk. It allowed us to see when we were online and effortlessly kept a log of everything we were saying. Alex and Tony also seemed to prefer to use it, too. Because it was attached to our personal email addresses, rather than work addresses, it also meant that we were available after 5pm and over the weekends. The convenience of using GTalk meant that on many occasions, I’d be contacted by Eddie as I was going to bed and end up spending late nights testing his latest code revisions and feeding back comments via GTalk. Had we relied on other tools, I wouldn’t have been in the habit of using them at home and consequently Eddie and I would have chatted far less than we did.

Our ‘office’ consisted of the Google Code svn repository and Google Talk. Eddie would check code in, I’d be waiting on the jiscpress.org server to check it out and then we’d chat about it over Google Talk. Anything that required time to fix, was added to the list of Issues. Alex and I worked differently. He’d usually come to my office on campus and show me what he’d done. I preferred to work over svn and Alex taught himself how to use it, but we never worked in real-time over svn as Eddie and I do.

Tony’s role was always one of provocateur and champion for the project. A lot of the best ideas (and hardest to implement) originated with Tony and then I would try to push them into Eddie’s list of tasks, which were hosted on the digress.it Google Code site.

Working with students

I think we’ve been really lucky to have Alex on the team. In the Centre for Educational Research and Development, where I work, we’re keen to work with and undertake research with students and it seemed like a natural extension of this  to be working with Alex. The project was nicely timed over the summer holidays, too, so Alex had quite a bit of time to spare during July – September. Noticeably, when the semester started again, he was very pushed for time partly because I’d found him other paying projects to work on and regrettably I ended up competing for his time! I should add that both Eddie and Alex worked more hours than they were actually paid for – not unusual when working in education, but from my point of view, it meant I was juggling a lot of good will at times.

Alex also came to the JISCRI and CETIS conferences where he seemed to really get a lot out of meeting other developers and talking about the work he has been doing on the project. I think all of us would say that the project has been an opportunity to learn from each other and hopefully we’ve documented the useful bits on this blog for others to learn from too. Eddie talks about his work for the project in a video interview I made with him. I’ll link to it from here when it’s ready.

Who were our users?

Agile methodologies put a lot of emphasis on the user and their stories. We didn’t arrange for a distinct set of users to feed back to the project. The main Stakeholders for the project were JISC and the JISC community, not Lincoln staff and students, so we needed to engage them as much as possible. We had already done this to a certain degree through WriteToReply and had republished the JISCRI call specifically to see how the community might use the platform. During the project, JISC launched their draft Strategy and a consultation on the Google Book Settlement, both of which used early versions of the digress.it code we were were developing. On a few occasions, I pointed people via Twitter and the blog to our UserVoice site, but no-one offered anything this way. Tony acted as an uber-user, looking at the platform from the point of view of a developer/mashup-artist and consequently fed back lots of thoughts for how it could be improved. Tony and I have also spent hours and hours publishing documents via WriteToReply and we’re acutely aware of the issues that publishing users might face. We also gained a lot of users following the launch of digress.it and it was embraced by the WordPress community (931 downloads and counting) as well as people signing up to Eddie’s site http://digress.it A mailing list specifically for digress.it was set up and people would feedback issues and requests that way. We also have a public mailing list for the JISCPress project as a whole. It has a few people lurking on it, but has not been a particularly effective method of communication. Perhaps it will be if people pick up the work we’ve done and try to set up their own version of JISCPress, as I hope they will do. As a public open source project, we pursued our original ideas in public and as the code was released, waited for users to feed back to us. Alex, too, got feedback in this way for his OpenCalais plugin and was made changes to his code based on a suggestion from another developer who is using it.

Misc.

A couple of things I didn’t plan for when working with Alex and Eddie were payroll related. Customs and Excise have told us that we have to pay VAT on Eddie’s invoices.  Fair enough. It didn’t occur to me because he doesn’t reside in the UK, but it makes sense. I learned a lesson and it didn’t affect the budget because the exchange rate improved in our favour over the course of the project and what we saved there, we spent on VAT.

Alex began as a freelancer with the intention of invoicing the project for his work at the agreed amount, but the university said it would be easier for Alex if he was put on the payroll and then his tax and NI would be dealt with. It affected the project by adding about £30 in NI contributions to the total amount we’d agreed – good value, I think. One less set of invoices that I had to deal with.

Did it work?

As I said at the top of this post, one of the benefits for me and the University of Lincoln, was to see how effectively we could run a development project in this way. It’s likely that any development project I get involved in will be based on existing open source software that has an existing community of developers and users, just as WordPress does. I’m just not interested in re-inventing the wheel on something completely new. I prefer to contribute rather than invent. In this respect, I feel like it’s been a success. Within a few days, we’ll have the prototype we set out to create and within a week or two, it will be documented so that anyone should be able to pick JISCPress up and make it their own. JISC have seen the benefits of using it on three documents so far and in one shape or another, I think they’ll continue to use it. We’ve shared our ideas around the project widely, both on this blog and the other sites we use (see sidebar) and have discussed and demonstrated our work at two conferences. digress.it is now a well-regarded WordPress plugin with a growing number of users. Recently, Cornell University have employed Eddie to continue to develop digress.it for a project they are running with the Whitehouse. Through the JISCPress project, JISC have directly contributed to the work Cornell are doing. Likewise, the New York Public Library also want Eddie to develop digress.it for them, so I’m confident that this part of our work will be maintained and fed back into the core tree of the code for everyone to benefit from.

WriteToReply has benefited of course. We’ve been able to spend time thinking about and working on ’stuff’ which we intend to use on http://writetoreply.org Recently Eduserv offered to help us with hosting WriteToReply which has encouraged Tony and I to keep pursuing our interests in document publishing and public engagement on the web. I hope that we can work with Eduserv over the coming months and pass on what we know about WordPress hosting.

Finally, based on the work we’ve done on JISCPress with Triplify and WPMU, I plan to apply for funding from Talis to develop a ‘wordpress.com for OPACs’. It sounds ambitious, but a lot of the work has already been done on this and the Scriblio project and I think we can show it is a viable idea. You can read more about what we have in mind, here.

One of the things I’ve been doodling with but not managing to progress much thinking wise (not enough dog walking time lately!) is how we might be able to use the digress.it WordPress theme to support various course related functions in ways that exploit the disaggregating features of the theme.

Chatting with Huw Jones last week about his upcoming Arcadia seminar on “The Problem of Reading Lists” (this coming Tuesday, Nov 24th – all welcome;-) I started thinking again about the potential for using digress.it as a means of publishing, and collecting comments on, reading lists.

So for example, over on the doodlings WriteToReply site I’ve posted an example of how a reading list posted under the theme is automatically disaggregated into separate, uniquely identified references:

The reading list was generated simply by copying and pasting a PDF based reading list into a WordPress blog post. Looking at the format of the list, one could imagine adding further comments or notes relating to each reference using a blog comment. Given that the basis of each paragraph is a citation to a particular work, it might be possible to parse out enough information to generate a link to a search on the University OPAC for the corresponding work (and if so, pull back an indication of the availability of the book as, for example, my Library Traveler script used to do for books viewed on Amazon).

Under the current in-testing digress.it theme, each paragraph on the page can be made available as a separate item in an RSS feed; that is, as well as the standard ’single item’ RSS page feed that WordPress generates automatically, we can get an N-item feed from the page for the N-paragraphs contained on a page.

Which in terms means that to generate an itemised RSS feed version of a reading list, all I need to do is paste the reading list – with each reference in a separate paragraph – into a single blog post. (the same is true for disaggregating/feed itemising previous exam papers, for example, or I guess video links in order to generate a DeliTV programme bundle…?!)

(For more details of the various ways in which digress.it can automatically disaggregate/atomise a document, see Open Data: What Have We Got?.)

PS just a reminder again – Huw’s Reading List project talk, which is about far more than just reading lists, is on Tuesday in the Old Combination Room, Wolfson College, Cambridge, at 6pm.

describes a common approach to measuring website traffic [for central government]. This enables departments to answer Parliamentary Questions and Freedom of Information Requests about website usage consistently and reliably

I’ll also start to explore how to generate reports that satisfy those guidelines using Google Analytics.

The proposed metrics “are defined according to industry standards set by the Joint Industry Committee for Web Standards (JICWEBS)” and specify the following minimal level of reporting (Measuring Website Usage – Reporting requirements):

1. The following web metrics, as defined by the Joint Industry Committee for Web Standards (JICWEBS), must be measured for each and every publicly accessible website operated by an organisation:
   • Unique User/Browsers
   • Page Impressions
   • Visits
   • Visit Duration
2. Central government departments must measure Unique User/Browsers, Page Impressions, Visits and Visit Duration starting from 1 April 2009 for every website open on 1 April 2010.
3. Executive agencies and non-departmental public bodies (NDPBs) must measure Unique User/Browsers, Page Impressions, Visits and Visit Duration starting from 1 April 2010 for every website open on 1 April 2011.
4. The following information must be provided to COI at the end of each quarter:
   • Number of monthly Unique User/Browsers
   • Number of monthly Page Impressions
   • Number of monthly Visits
   • Number of Visits of at least two Page Impressions
   • Total time in seconds for all Visits of at least two Page Impressions
5. Each report should contain figures for each of the previous three months. This information should be provided in the format shown in the reporting template in Appendix A.
6. All figures should exclude internal web development activity, performance monitoring, automated broken link detection and other types of non-human activity (e.g. robots and spiders). Further details on what to exclude are found in the Page Impressions section.

So what does Google Analytics offer “out of the box”?

The Visitors Overview repeats these figures and additionally provides an indication of the number of ‘unique’ visitors:

At face value then, it would appear that the Google Analytics are providing at least some of the required stats (though we need to clarify that the numbers as recorded by Google Analytics conform to what the COI has in mind for those reports as described in their guidance on the Minimum standard for web metrics!) But what does that guidance relating to “at least two web pages” mean?

To understand the emphasis on “at least two pages”, it’s worth reflecting on the notion of bounces and the bounce rate. Bounce rate refers to the proportion of visitors to a site who only visit one page on a website before leaving that site, and as such tend to leave no meaningful analytics behind.

According to the ClickTale blog (What Google Analytics Can’t Tell You – Part 1), Google Analytics “has no way of knowing how long a bounced visitor, who only visits one page, spent on your website”. That is, it appears that the time spent looking at a page appears not to be based on the difference between the time when a page has fully loaded (and generated a trackable onload event) and its unload event; instead, it is calculated as the time between two loading one page and clicking through to and loading a second page on the sam site.

Which is why the emphasis on collecting stats from at last two pages: given the current crop of analytics tools that struggle to do anything meaningful with single page visits, specifying a two page visit means that not only visits to the site that are likely to be meaningful are reported, but also that the reports are more likely to contain meaningful data too. (There is an obvious problem here: if visitors visit two pages, and quickly click to the second from the first before exiting the site from the second page, the time spent on the second page won’t be captured? See for example Time on Site & Time on Page – Google Analytics metric mystery)

One of the nice things about Google Analytics is that it lets you create custom views, or “segments” of the data in which you can specify things such as the minimum number of pages visited when generating a particular report. In order to do this, you specify an “Advanced Segment”. Here’s what an Advanced Segment for a “minimum of two pages visited report” might look like:

Applying this segment to the same data charted above gives these results:

So for example, in this version of the report we see that the average number of page views and the average time on site has gone up.

Something I don’t think Google Analytics report is the total time on site. Bearing in mind the lack of data regarding the time spent on exit pages, the best we can do is multiply the number of visits by the average time on site to get an estimate of the total time on site.

With just this single advanced segment, a simple calculation, and the out of the can reports from Google Analytics, I think we can deliver on the suggested stats based on a literal reading of the headings, though in a follow up post I’ll check to see if the more detailed spec on the metrics matches the way that Google ANalytics defines its metrics.

PS Unfortunately, the segmented report appears to have lost the number of absolute unique visitors (although I think the recommended report wanted the number of uniques, including bounces, to the site?) Anyway, let’s play: the number of visits gives the upper bound on the number of unique visitors, but can we also estimate the lower bound? One heuristic might be to look at the number of visits and uniques in the original report (176 uniques, 245 visits), see how many visits were lost in discounting the bounces (245-104 = 141), assume these were all unique and subtract these from the original number of uniques (176-141=35). I think this gives the lower bound on uniques as recorded by Google Analytics for non-bouncing visitors?

Over the last couple of weeks, it seems as if the Goog has been doing a bit of reconciliation on the old analytics front, in particular the ability to track traffic driven back to your website from links contained within a feed published from that site using Feedburner…

The first thing I’d noticed as being different was the appearance Google Analytics tracking codes on Feedburner powered posts that I was reading in Google Reader – opening such a post in a new window seems to display it with a set full blown set of GA tracking attributes. So for example, opening a post from the Feedburnered OUsful.Info feed results in a URI like this:

http://ouseful.wordpress.com/2009/11/18/under-the-radar/?
utm_source=feedburner&utm_medium=feed
&utm_campaign=Feed%3A+ouseful+%28OUseful+Info%29&utm_content=Google+Reader

…and I’m pretty sure I didn’t put those tracking codes in there explicitly…

In “Campaign” Tracking With Google Analytics, I started sketching out how it might be possible to use Google Analytics campaign tracking codes to to track the spread of referrer links to documents or document fragments hosted on WriteToReply or JISCPress, so let’s see how the Feedburner annoations are structured:

• utm_source=feedburner (that is, the originator of the feed);
• utm_medium=feed (that is, the means by which the content was transported/syndicated);
• utm_campaign=Feed: ouseful (OUseful Info) (that is, the name of the Feedburner feed (I think: the feed URL is http://feedburner.com/ouseful), followed by the feed title (OUseful Info);
• utm_content=Google Reader (that is, the place where I viewed the link).

Compare this with the suggestion I made for annotating WriteToReply links:

• utm_source=twitter.com (that is, the place a link was ‘launched’);
• utm_medium=question (that is, the type of slug content used to qualify the link);
• utm_campaign=jiscri (that is, the consultation document linked to, e.g. for the link <em.http://writetoreply.org/jiscri/2009/03/11/rapid-innovation-projects/);
• utm_content=slug3 (that is, a unique ID to identify the text used to qualify the syndicated link).

So how can you get Googalytics tracking codes on your Feedburner feeds? Details are still sketchy, (e.g. see the original announcement on the Goole Analytics blog here: An Integration With Feedburner, and the Google AdSense for Feeds blog here: “Afternoon, Frank.” “Hey howdy, George.”) but this Google FAQ post on How do I set up my FeedBurner feed to report feed clicks in Google Analytics?:

If you use Google Analytics to track web site visitors, you can see feed clicks originating from your FeedBurner feed by activating an option on the Analyze tab.

When someone clicks one of your feed items and ends up back on your web site, Google Analytics will track that activity and include it in the “Traffic Sources” section.

The post also tells you where you can set up the tracking details – from the Configure Stats menu option. And selecting that, I can now see why my feed links are annotated as they are:

(I’m not sure how the $distributionEndpoint is treated for none Google properties?)

The Google AdSense for Feeds post suggests that:

By default, these analytics will show up in the “All Traffic Sources” and “Campaigns” views in Google Analytics. You can filter the results just to only the traffic that comes from Google FeedBurner by filtering on “feedburner” on the All Traffic Sources page or “Feed:” on the campaigns view. You can also use these sources in the Advanced Segments views.

which suggests that for sites like JISCPress/WriteToReply that use Google Analytics on the main site and Feedburner for the public/promoted feeds, the Feedburner integration will automatically annotate feed links with tracking codes that can be tracked from the site’s Google Analytics dashboard.

***
The website looks as though it has some good accessibility built in, but the
JavaScript it uses isn’t particularly accessible. To fully understand, it would be
worth getting the site audited against a recognised benchmark such as the Web Content
Accessibility Guidelines:

http://www.w3.org/TR/WCAG20/

The JavaScript doesn’t seem to be keyboard/screen reader friendly, particularly
on the comments pages. If the jQuery libraries have been used out of the box, it’s
likely they’ll need adapting.

Form fields will also need text labels associating with them. Guidance on
creating accessible forms can be found here:

http://www.accessify.com/features/tutorials/accessible-forms/

It’s great to see that ARIA landmarks have been used. If the main landmark could
be applied consistently across all pages, it would help people rely on it as a means
of navigation. Adding in more landmarks, particularly for search and navigation,
would also help.

IE6 and IE7 make up an enormous chunk of the browser market. Most IE6 users are
within the corporate and public sectors, where upgrading isn’t an option because of
organisational policy.

Representing the full visual design can sometimes be difficult for these
browsers, particularly IE6. In this case though, that doesn’t look as though it
should be the case.

The general building blocks of the website are good. The separation of
presentation from content has been done well, and the code is reasonably clean.
Headings, lists and other standard elements have all been used well.
***
My notes:

The digress.it theme is based on the default theme for WordPress (‘Kubrick’), which is considered a solid and accessible design which many themes are built on. digress.it has also been consciously developed to be relatively easily styled using CSS.

Accessibility has been a constant, though admittedly secondary, requirement in the JISCPress project and Eddie has made specific efforts to improve the accessibility of the plugin over the original CommentPress. I believe digress.it is partially using the ‘accessible jquery’ library, too. I’ll be looking at the WCAG2 document and reviewing, as best as I can, the areas of improvement that can still be made.

Anyway, what I would have shown was how each document site on jiscpress.org uses Triplify to provide Linked Data in the form of RDF/N3 triples, which we store on the Talis Platform using a plugin Alex wrote.

Using Alex’s config file for WordPress MultiUser, we drop the triplify directory into the WPMU root directory, alongside wp-admin, wp-includes and all the other WordPress files. You should take a look at the config file and make sure it’s doing what you want it to do, but it will work as it is.  With this in place, Linked Data in the form of an RDF flat file for each document site (blog) is available at http://document.jiscpress.org/triplify or http://jiscpress.org/document/?triplify

(I should warn you that none of the URLs in this post are genuine URLs. They’re examples of syntax. The server at jiscpress.org will stop running at the end of December).

Now, to get that same data onto the Talis Platform, Alex has written a plugin for WPMU that periodically crawls the documents for changes and pushes the new data to a Talis Platform account.  Here are the WPMU site-wide admin options:

and here are the per document site user settings:

I won’t explain what the plugin does in detail. Just click on those images above and you’ll see the options that are available and if you’re reading this stuff, you know what it’s all about.  The Talis/Triplify plugin for WPMU will appear on  http://wordpress.org/extend/plugins in the next couple of weeks. It’s been tested and it does what we expect it to do but we want to test it more on sub-directory installs before it’s publicly available. Full documentation will appear soon on http://code.google.com/p/jiscpress/wiki/Documentation

We have also developed a WPMU plugin for Open Calais and the Yahoo! Term Extraction API. This provides a background service which indexes each document section (blog post) and creates relationships between content across the platform. We’ll post here about that very soon.

In addition to the Linked Data, JISCPress, using digress.it on WordPress, provides a long list of other open data (not Linked Data) end-points which might be put to good use. Here you go..

Document paragraphs

These are switches that provide individual paragraph data in different formats.

http://test.jiscpress.org/?p=15&digressit-embed=1&format=xml

http://test.jiscpress.org/?p=15&digressit-embed=1&format=text

http://test.jiscpress.org/?p=15&digressit-embed=1&format=rss

http://test.jiscpress.org/?p=15&digressit-embed=1&format=html

http://test.jiscpress.org/?p=15&digressit-embed=1&format=json

Document sections

This is just the regular WordPress post content in RSS format. In JISCPress terms, it’s the document section which is a single feed item.

http://test.jiscpress.org/2009/07/28/6-how-jisc-invests/feed/?withoutcomments=1

and this is the normal WordPress feed of comments on a particular post/document section.

http://test.jiscpress.org/2009/07/28/6-how-jisc-invests/feed/

We’ve also added the provision of a feed for each document section (‘post’), where each paragraph is a feed item. Note that this makes digress.it a nice tool for building your own feeds out of a single WordPress post.

http://test.jiscpress.org/feed/paragraphlevel/3-jisc-vision-mission-and-objectives/

Per paragraph comments/discussions

For each paragraph, there’s a feed of the comments/discussion.

http://test.jiscpress.org/feed/paragraphcomments/3-jisc-vision-mission-and-objectives,1

Commenter feeds

For each person that comments, there’s a feed of their comments

http://test.jiscpress.org/feed/usercomments/Joss%20Winn

All the other stuff

Don’t forget that the entire document content is also available as a feed

http://test.jiscpress.org/feed/

http://test.jiscpress.org/feed/rss

http://test.jiscpress.org/feed/rss2

http://test.jiscpress.org/feed/atom

http://test.jiscpress.org/feed/rdf

as are all comments from the site, too:

http://test.jiscpress.org/comments/feed

with WordPress, tags also have feeds

http://test.jiscpress.org/tag/tag1/feed

and so do categories

http://test.jiscpress.org/category/category1/feed

You can also combine tags

http://test.jiscpress.org/tag/tag1+tag2+tag3/feed

and you can combine tags and categories

http://test.jiscpress.org/?category_name=category1&tag=tag2,tag3&feed=rss2

Finally, authors have a feed, too

http://test.jiscpress.org/author/joss/feed/

Summary

WordPress is a versatile CMS for organising/designing and publishing data as feeds and therefore a useful source of Open Data. JISCPress has extended this versatility by choosing to develop further data end points using digress.it and offering a simple way of publishing Linked Data to the Talis Platform RDF triple store where is can be queried and mashed up using the platform’s API.

]]> http://jiscpress.blogs.lincoln.ac.uk/2009/11/18/open-data-what-have-we-got/feed/ 2 Admin settings User settings http://jiscpress.blogs.lincoln.ac.uk/2009/11/18/digress-it-version-2-3-was-released-last/ http://jiscpress.blogs.lincoln.ac.uk/2009/11/18/digress-it-version-2-3-was-released-last/#comments Wed, 18 Nov 2009 09:06:38 +0000 Joss Winn http://jiscpress.blogs.lincoln.ac.uk/2009/11/18/digress-it-version-2-3-was-released-last/ digress.it version 2.3 was released last night and this marks the last major release of this WordPress plugin funded within the time frame of the JISCPress project. It is worth pointing out that our project funding effectively boot strapped the re-birth of CommentPress and paid for Eddie Tejeda, the original CommentPress developer, to rewrite CommentPress from scratch into digress.it. I was recently told that this work has led to Eddie being asked by Cornell University to work on a really interesting and high-profile digress.it-based project for them which we’ll be announcing soon. It’s great to see JISC’s work sustained in this way and hear that digress.it will be properly maintained through additional funding.

This release brings better IE6 & 7 compatibility, a smoother, better Comment Box, a document section level comment view, an option to parse lists into separately commentable points, BuddyPress compatibility, document section level feeds and a bunch of bug fixes. Overall, it feels like stable, feature rich code.

As noted above, we added one more RSS feature which now means digress.it can be used as an RSS feed builder. Each paragraph in any given blog post/document section, can be extracted as an RSS feed ‘item’. See http://writetoreply.org/jiscstrategyreview/feed/paragraphlevel/8-measuring-success/ for an example (and note the /feed/paragraphlevel/post_slug/ syntax used!)

I’ll be writing more in the next day or so about all the other ‘open data’ end points that we’ve developed during the JISCPress project.

If a link to a JISCPress document appears on a third party webpage, and somebody clicks on that link and then lands on the corresponding JISCPress page, Google Analytics will capture where that incoming visitor cam from via the Referring Sites report. At the top level this is organised by domain:

We can then tunnel down to the page level:

This is all well and good, but sometime we also might want to know where the person who posted the referring link on their web page got hold of it. Did they capture it from a tweet, for example, or via an email list? When we releas a URI into the wild via some sort of marketing campaign, what sort of life does that URI have, and where will it end up sending traffic back from?

In the Googe Analytics FAQ answer How do I tag my links?, a method is described for adding additional tags to a referral URL (that is, a URL that you publish and/or distribute more widely that refers back to your website) that Google Analytics can use to segment traffic referred from that URL. Five tags are available (as described in Understanding campaign variables: The five dimensions of campaign tracking):

Source: Every referral to a web site has an origin, or source. Examples of sources are the Google search engine, the AOL search engine, the name of a newsletter, or the name of a referring web site.
Medium: The medium helps to qualify the source; together, the source and medium provide specific information about the origin of a referral. For example, in the case of a Google search engine source, the medium might be “cost-per-click”, indicating a sponsored link for which the advertiser paid, or “organic”, indicating a link in the unpaid search engine results. In the case of a newsletter source, examples of medium include “email” and “print”.
Term: The term or keyword is the word or phrase that a user types into a search engine.
Content: The content dimension describes the version of an advertisement on which a visitor clicked. It is used in content-targeted advertising and Content (A/B) Testing to determine which version of an advertisement is most effective at attracting profitable leads.
Campaign: The campaign dimension differentiates product promotions such as “Spring Ski Sale” or slogan campaigns such as “Get Fit For Summer”.

(For an alternative description, see Google Analytics Campaign Tracking Pt. 1: Link Tagging.)

The recommendation is that campaign source, campaign medium, and campaign name should always be used.

Elsewhere, (Library Analytics (Part 7), from which elements of this post have been taken), I considered how these codes might be used to track course referrals to Library resources from a VLE (something I need to revisit, now I’ve had a little more time to consider the possible role(s) of these tracking codes). But it also seems to me to be reasonable to raise a few questions about how we might use these tracking codes in the context of a document on JISCPress or WriteToReply in order to track referrals back to the site from social media campaigns highlighting a particular document or section of a document.

So, what are sensible mappings/interpretations for the campaign variables? Remember, these tracking variables are parameters that we might add to a link that we have posted somewherethat is intended to drive traffic back to the site. The tracking variables are there to allow us to see how different links are performing. Thinking about how we might use these five tracking dimensions, whether or not we use them in the “intended” Google Analytics way, may also provide us with some ideas about how to use links to drive traffic back to our site.

To try and ground the exercise, consider this example: a new document is published on JISCPress and we want to compare how well links posted on Facebook compare with links posted on Twitter for driving traffic back. For tracking to be most effective, we hope that if a link is rebroadcast or shared, the tracking variables are carried along with it. This means that if a link is posted to Twitter, that gets shared onto Facebook and onto a blog, we can look at the traffic that comes back, and from where (via the Referral tracking described at the start of this post), for each of the separately released URIs. A second example might relate to a campaign intended to drive traffic back to a particular section or paragraph of a document. This campaign might involve publishing a link back to the same paragraph in a series of separate posts or status updates, each with a different slug or call to action message. That is, each link+message may be published in the same place (and hence have the same referrer information), but at different times and with different link text, or contextual information. A third example might be where there is more than on link back to the same document on a web page, and we want to track how effective each link is compared to the others?

Here are the supported variables again:

• source: the obvious thing to use this variable for is the domain or URI of the page where the link is published to. So if we tweet a link, twitter.com might be sensible. If we blog it, actually might be best?
• medium: this is intended to refer to the sort of link that has generated the traffic, such as a banner ad. In our case, we might clarify the intent with which the link was posted, such as announcement, or question;
• term: this is an optional parameter, and I’m not sure how it should be used or whether it conflicts with other Google services. If we post something with a hashtag on twitter, or a st of tags on delicious, might we use those tags are terms?
• content The second optional variable, this is often usd to discern A/B test ads. If we tweet the same link with different call to action/prompting questions, maybe this differential content should be uniquely identified with the content field?
• campaign: typically used for tracking a promotion or campaign, this field might be used to identify a different document when, for example, a link to the top level JISCPress is referred to in a announcement about a particular document?

So for example, we might have something like:
http://writetoreply.org/?utm_campaign=ukgovurisets &utm_medium=announcement&utm_source=actually
appearing as the link for WriteToReply in an announcment about the hosting of the UK Government URI Sets document.

Or maybe a call to action on twitter relating to a particular part of a document:
What benefits would you like to see from #JISCRI calls? http://writetoreply.org/jiscri/2009/03/11/rapid-innovation-projects/#3?utm_campaign=jiscri &utm_medium=question&term=JISCRI&utm_source=twitter.com&utm_content=slug3

To support the generation of tracking URIs, a URL Generator Tool (like the official Tool: URL Builder) that will accept a tweet, for example, along with a JISCPress/WriteToReply URL and then automatically create tracking variable values might be worth considering?

As we come to the final month of the JISCPress project, we had some great news over on WriteToReply last week where we were able to announce that Eduserv would be covering our hosting costs for the immediate future (Eduserv funds hosting for WriteToReply, eFoundations: Write To Reply).

So what exactly does the platform we’ve been working on have to offer? Here’s one of the ways I think of it…

A document publishing platform that automatically atomises documents to the paragraph level, allows aggregated commenting at the paragraph and ‘user’ level, and supports the republication and re-presentation of documents in a variety of standard formats at the document level.

The first part of the process is the (manual assisted) ingress stage, in which documents are imported into the WordPress environment such that each substantive document section ideally maps onto a single WordPress “blog post”:

An RSS for the document as a whole, with one item per section, is generated automatically by the WordPress platform. A single item RSS feed is also generated for each page (so the content of each page can be easily transported around the web).

The second part of the process is the atomisation of each post, carried out automatically by the Digress.It theme, in which each paragraph in the document is given its own unique URI, derived from the URI of the web page (“blog post”) the paragraph appears on:

Potentially, an RSS feed can also be produced for each page in which each paragraph is a separate feed item, thus allowing a page/section to be transported around the web via a single feed, but in atomised form.

The paragraph level chunks produced by the atomistation process can be transcluded as independent elements in independent web documents in other documents by a variety of means (as an embeddable object, via XML, txt, JSON, etc):

The default nature of the WordPress platform allows comments to be made at the level of each web page, with an RSS feed of comments for each page being published ‘for free’. JISCPress extends this functionality by allowing comments to be associated with discrete paragraphs. Views over the comments are also available at the user level, (that is, grouped according to the user who made the comments, wheresoever they are made in the document). An additional RSS fed of comments by user is also available, which means that a document on the platform can actually be used as a scaffold for a critical response to the document by a particular user.

A further level of innovation is based on the automated generation of ’semantic tags’ at the page level. Once generated, tag based collections of posts can be syndicated in the normal way via WordPress generated tag based RSS feeds:

JISCPress also benefits from the Trackback mechanism implemented by WordPress. When a page or paragraph URI is linked to from a third party web page, a trackback to the originating page may be captured, which we interpret as the automated capture of links remote annotations or comments about the document.

When considered in these terms, the JISCPress/WriteToReply platform is seen to provide a powerful means of publishing documents in which individual sections may carry their own unique URI, and individual paragraphs within a section also contain their own unique URI (which in many situations may be rooted on the section URI).

The platform can also be regarded as republishing – or re-presenting – each section (i.e. page) and each paragraph as an independent entity. That is, whenever a document is published via the platform, each separate paragraph may also be thought of as being independently published “for free”, in the sense that:

- each paragraph is independently addressable,
- each paragraph is independently commentable, and
- each paragraph is independently republishable/syndicatable.

So, given that, can you think of any ways in which the JISCPress/WriteToReply platform can support your document publishing and comment gathering strategy?

As Joss announced on the JISCPress blog, We’ve got paragraph data output switches! that expose paragraph level content through a unique URI in a variety of formats (xml, txt, html, rss and json), as well as object embed codes for each paragraph, though I’m not sure if this is going to be maintained…? e..g at the moment, I think we’re trialling literal text blockquote embeds:

(If the object embed does disappear, similar functionality could be achieved using the JSON feed and a Javascript function, though I guess we need JSON-P (i.e. support for something like &callback=foo to make that really easy.)

See also: A Quick Update for a review of the latest feature releases within the digress.it theme we’re using.

To demonstrate one possible use case for object embedding, see the post Engaging With the Issues Raised By The Google Book Settlement which includes three embedded paragraphs from the JISC’s current consultation around the Google books settlement.

Here’s the actual HTML:

Note that currently there is an issue with sizing the embed container (can any CSS gurus out there give us a fix?

Ideally we need to identify the container height and then size it automatically so there are no scrollbars? I’m guessing .scrollHeight might have a role to play in autodetecting this?)

One thing you might notice is that the URIs for the embedded consultation questions follow a similar pattern – only the paragraph number identifier changes:
http://writetoreply.org/googlebooks?p=8&digressit-embed=4

What this means is that we should be able to pull in a random paragraph by constructing a URI with a randomly generated paragraph number. So for example:

If you reload the page, you have an 80% chance of seeing a different question…

Here’s the Javascript snippet:

var n=2+Math.floor(Math.random()*5);
var o=document.createElement('object');
o.setAttribute('style','width: 100%; height:70px;');
o.setAttribute('id','61c197964762012d4819093ebeee4fcf');
var p='http://writetoreply.org/googlebooks?p=8&digressit-embed='+n;
p=p.replace(/#038;/,''); //get round Wordpress escaping everything...
o.setAttribute('data',p);
document.getElementById('wtr_embed').appendChild(o);

//There’s a div with an appropriate id attribute (’wtr_embed’) also added to the page…
//Note that the div needs to be placed before any inline Javascript in the page;-)

I’m not sure yet if we can track the use of embeds (certainly server logs should be able to track calls, but these probably can’t be captured using Google Analytics?), but it’s still early days…