//We finished JISCPress. If you’re interested, I’ve written a long overview of the work we’ve done with WPMU as a document discussion platform, based on WriteToReply. You’ll see that the project has, among other things, produced three plugins: digress.it, and two Linked Data plugins that run as background services across the platform, create relationships between documents and document sections and post RDF to the Talis Data Store. Fancy!
//As is often the case, I struggle at first glance to see the full implications of a new development in technology, which is why I so often rely on others to kick me up the arse before I get it. ((I am not ashamed to admit that I’m finding that my career is increasingly influenced by following the observations of Tony Hirst. Some people are so-called ‘thought-leaders’. I am not one of them and that is fine by me. I was talking to Richard Davis about this recently and, in mutual agreement, he quoted Mario Vargas Llosa, who wrote: “There are men whose only mission is to serve as intermediaries to others; one crosses them like bridges, and one goes further.” That’ll do me.))
I first read about web hooks while looking at WordPress, XMPP and FriendFeed’s SUP and then again when writing about PubSubHubbub. Since then, Dave Winer’s RSSCloud has come along, too, so there’s now plenty of healthy competition in the world of real time web and WordPress is, predictably, a mainstream testing ground for all of it. Before I go on to clarify my understanding of the implications of web hooks+WordPress, I should note that my main interest here is not web hooks nor specifically the real time web, which is interesting but realistically, not something I’m going to pursue with fervour. My main interest is that WordPress is an interesting and opportunistic technology platform for users, administrators and developers, alike. Whoever you are, if you want to understand how the web works and how innovations become mainstream, WordPress provides a decent space for exercising that interest. I find it increasingly irritating to explain WordPress in terms of ‘blogging’. I’ve very little interest in WordPress as a blog. I tend to treat WordPress as I did Linux, ten years ago. Learning about GNU/Linux is a fascinating, addictive and engaging way to learn about Operating Systems and the role of server technology in the world we live in. Similarly, I have found that learning about WordPress and, perhaps more significantly, the ecosystem of plugins and themes ((Note that themes are not necessarily a superficial makeover of a WordPress site. Like plugins, they have access to a rich and extensible set of functions.)) is instructive in learning about the technologies of the web. I encourage anyone with an interest, to sign up to a cheap shared host such as Dreamhost, and use their one-click WordPress offering to set up your playground for learning about the web. The cost of a domain name and self-hosting WordPress need not exceed $9 or £7/month. ((I am thinking of taking the idea of WordPress as a window on web technology further and am tentatively planning on designing such a course with online journalism lecturer, Bernie Russell. It would be a boot camp for professional journalists wanting (needing…?) to understand the web as a public space and we would start with and keep returning to WordPress as a mainstream expression of various web technologies and standards.))
Within about 15 minutes of Tony tweeting about HookPress, I had watched the video, installed the plugin and sent a realtime tweet using web hooks from WordPress.
It’s pretty easy to get to grips with and if a repository of web hook scripts develops, even the non-programmers like me could make greater use of what web hooks offer.
Web hooks are user-defined callbacks over HTTP. They’re intended to, in a sense, “jailbreak” our web applications to become more extensible, customizable, and ultimately more useful. Conceptually, web applications only have a request-based “input” mechanism: web APIs. They lack an event-based output mechanism, and this is the role of web hooks. People talk about Unix pipes for the web, but they forget: pipes are based on standard input and standard output. Feeds are not a sufficient form of output for this, which is partly why Yahoo Pipes was not the game changer some people expected. Instead, we need adoption of a simple, real-time, event-driven mechanism, and web hooks seem to be the answer. Web hooks are bringing a new level of event-based programming to the web.
I think the use of the term ‘jailbreak’ is useful in understanding what HookPress brings to the WordPress ecosystem. WordPress is an application written in PHP and if you wish to develop a plugin or theme for WordPress you are required to use the PHP programming language. No bad thing but the HookPress plugin ‘jailbreaks’ the requirement to work with WordPress in PHP by turning WordPress’ hooks (‘actions’ and ‘filters’) into web hooks.
WordPress actions and filters, are basically inbuilt features that allow developers to ‘hook’ into WordPress with their plugins and themes. Here’s the official definition:
Hooks are provided by WordPress to allow your plugin to ‘hook into’ the rest of WordPress; that is, to call functions in your plugin at specific times, and thereby set your plugin in motion. There are two kinds of hooks:
- Actions: Actions are the hooks that the WordPress core launches at specific points during execution, or when specific events occur. Your plugin can specify that one or more of its PHP functions are executed at these points, using the Action API.
- Filters: Filters are the hooks that WordPress launches to modify text of various types before adding it to the database or sending it to the browser screen. Your plugin can specify that one or more of its PHP functions is executed to modify specific types of text at these times, using the Filter API.
So, if I understand all this correctly, what HookPress does is turn WordPress hooks into web hooks which post the output of the executed actions or filters to scripts written in other languages such as Python, Perl, Ruby and Javascript (they can be written in PHP, too) hosted elsewhere on the web. In the example given in the HookPress video, the WordPress output of the action, ‘publish_post‘, along with two variables ‘post_title’ and ‘post_url’, was posted to a script hosted on scriptlets.org, which performs the event of sending a tweet which includes the title and URL of the WordPress post that has just been published. All this happens as fast as the component parts of the web allows, i.e. in ‘real time’.
In other words, what is happening is that WordPress is posting data to a URL, where lies a script, which takes that data and creates an event which notifies another application. Because the scripts can be hosted elsewhere, on large cloud platforms such as Google’s AppEngine, the burden of processing events can be passed off to somewhere else. I see now, why web hooks are likened to Unix pipes, in that the “output of each process feeds directly as input to the next one” and so on. In the case of HookPress, the output of the ‘publish_post’ hook feeds directly as input to the scriptlet and the output of that feeds directly as input to the Twitter API which outputs to the twitter client.
Besides creating notifications from WordPress actions, the other thing that HookPress does (still with me on this ‘learning journey’ ??? I’ve been reading, writing and revising this blog post for hours now…), is extend the functionality of WordPress through the use of WordPress filters. Remember that filters in WordPress, modify text before sending it to the database and/or displaying it on your computer screen. The example in the video, shows the web hook simply reversing the text before it is rendered on the screen. ‘This is a test’ becomes ‘tset a si sihT’.
The output of the ‘the_content‘ filter has been posted to the web hook, which has reversed the order of the blog post content and returned it back to WordPress which renders the modified blog post.
Whereas the action web hooks are about providing event-driven notifications, the filter web hooks allow developers to extend the functionality of WordPress itself in PHP and other scripting languages. In both cases, web hooks ‘jailbreak’ WordPress by turning it into a single process in a series of piped processes where web hooks create, modify and distribute data.
Finally, I’ll leave you with this presentation, which is all about web hooks.
In the presentation, there are two quotes which I found useful. One from Wikipedia which kind of summarises what HookPress is doing to WordPress:
“In computer programming, hooking is a technique used to alter or augment the behaviour of [a programme], often without having access to its source code.”
and another from Marc Prensky, which relates back to my point about using WordPress as a way to learn about web technologies in a broader sense. WordPress+HookPress is where programming for WordPress leaves the back room:
As programming becomes more important, it will leave the back room and become a key skill and attribute of our top intellectual and social classes, just as reading and writing did in the past.
//Working on the JISCPress project, I’ve been thinking quite a lot about scholarly publishing on the web, and in particular with WordPress. This morning, I read a post over on the ArchivePress blog about some WordPress plugins which are useful additions for creating a scholarly blog and it got me thinking a bit more about what features WordPress would need to support scholarly publishing.
JISCPress does away with the idea that WordPress is a blogging tool, and instead uses WordPress Multi-User as a document publishing platform, where one site or ‘blog’ is a document. The way WPMU is structured means that despite serving multiple (potentially millions) of document sites, the platform remains relatively ‘lightweight’ as each document site generates just a handful of additional database tables, while sharing the same administrative core as a single WordPress install. So, 100 WordPress blogs on WPMU is nothing like the equivalent of running 100 separate WordPress blogs, both from the point of resource requirements and administration. In fact, quite soon, there will be no such thing as WPMU as the two products are going to be merged and because they share 90%+ of the same code already, it’s not too difficult to achieve. ((Has anyone done a diff on the two code bases to measure exactly what percentage of the code is shared between WP and WPMU?))
Anyway, my point here is to discuss whether WordPress can be extended to accommodate most conventions found in scholarly publishing and where it is lacking, to identify the development work required to meet the needs of most academic who wish to write on and publish to the web. ((Actually, I think I’ll save the discussion of its shortfalls for my next post. This one is already long enough.))
Scholarly publishing extends to a wide variety of published outputs. As a Content Management System (CMS) and technology development platform, I believe that WordPress has the potential to support any type of scholarly publishing that the web supports. It is extremely extensible, as can be seen from the 6000+ plugins that are available. However, what I’m interested in is what can be done now, by an academic wishing to publish their work through the use of WordPress acting as a CMS. What can be achieved with a few quid ((I pay $5/year for my domain name and as many sub-domains as I need. I pay $10/month for my hosting with unlimited storage and bandwidth.)) to self-host WordPress so that a few plugins can be installed and a well structured, typical, scholarly paper can be published.
For some time, I’ve been meaning to publish my MA dissertation. Back in 2002, I undertook some unique research which has not, to my knowledge, been repeated and I think there is some value in having it easily accessible on the web. I have an OpenOffice file and a PDF and, in the course of a morning, have published it under my own domain. The reason I did not publish it on the university WPMU platform is because I have been experimenting with different plugins and did not want to install plugins that were untested or we may not support long-term. In this case, I’ve used a single WordPress installation, but ideally an individual researcher, group of researchers or research institution, would run a WPMU installation which allowed multiple documents to be authored individually or collaboratively ((Like any decent CMS, WordPress supports role-based authoring and editing and maintains a revision history of edits, auto-saved once per minute. Revisions can be compared alongside of each other.)) and published directly to the web as XHTML.
BuddyPress, by the way, can make the experience even more natural, not only because it is based around a community of like-minded people writing together on the same web publishing platform, but also because, with a few tweaks here and there, we can move away from the language of blogs and towards the language of documents.
Enough of BuddyPress on WPMU for now and back to my dissertation. I set up the site in ten minutes, without using FTP or a command line because I use a host that provides a one-click install of WordPress and WordPress allows you to search for and install plugins from its Dashboard, rather than having to use FTP. Once the site was installed, I then made some basic changes to the settings, turning on XML-RPC and AtomPub, so that, if I decided to, I could publish to the site using my Word Processor. ((On a scholarly WPMU installation, plugins could be pre-installed and activated, a default theme selected and settings tweaked so very little work is required by the academic author prior to writing her document.)) I didn’t use this in the end, but trust me, it works very well using recent versions of MS Word, Open Office (free) and other blogging clients such as MS Live Writer (free).
So, what are the common characteristics of an academic paper? What does WordPress have to support to provide functionality that meets most scholars’ publishing requirements? I scratched my head (and asked on Twitter) and came up with the following:
Many of these are supported in WordPress by default and don’t require any additional plugins (tables, images, sub-headings, annexes, appendices, dedication, abstract, introduction, exposition, conclusion, are all either basic literary conventions or just part of a simply structured document).
For additional support, I installed digress.it, which we have funded through the JISCPress project. This is a WordPress plugin which allows readers to comment on the paragraphs of a document, rather than at the document section level. We’re adding a lot more functionality to meet the objectives of the JISCPress project, but I chose digress.it, principally for the reason that it is designed to turn a WordPress blog into a document site. I could have used any other WordPress theme, but digress.it automatically creates a Table of Contents and allows you to re-order WordPress posts when they are read so that you don’t have to author your document in reverse or adjust the publication dates so the document sections appear in the correct order.
I added the abstract for my dissertation to the ‘about’ page, so it shows up on the front of the site. I also uploaded a PDF version so that people can download it directly. You’ll see that I also added some links to a related book and DVD, which will certainly appeal to people who are interested in my dissertation. The links pull an image and some basic metadata from Amazon, using the Amazon Machine Tags plugin. This could be used to link to the book in which your article is published and earn you money in click referrals. An alternative, would be the Open Book Book Data plugin, which retrieves a book cover and metadata from Open Library, where your book may already be catalogued. If it’s not on Open Library, catalogue it!
After setting this up, I installed a few more plugins:
Dublin Core for WordPress: Automatically adds ten Dublin Core metadata elements to the document mark up.
wp-footnotes: This allows you to easily add footnotes to your document by enclosing your footnote in double parentheses. ((I am using the plugin on this blog!))
OAI-ORE Resource Map: Automatically marks up the document sections with a OAI-ORE 1.0 resource map.
Google Analyticator: Adds Google Analytics support so you can collect statistics on the readership of your document.
WP Calais Archive Tagger: Analyses your entire document and automatically keywords each section, using the Open Calais API.
Search API: WordPress comes with search built in, but there is a new search API which will eventually make its way into the WordPress core. I’ve installed the plugin to provide full-text search across the document. It can also add Google Search to your document site.
wp-super-cache: This is simple to install and will significantly speed up your document site, making it a pleasure to navigate through and read 🙂
wp-latex: Although I didn’t need it for my dissertation, it’s worth noting that WordPress supports the use of 
Academic Citation: You need to add a line of code to your theme for this to display. It supports the concept of an article being a single blog post, rather than a ‘document site’ and displays a variety of citation formats for readers to use.
Do you know of any other plugins for a scholarly blog?
The other useful thing about managing a document using WordPress and in particular, using digress.it, is that you automatically get RSS/Atom feeds for the document. I’ve already discussed these in detail. It means that I was able to read my document in my feed reader, with footnotes and images displayed correctly.
See how nicely the formatting is preserved.
You’ll see that the document sections are listed in order; that is, first section on top. As I noted above, blogs list posts in reverse (most recent first), so I sorted the feed items in Yahoo Pipes and sorted it in ascending order. Yahoo Pipes exports as RSS and it’s that feed that I subscribed to in Google Reader. Wouldn’t it be nice, if I could import my document feed into an Institutional Repository? Wait a minute, I can! 🙂
When importing the default feed, the HTML output is accurate but in reverse order, while the RSS output from Yahoo Pipes didn’t import into EPrints very cleanly at all. I’ll work on this. UPDATE: Forget Yahoo Pipes. WordPress feeds can be sorted with a switch added to the URL: http://example.com/feed/?orderby=post_date&order=ASC
So there it is. An academic paper, published to the web using a modern CMS which supports most authoring and publishing requirements. I would favour an institutional WPMU platform for academics to author directly to, publish their pre-print to the web for open access and detailed comment, and import their RSS feed into the repository. As a proof of concept, I’m quite pleased with this. We are currently developing a widget that can be embedded in a web page or WordPress sidebar and allow a member of staff to upload a document or zipped folder of documents to the Institutional Repository. I wonder if we can also support the import of a feed from the widget, too?
So, what would your requirements be? Tell me and I’ll do my best to test WordPress against them.
//I’m very pleased to announce that my bid with Tony Hirst at the Open University, to develop a community platform for the JISC funding call process based on WriteToReply, was successful. The original bid document is publicly available and currently offers the most information on this six month, £32,500 project.
Note that this is an open project using open source software and we welcome volunteer contributions from anyone. I’ve set up a project blog, mailing list, wiki and code repository. Feel free to join us if this WriteToReply spin-off appeals to you. If you know anyone that might be interested, please do let them know.
If you’ve been following WriteToReply, you’ll know that we use WordPress Multi-User and CommentPress. Eddie Tejeda, the developer of CommentPress will be working with us on the project and this will result in significant further development of CommentPress 2. So, if you’re interested in CommentPress (as many people are), please consider following, contributing to and testing JISCPress.
I should also note that while the project is a spin-off of our work on WriteToReply, neither Tony or I are personally receiving any funds from JISC. The contributions from JISC to cover our time on this project are paid directly to our employers and does not result in any financial benefit to us or WriteToReply (which is in the process of being formalised as a non-profit business). In other words, while WriteToReply is a personal project, JISCPress is part of our normal work as employees of our universities (both Tony and I are expected to bid and win project funds – you get used to it after a while!). Money has been allocated to fund dedicated developer time to the project, which will pay Eddie and Alex, a student at the University of Lincoln, for their work.
Anyway, on with the project! Here’s the outline from the bid document:
This project will deliver a demonstrator prototype publishing platform for the JISC funding call and dissemination process. It will seek to show how WordPress Multi-User (WPMU) can be used as an effective document authoring, publishing, discussion and syndication platform for JISC’s funding calls and final project reports, and demonstrate how the cumulative effect of publishing this way will lead to an improved platform for the discovery and dissemination of grant-related information and project outputs. In so doing, we hope to provide a means by which JISC project investigators can more effectively discover, and hence build on, related JISC projects. In general, the project will seek to promote openness and collaboration from the point of bid announcements onwards.
The proposed platform is inspired and informed by WriteToReply, a service developed by the principle project staff (Joss Winn and Tony Hirst) in Spring 2009 which re-publishes consultation documents for public comment and allows anyone to re-publish a document for comment by their target community. In our view, this model of publishing meets many of the intended benefits and deliverables of the Rapid Innovation call and Information Environment Programme. The project will exploit well understood and popular open source technologies to implement an alternative infrastructure that enables new processes of funding-related content creation, improves communication around funding calls and enables web-centric methods of dissemination and content re-use. The platform will be extensible and could therefore be the object of further future development by the HE developer community through the creation of plugins that provide desired functionality in the future.
//I’ve just re-submitted this proposal for a demonstration at ALT-C 2009. It’s called WordPress Multi-User: BuddyPress and Beyond. It won’t be confirmed until June, but for the record, here it is…
‘BuddyPress’ is a new social networking layer for WordPress Multi-User blogs. It provides familiar, easy to use social networking features in addition to a high-quality and popular blogging platform. The University of Lincoln have been trialing WordPress MU since May 2008 and have been using BuddyPress since February 2009 to promote an institutional social networking community built around personalised and collaborative web publishing.
This session will demonstrate the versatility of the WordPress MU platform. We’ll look at an installation that is enhanced with BuddyPress, LDAP authentication, mobile phone support and advanced privacy controls. You’ll see how simple it is to set up site-wide RSS syndication and aggregation, enhance your blog with semantic web tools, publish mathematical formulae with LaTeX, send realtime notifications to Facebook, Twitter and IM, publish podcasts to iTunes, and embed GPX and KML mapping files. We’ll also look at how to embed WordPress content in your VLE and other institutional websites. The use of a temporary ‘ALT-C 2009 BuddyPress’ installation will be encouraged.
There will be opportunities throughout for questions and answers and participants will leave with a good understanding of the advantages and disadvantages of WordPress and the resources and skills required to provide a social networking and blogging platform in your institution.
//In February, I wrote a brief post about setting up BuddyPress with LDAP authentication within a university context (you’re looking at it). Four days ago, BuddyPress reached maturity by hitting version 1.0, marking a time to reflect on what I’d like to see developed for BuddyPress for use within a university context. This is an initial wish list. I’m not looking for BuddyPress to be an all singing, all dancing, social network. I don’t care about image collections and status updates (Flickr and Twitter do those jobs nicely) I would, however, like to see it being used for building group identity (projects, special interest groups, classes, courses) and portfolio/resume building. Right now, it’s pretty limited in those areas.
As I mentioned previously, our social network is private, while the blogs have five levels of optional privacy controls, ranging from public and indexed by Google, to private, single-user blogs. However, privacy within the social network is currently all or nothing. It’s a hack that works but has no flexibility. The BuddyPress activity plugin is currently turned off because the privacy plugin I use, doesn’t account for the feeds that the plugin exposes. It would be nice to be able to have the site-wide social activity visible when logged in. Currently, only information about new blog posts is published site-wide. What I would like is for everything that the activity plugin logs, to have site admin options to be 1) visible to non-logged in users/public; 2) visible to logged in users; 3) visible to my groups and friends,4) visible to my friends and 5) not visible. In addition, the feeds that are exposed of site-wide activity and member activity, could also be configurable so that 1) a site admin can choose to expose them or not; 2) if allowed, a member can choose to expose their personal feed or not; 3) a feed key could be used in place of the normal feed URI so that private member feeds could be created. Finally, groups and member profiles could optionally be made public or private. So anything following /groups/ or /members/ has an option to be visible outside the community.
Currently, groups don’t publish very much information and you can’t aggregate information from elsewhere into a group profile. I submitted a ‘wishlist’ ticket to BuddyPress for group activity feeds, requesting that feeds for when a new group member joins and changes to the group wire. It would also be nice to be able to aggregate content from other sites via RSS into the group ‘news’ field, or a new lifestream-like field so group photos or videos or whatever, could be sucked in. It was possible to do this via a Yahoo! Pipe which combined various feeds which could then be put through feed2js and dropped into the ‘News’ field. However, embedded javascript is now intentionally blocked 🙁 I guess I could find a work-around.
For both teachers and students, the profile pages could be effective resumes. Currently, the site-admin can build basic grouped fields and there’s a choice of field types, too. I’d like members to be able to build their own fields and for there to be pre-built field types to choose from. It’s possible for the site admin to pre-build fields and probably easy enough for me to pre-build specific fields to design a resume (the examples given of language, country and state are just .csv lists). However, currently, if I provide three ‘Employment’ fields, a member can’t add a fourth ‘Employment’ field, nor can they select dates to correspond to when that employment was. I’m pretty sure I could create the fields, but it’s beyond me to allow a member to build their own profile pages from a selection of pre-built fields.
Finally, in addition to my request for members to be able to make their profiles public, I’d like the member profile to be marked up with hResume markup and exportable in a variety of styled formats: xhtml+css, xml, pdf, txt, doc and rtf.
The entire member profile should use microformat markup where possible. Currently, the profile can export a simple, personal hCard but could also use hCard for company and school addresses, hCalendar for dates, and rel=”tag” for creating a set of tagged skills. LinkedIn partially implements this, by the way.
So, privacy controls, group feeds and a resume builder. Not too much to ask is it? I’d probably be able to pay for the resume builder if anyone is interested…
//CommentPress is, for educators, one of the most important developments to come out of the WordPress community and one of the most significant innovations that I know of in online publishing. I first learned about it when I saw that Yale University Press were using it to invite comment on Yochai Benkler’s book, The Wealth of Networks. In its original form, CommentPress is a theme for WordPress that allows readers to comment on, annotate and discuss paragraphs of text. In fact, although installed as a theme, it transforms a site not only by design, but with functionality you’d normally expect from plugins. In CommentPress v1.x, form and function came as a single package. It’s worth reading about the background to CommentPress. You’ll see that it’s part of a larger course of research by the Institute for the Future of the Book.
Institute for the Future of the Book was founded in 2004 to [… stimulate] a broad rethinking—in publishing, academia and the world at large—of books as networked objects. CommentPress is a happy byproduct of this process, the result of a series of “networked book” experiments run by the Institute in 2006-7. The goal of these was to see whether a popular net-native publishing form, the blog, which, most would agree, is very good at covering the present moment in pithy, conversational bursts but lousy at handling larger, slow-developing works requiring more than chronological organization—whether this form might be refashioned to enable social interaction around long-form texts… We can imagine a number of possibilities: scholarly contexts: working papers, conferences, annotation projects, journals, collaborative glosses; educational: virtual classroom discussion around readings, study groups; journalism/public advocacy/networked democracy: social assessment and public dissection of government or corporate documents, cutting through opaque language and spin (like the Iraq Study Group Report, a presidential speech, the federal budget, a Walmart or Google press release); creative writing: workshopping story drafts, collaborative storytelling; recreational: social reading, book clubs.
You can also read about CommentPress in The Chronicle for Higher Education and The Journal of Electronic Publishing.
We have started to use CommentPress at the University of Lincoln for the discussion of internal documents and feedback from staff has been good. Many are astonished at what it makes possible. A departmental research strategy paper received over 100 comments from nine staff; something we’d never have had by emailing the document out for comment. Of course, I am keen to use it to support courses and a colleague and I have recently applied for funding to use CommentPress in a course with over 100 Criminology students, who are normally asked to critique texts and respond by emailing Word documents to their tutor. Using CommentPress allows for transparent and open, formative feedback and assessment by both staff and student peers.
Outside of my work for the university, I’ve been developing WriteToReply, with Tony Hirst from the Open University. You can read about how we started WriteToReply and you’ll see that CommentPress is fundamental to what we’re trying to achieve and we’re using it for networked democracy, as suggested above. CommentPress is in fact, a comment engine for each document site. Two things make this possible. First, and most obvious, is the fact that readers on a document site can direct comments to specific paragraphs of text. Readers can also respond to other readers’ comments and a happy by-product of our re-publication of the Digital Britain – Interim Report, is that the discussion still continues, despite the consultation period being over. So CommentPress is an engine for on-site comment and discussion. Texts are dissected but remain whole; they also become social objects.
The second important contribution CommentPress has made is the provision of permalinks for each paragraph in the text. This provides a unique URI or URL for each paragraph of text, making linked references from third-party web sites possible. Combined with the trackback/pingback system built into decent web publishing platforms, CommentPress makes remote commenting on text possible, as Tony explains on his blog.
What this means is that the paragraph, action point, section or whatever can become a linked resource, or linked context, and can support remote commenting. And in turn, the remark made on the third party site can become a linked annotation to the corresponding part of the original report… How? Well through the judicious use of trackbacks… So even if you don’t want to comment on the Digital Britain Interim report on the WriteToReply site, but you do care, why not post your thoughts on your own blog, and link your thoughts directly back to the appropriate part of the report on WriteToReply?
It’s this feature, so easily missed, which makes CommentPress a comment engine. An engine suggests an underlying technology that drives something greater. By introducing paragraph permalinks, text can now be linked at a much more accurate and deeper level than was previous possible. Texts are transformed into uniquely identifiable resources of data. Academics can now reference paragraphs rather than page numbers and readers can reflect, comment and participate in the analysis of texts from their own site. For the reader, CommentPress provides a fluid interface to the document as a whole but at a technical level, explodes it across the Internet.
In the running of WriteToReply, we’ve tested CommentPress quite hard and found it to be a complex and fragile tool. Until recently, it hasn’t been updated to reflect the fast changing development of WordPress and because of its extensive use of Javascript, it clashes with other plugins, so while it transforms a WordPress site, it also restricts functionality otherwise possible. Fortunately, CommentPress 2 is being actively worked on and I’ve been helping to test it with Eddie Tejeda, the original developer. It’s currently in beta, but Eddie is responding to my feedback and fixing issues rapidly. There is a mailing list for CommentPress and the code is publicly accessible.
If you test CommentPress 2, you’ll immediately see that it’s been split into a suite of plugins and themes and that it’s now much more flexible in terms of compatibility with other WordPress plugins and in being able to select different components, options and themes. Notably, paragraph permalinks are available as a separate plugin, which means that any WordPress blog will be able to have paragraph-level URIs, without necessarily supporting paragraph level commenting. My test site is on WriteToReply. Feel free to have a look and post comments, if you wish. As I write, it’s not quite ready for everyday use, but at the speed which Eddie has been working over the last few days, I’m confident that I’ll be able to use it here at the university and on WriteToReply before the month’s out. If you’re used to using v1.4.1, you’ll notice a lot of change. Remember that it’s still beta software and that not all of the features have been fully implemented yet. It would be great if other people could help test it across various browsers and with different documents. Multimedia is not something I’ve yet been able to throw at it, for example.
Finally, CommentPress needs continued support in terms of testing, reporting issues, bug fxes and feature development. This can be done voluntarily, but given it’s potential to support education, business and government consultations, I for one, will be looking for ways to raise funding to help support all of this. If you know of any possible funding opportunities within UK Higher Education, please do let me know.
//“When I learned electronics, we soldered together discrete components to make a product. Today, we combine copyrighted units of other people’s intellectual property to make a product. And we don’t really have the tools to carry out that task properly.”
A comment by Bruce Perens, the creator of the Open Source Definition, the manifesto of Open Source and the criterion for Open Source software licensing. Perens represented Open Source at the United Nations World Summit on the Information Society, at the request of the United Nations Development Program.
