Social Communications

Many of you will know that the OU has produced a few Facebook apps (blogged previously here and has since added others such as Devolve Me) and have had a recent foray into iphone / ipad apps. I’ve only been involved in the development of one (Study at the OU) which tony has blogged about – actually that reminds me there was a bug reported which I must investigate…).

The point of the apps that I’ve been involved with have been 1) to learn how to work in these environments (both in terms of development and in terms of what it means for the OU to contribute and promote activity in these external environments) and 2) to provide some kind of added value to members of the OU community in external social networks (whether they choose to engage with it is obviously entirely up to them).

So for me as a comms person I suppose the big benefits of engaging in this kind of activity might fall into the areas of ‘relationship management’, ‘brand reputation’ and a general early investigation of possible new channels of communication. What I have never seen (especially the FB apps) as being explicitly about is selling courses. In a doc I’m drafting at the moment about why the OU engages with social media i’ve put

…Not all of the fans / followers of OU social media accounts are staff, students or alumni of the OU however. A number are interested in the research and content (itunesU, YouTube, study tips) that the OU makes available. Some of these individuals (via engagment with the OU community of staff, students and alumni in social media environments) will go on to become students. It is not however the purpose of the OU’s engagement with social media to recruit students, rather the intention is to provide added value to our followers or fans and enrich the conversations, groups and relationships that users choose to form for themselves. In so doing it is hoped that the reputation of the university is enhanced and communicated and that new people are introduced to OU content.

I may end up changing this a bit (and maybe improve the sentence structure!) but I think I’m reasonably happy with this as a summation for now.

Anyway, where I’m trying to get to with all this rambling is that I think by explicitly not trying to sell modules to people but rather acting in a way which we believe adds value to them we may infact be having the happy side-effect of driving people to the OU online prospectus and generating registrations. Let me share a few facts that were pulled out of our analytics package by a colleague in jan of this year.
The two Facebook apps that I have been involved with (My OU Story and Course Profiles) have generated 943 unique visitors to to Study at the OU – this represents a 12% click thru rate. They have also generated 10 registrations to OU modules (which represents a 1% recruitment conversion rate). Now, as you will have gathered i’m not a marketing person, but I would imagine that both of those figures stack up pretty reasonably against traditional marketing activity – if anyone has any average figures for this kind of stuff I’d love to hear it!

The Study at OU iphone app (which was release late last year) had generated 429 unique visits to the website (19% ctr) and generated 4 registrations (0.9% rcr). As a little bit of extra data for you 55% of the visitors to the site were visiting the OU site for the first time on a mobile device, and only 25% of visitors were existing students. We haven’t really told anyone about this app yet so I’m confident that we can increase the number of users quickly when we do.

So, what to take from this? Well 1) I think I can safely say that if someone were to ask what the ROI of this activity is in terms generating enough income to cover development time we are well and truly covered. Without divulging too many details the income the OU will have received from those registrations (not to mention future study those people may go on to undertake with us) well and truly covers costs. In addition to this the relative value of visits to the site in terms of what we may pay google to drive traffic to us further tips the balance in our favour. 2) I think that building apps that are designed to sell may ultimately have less success in doing so than those that are designed to add value. Again I don’t have any stats from other FB applications to back this up so if you are able to share some that would be great.

So, I’m about to begin a new bit of development working towards an HTML5 app that we can build on top of to provide android, iOS, Chrome etc applications (instead of starting from scratch for each). Once that’s done I’ll share more about the details and posts some stats. Similarly I know that Liam is doing some very cool stuff using OU data from data.open.ac.uk and building android apps and TV-based apps. Also Mathieu (@mdaquin) has a very cool OU-based android app that he is not sharing at the moment cos it’s buggy. I suggest we lobby him to adopt a ‘release early and often’ strategy

Share This

So, you know the story of why the Open University has been using Drupal as its web CMS for the last year or so, now I want to say a little about how it actually works. First up – Module selection!

If you are reasonably familiar with Drupal (or any open source application) you will know that one if it’s great strengths is also one of it’s possible weaknesses: almost anyone can provide additional functionality to the core offering and this leads to many different possible ways to achieve the same functionality.

At the OU the maintenance of Drupal falls to the doughty souls in IT (and in particular mostly to @stuartcrouch!). This is a tricky enough task as it is what with all the various point updates that come along for various contrib and core modules but it’s even harder if as a university / organisation we support 5 different modules that serve the function of adding an image to a page. For this reason one of the first steps we took when ‘officially’ supporting Drupal was to do a lot of research around ‘best of breed’ modules for certain functionality and work internally to ensure that developers knew that if they wanted to achieve certain things they should use the supported module. Alongside the tech maintenance advantage this has it also help with end user training since they are presented with a common workflow regardless of the Drupal site they are admin’ing.

In order to help developers to know what the supported module were we made this info available on the Drupal wiki which includes a page on supported modules. You can see that there are a number of modules that are listed as mandatory (most for SEO reasons).

One of the unfortunate side effects of this approach (which has become known internally as ‘standard modules’) is the urban rumour that went around the university (and still persists in niches) that you were not allowed to use any other module than those listed on the wiki. This isn’t true (or isn’t completely true!), the basic rules are:
- Where functionality is provided by a module is in the ‘standard’ list that module should be used.
- If there is a conflict between the version supported by the OU and another module you are using don’t update to the next version without consulting IT (because other sites may be relying on a particular version for their functionality).
- If you want to use a module that isn’t on the list let us know before you go too far down the development track. There are a couple of things that we look for in good module: compatability / complementary with existing standard module set and active maintenance (i.e. an active issue queue and a history of regularish updates).

One of the issues related to this (and it generally comes from those unfamiliar with Drupal or those who don’t want to spend time looking for contrib modules / learning how to use their full feature set) is the request to build custom modules. To be clear, there is no rule against building custom modules at all, but there are a few criteria that need to be met before such modules will be supported:
- Functionality does not already exist within the Drupal community.
- Functional spec should be agreed and the benefit derived from the module should be as broad as possible. For example if faculty a wants to build a module that hooks into a particular OU system it should not be built as a faculty-specific solution but rather a general solution which can be used by others around the university.

So far we have 4 (well, 3 1/2! OU custom modules):

1. SAMS authentication – So OU staff and students etc can sign into Drupal using their OU identity
2. Documentum – The OU uses documentum as its document store / lifecycle management thingy and the DocuDrupal module allows people to link Drupal to Documentum. This means that if I have documents published on my website I don’t need to update my site when I edit documents in Documentum, it all happens automatically
3. Brand module – this helps our sites to adhere to the OUICE (i.e. online brand) style by overriding some of the HTML that Drupal / Views spits out. The module also helps us out with some of the issues we have had with Drupal around how menus (and menu hierarchy) and breadcrumbs work. It also has a nifty way of allowing us to tokenise CCK fields so they can become ‘common content’ that can be added anywhere within a site.
4. People Profiles – is a module that is under development that will enable OU sites to access staff data from a central data source for presentation on their site. There are basic permissions around this both for the individual (‘I don’t want my email address to be available’) and the site (I am the Arts faculty site please give me the details of all members of the Arts faculty’)

As I said waaaaaay back at the beginning of the first post on this, one of the reasons the OU has adopted Drupal as it’s web CMS for non teaching and learning sites has to do with cost-saving. To further help with this we have done some development work and created two skeleton sites (one for intranet and one for public) which, experience has shown, actually work for 90% of the end users out of the box. This means that various projects and even faculties are able to get the website they want without having to spend a penny on development time (content-creation and IA work yes, but not development!). Some time later down the line I’ll post on exactly how these skeleton sites work for us.

Share This

For many years here at the OU there were a couple of ways of creating non teaching and learning websites: you either used a tool called APS (Assisted Publishing System??? Which I think was developed in house), or you went to an agency.

APS spawned a couple of related systems and was is great for easily publishing very simple sites that are primarily text-based. Over the years however as the web has moved on and expectations have grown on the part of both site owners and visitors about what websites ‘should’ do, increasingly external agencies began playing a role in web development.

I have absolutely nothing in the world against agencies (some of my best friends are agencies…), but the involvement of numerous externals started to bring a few issues along with the obvious benefits. Some of these included:

• Cost – Everyone knows that HE funding is reducing and costs associated with having agencies build each new site that comes along de novo is obviously one of the areas where HEs need to try and find ways to reduce costs.
• Tie-in and Tech proliferation – Frequently agencies sell you a website built on top of their in-house CMS. The huge benefit of this is that frequently these systems are better and more user-friendly that the existing university in-house solutions. The downsides however are that you then become tied to that agency for tweaks to the CMS, upgrades/new features that are produced, compatability issues (what happens when your sysadmin upgrades PHP and your CMS breaks?)
• Training – With each new CMS comes another round of training; frequently with there being a small number of ‘web people’ in each faculty having to learn the ins and outs of a number of CMS’. This is sort of OK whilst they are there but then one day they leave and take all of their knowledge with them. Now the original documentation is lost, and the only person who knew how to update the site has gone……

So, it was against this background that the OU experimented with using Drupal as a CMS on the basis that:

• The software at least was free. You may need to engage with an agency or freelance developers to build your site, but you could guarantee that you would benefit (for nothing) from the upgrades provided by the Drupal community
• We could limit the amount of applications that our IT dept were required to support. By saying that the default OU position was that non teaching and learning sites should be built in Drupal the support overhead could be significantly reduced. This obviously isn’t a benefit of Drupal per se but a benefit of taking a clear and internally supported position.
• There is loads of free training available online for Drupal (both within and without the Drupal community) so the chances of knowledge vanishing from with the university is reduced. Similarly if you are the person who ‘does web stuff’ in your faculty you know that if future sites are built in Drupal the training that is required to administer/update those sites is minimal or non existent. A time and money saver!

I’ve gone on a little more than I planned without getting to the real meat of the post so I’m going to split this into two posts. In part two I’ll explain a little more about the Drupal setup here at the OU and address issues such as contrib module selection and support, updates to core Drupal and contrib module, the development environment (dev, acc test, live), problems we’ve faced, the custom modules the OU has developed (one of which we hope to contribute back to the Drupal community).

Share This

I’ve been hitting snooze on the annoying reminders to blog for (after a quick check) just over a year now. For much of that time (and before) twitter has served the purpose for me that blogging used to but recently it’s become increasingly clear that 140 chars just doesn’t work for everything.

So, I’m not going to go into a great long post here, but I am going to list a few things that I want to cover in future posts (and hence shame myself into actually having to post on them). Here’s the list in no particular order:

• Implementing Drupal as main CMS at the OU – actually I think that may have to be a few posts but I’ll have to think about how to break it down!
• data.open.ac.uk AKA LUCERO – The OU’s venture into linked data (and becoming the first university to mint a data.*.ac.uk URI)
• How we are going to make use of data.open.ac.uk
• Study at the OU iphone app released a few weeks back (with added google analytics goodness)
• OU online identity – mostly in the form of OUICE masterminded by @guyweb
• OU staff people profiles – not yet finished, but hopefully a building block for great stuff in the future!
• OU web2 identity – different from OUICE (but related). An imagining of OU identity in web2 world (and the intro of open.edu)
• Web2 guidelines for staff and students – again not something that is finalised yet (and maybe something I duck out of posting about!)
• Dreams of aligning web publishing strategy with feeding data.open.ac.uk (AKA must I really wait for Drupal 7??)
• Workshoping applications / new data to expose as LOD with Talis. Who actually does the work of modelling / building is a different matter!
• Redevelopment of Platform, the OU’s online student and alumni community site

That’s probably enough for now – I’m starting to scare myself.

Share This

Some stream of consciousness thinking here about the opportunities that may (or may not) have opened up with Twitter lists.

I look after the open university twitter account and today put out a tweet asking followers what lists they wanted creating / or wanted to join. I got a massive response and although the majority of the lists currently contain only one or two people I hope that will change over time.

The types of lists that I was asked to create fall into one of two categories: either a course list DD101 or a subject list Law. The first of these is great because it enables students / alumni of particular courses to come together in a way which is a little more formal than following a hashtag.

I wonder if it’s the second of these (law / psychology etc) which holds real promise for the organisation using twitter in terms of developing partnerships with other organisations in order to create extra value.

Already students have asked me to add the law society gazette , the law society for england and wales and the law society scotland to the law list and this made me wonder about how this might encourage us (as twitter accounts or even organisations) to work together to create extra value for people who use the list. I’m sure we all have resources available to us that each others’ followers would appreciate.

On a related note, if we are to build up really valuable subject-specific lists I really need a way of feeding content into specific lists using hastags. Anyone know of a way?

EDIT: I also need a way to sync list followers with people in the list – kind of like an automatic follow for people who follow you. I think this is the right approach to take (is it?!?) since the value of the list is created by the people in it. you follow the list because it is something you re interested in / tweet about. If that is the case other people on the list want to know about that content.

Share This

Yesterday was a happy day; I was one of a lucky 100,000 to get an invite to try out Google Wave. Seeing as I’ve given over almost every aspect of my life to Google I suppose this is probably the least they could do in return!

So, what’s it like? Well I suppose the answer is ‘I’m not sure yet’. This is what it looks like though!

The first problem you come across when you start up your account is that you can’t import your existing contacts list (cos they’re not using wave) and so you don’t actually have anyone to communicate with! A quick post to twitter of your address (jinky32@googlewave.com – add me to your contacts list!) and all of a sudden you have LOADS of people to talk to – hence the ‘drowning’ part of the title!

The big wow moment for everybody that I’ve spoken to is that you can actually see people’s messages appear on screen as they type them!. Ok, so this isn’t a killer feature in terms of functionality, but I think something that claims to be as revolutionary as Wave needs something new and shiny that people haven’t seen before.

Wave also allows you to add a (currently small) number of widgets so that you can add them to your waves. By default you have a polling app installed (‘yes’, ‘no’, ‘maybe’ for respondents) and a google map app which doesn’t work for me; it works for others tho so presumably i’m being dumb.

One of the problems I’m finding at the moment is that it’s quite difficult to stay on top of what’s going on in a wave since communication happens at all levels

People can reply to the wave, reply to individual wavelets (particular ‘posts’ in the ‘thread’ to use old fashioned terminology ) and can also edit wavelets that you have posted! It’s a bit artificial to criticise over this at the moment tho I think cos the early waves i’ve been involved with have included about 15 people all doing all sorts of crazy stuff within waves simply to test out what they can do. Once we all get used to using Wave as a tool rather than a plaything it’ll be easier to assess how the new features stack up.

Share This

I’ve long been a fan of services such as Scribd and Docstoc which allow people to share their (or in some cases, other peoples’!) material and enable a whole load of those lovely web2 features that we’ve become so used to that we expect to be almost everywhere.

Obviously given my overwhelming enthusiasm to give my entire life data to H.A.L I’m also a huge fan of Google docs so I was pleased to get an email from another of everyone’s favourite embeddable service Slideshare telling me that I can now upload and share my Google docs via their service.

The process is reasonably slick. Choose the ‘upload google doc’ option and you’ll be asked to give slideshare permission to access your Google docs account. Once done you can choose to upload al or some of your docs and change the titles by which they will be known on Slideshare.

It took a fair old while to convert the doc into Slideshare format (about 12 minutes), but the end result is pretty satisfying. Here’s a copy of the OU Boxee submission document.

Boxee submission

View more documents from jinky32.

Hmm, there seems to be some issues with embed code on the page, but you get the idea!

Share This

Great title huh? You can take the boy out of SEO…etc etc

So, it’s been pretty much radio silence from me over the past few months – essentially I suppose because my role has changed a little and I’m involved in more long term projects than quick win ones.

One of the major things I’ve been working on at the OU is the mainstreaming of Drupal at the OU as the CMS of choice (for non-learning sites). There’s a whole load of work around the environment, security, update strategy, single sign on, theming etc that we’ve had to get to grips with but I’m reasonably confident that come October time the OU will have a great Drupal infrastructure in place. The aim is that it will be both manageable from a support perspective and flexible enough to give developers what they need in order to build the sites they want.

Alongside this work there’s also been some pretty deep thinking about how Drupal can serve as a web publisher for our Content Management System Documentum. Now I’m no Documentum expert so for those of you like me, think of Documentum as basically a big file repository (docs, images etc) where instead of 10 people each having a copy of a single document, the document exists in one place and ten people (if they have the right permissions) can work on it. This obviously means that there is a lifecycle to the document (i.e a version is no longer current) and that there is a LOAD of metadata associated with it.

The challenge in linking the two systems has been around how and where Drupal publishes the files, how Drupal accesses these files, how metadata is handled and how the lifecycle of a document is managed (e.g. how do i ensure that the version of the document on the drupal site remains the right one when the document is updated x number of times in Documentum?).

I’m not entirely sure how much detail I can go in to on this (at least one member of the team is hoping to do the conference circuit for a couple of years on the back of this work ) but it involves some nifty XML work and a rather cool OU custom Drupal module (which we will hopefully be sharing back to the community once we’re finished).

If everything goes well I’ll be able to show you the first live site with all this working by October. Fingers’ crossed!

Share This

Twitter is obviously one of the big online success stories of the past couple of years. Driving to work I’ve even heard Radio 4 talking about it, and various BBC 5 live ‘phone ins’ now actively ask people to tweet them messages. Seems like it’s becoming a recognised channel of communication.

About a year ago enterprise versions of twitter started cropping up, most notably Yammer. These are a great idea in principle because they offer a company all the advantages of twitter but also provide the nice cuddley blanket which is a feeling of control over the environment.

The big problem with these offerings of course is that hardly anyone used them. If you ‘got’ twitter (and are therefore a target audience of Yammer etc) then you have a twitter account and you don’t really want to have to go elsewhere to re-enter your 140 character aphorisms. If you don’t ‘get’ twitter then you are unlikely to be interested in Yammer either.

Back in the dim and distant past I set up the open.ac.uk yammer account. As suggested above some of those OU folk who get twitter created an account and then never used it! no-one who didn’t get twitter created an account (tho admittedly i think i only told people about it via twitter )

Anyway, today (or yesterday perhaps – I have no idea what the date is) Yammer announced Twitter integration by using the #yam tag. I’ve given it a quick test and it works perfectly. Whether it’s enough to reach the tipping point I don’t know, but it’s certainly a step in the right direction!

Share This

The BBC has joined the embed party and announced that BBC News videos will soon be available for UK bloggers to include in their posts.  There are details here on the hows around this, as well as details on the all important T&Cs.

Here’s an example of how it looks in the flesh – pretty nice huh?

Share This