IWMW 2009 – Steve's reflections

Posted on July 30, 2009 by steve

This afternoon web team returned from sunny Colchester, we have been attending the Institutional Web Management Workshop. We all heard useful stuff and came away with lots to think about (particularly as we embark on the website redevelopment). I’ll take a couple of paragraphs to share the things I am thinking about.

Machine readable content

This was something more than one session touched on. The idea is that you don’t just think about the humans that will be interacting with your content, but you also provide ways for computers to interact programmatically with it.

This can be as simple as providing RSS feeds for website searches, or any list of things on your site (e.g. courses?). Or as complicated as providing a full blown API (api.city.ac.uk anyone?) with a RESTful interface.

This was covered in some detail by Mike Ellis of Eduserve and Tony Hirst from Open University, and again by the BBC guys. The idea is that this will allow other people to re-purpose your content in ways you wouldn’t even consider, and even encourage people to do interesting things. One interesting example was Tony’s own mashup map of MPs travel expenditure, where he took data from the guardian, munged it and put it on a Google map.

Amazon Web Services

Mike Richwalsky gave us a compelling demonstration of AWS, including the serving of static content from S3 and the ultimate in virtual servers EC2.

S3 is a service where you can upload content to the Amazon infrastructure and have it served superfast from their data centres. It operates on a PAYG pricing scheme, you only pay for the bandwidth and storage you use. Mike is using it to serve flash video up at his University website. And I was impressed with the quality and speed in comparison with our offerings, might be something to consider as we decide how to take the video stuff forward.

EC2 is a service where you can create on-demand computing power, bringing virtual servers up and down on a whim through your browser. Again only paying for how much bandwidth/uptime you use.

In all it was a worthwhile couple of days where we could take time out to consider the bigger picture ad rub shoulders with others in a similar situation to ourselves. I look forward to implementing some of the things we’ve seen in the coming months.

Pragmatic accessibility

Posted on June 19, 2009 by dan

Web accessibility guidelines are a mess, but that doesn’t mean we can’t develop a coherent accessibility policy.

The second session of Headscape’s EdWeb 2009 conference (I previously blogged about the first session on becoming a user-centric institution) tackled the ever-thorny topic of accessibility. Why’s it such a difficult issue? Well for a start, web accessibility legislation has a reputation for being confused and ill-defined. Moreover, there’s just so much of it – anoyone for DDA, PAS78, Section 508, RNIB See it Right, WCAG 1, WCAG 2, etc etc?

While traditionally there has been disagreement within the web development community about the best way to act on the inherently weak W3C guidelines, WCAG 2.0 appears to be a positive step forward. Why? Because it stresses principles of web accessibility and non-measurable guidelines, and in doing so it:

  1. Takes us beyond a checklist mentality, and in doing so gets us thinking about the spirit rather than the letter of the law.
  2. Downplays the importance of conformance levels (“our website is AAA-compliant – ner ner ner ner nerrrrrrr!“)

But ‘web accessibility’ isn’t just about the traditional understanding of ‘disability’;  we should give it the widest possible interpretation – allowing our content to be accessed by the widest possible audience – and in doing so think about issues like bandwidth, old browsers, alternative devices, assistive technologies, mobiles, seach engine optimisation, etc. etc.

So what should our strategy be for dealing with this interpretation of web accessibility? Here’s the 6-Step Plan:

  1. Start with the basics (‘alt’ tags for images,resizable text, Standards-based code, etc).
  2. Educate out content editors.
  3. Create policies for hard to implement functionality, e.g. video captioning, media alternatives.
  4. Respond quickly to accessibility support requests. Fix or explain; never ignore.
  5. Introduce Graded Browser Support.
  6. Provide simple user guidelines for changing browser settings and making our website easier to read.

Serena Collage and Internet Explorer (IE) 8 Fix

Posted on June 11, 2009 by rik.williams

Microsoft have this week released a bundle of updates and security patches for Windows XP and Vista users that includes a recommended upgrade from Internet Explorer (IE) 7 to IE 8. Upgrading will probably mean that you will not be able to upload assets (e.g. images, Word files, Excel files) to the CMS. To fix this problem, you will need to follow the steps below:

  1. In IE, click ‘Tools’ then ‘Internet Options’.
  2. Select the ‘Security’ tab.
  3. Select the ‘Trusted Sites’ icon, then select the ‘Sites’ button.
  4. In the ‘Trusted Sites’ window, type ‘https://*.city.ac.uk’ (without the quote marks) into the ‘Add this website to the zone’ field, select the ‘Add’ button then the ‘Close’ button.
  5. Now select the ‘Custom level…’ button.
  6. Find the “Miscellaneous” section and look for the option that says “Include local directory path when uploading files to a server”
  7. Make sure that option is Enabled.
  8. Select ‘OK’, and ‘OK’ again to close all the Internet Options windows.

You should now be able to upload files to the CMS using IE8.

On using Zend Framework

Posted on June 5, 2009 by steve

We in webteam work on a lot of unseen stuff behind the scenes. I thought it might be nice to share a bit of this work with you (don’t feel obliged to feign interest though, you can leave now if you like)

Frameworks

The modern approach to web development seems to be all about frameworks; there are 96 listed on wikipedia’s comparison page.

Frameworks provide the bones of an application (on to which you build the flesh) – they do a lot of the repetitive hard work, leaving the bits that make your app unique down to you. This is good because nobody wants to waste time reinventing the wheel. They’re also very trendy at the moment, and we in the web world are a bit partial to shiny new things (and there is plenty of shine if you look at some of the frameworks home pages). It seems if your not using a web app framework you are behind the curve.

Zend Framework logo

Here in webteam we are not immune to a good idea, so we’ve been using one such framework to do a few things. The one that has been floating our boats is called Zend Framework. It is an PHP open-source framework, which works along the popular MVC paradigm. It’s not quite as trendy as Ruby on Rails (I can’t imagine A List Apart publishing articles about it), but useful none the less. It also suits our PHP skills, so we don’t have to learn a new language.

I don’t think it would be suitable to drive a large site with lots of content (like www.city.ac.uk, for example), but for a one off stand-alone project it’s very useful.

If you do any kind of server side coding in PHP, I can heartily recommend ZF to you. It will make your job so much easier and your code so much more structured.

Has anyone else had any positive experiences with application frameworks?

Becoming a user-centric institution

Posted on June 1, 2009 by dan

Over the last couple of weeks I’ve managed to get away from sunny Northampton Square to attend two excellent events for HEI ‘web professionals’. A couple of weeks ago I took part in the inaugural session of a networking meet-up for London university web people, organised by the excellent University of the Arts team. UCL’s Jeremy Speller has already blogged about this so I won’t go into too much detail, but it was fascinating to find that we’re all facing the same problems, and were eager to talk about the same issues, namely:

  • management of content (quality, quantity, CMS selection)
  • use of outsourced vs. internal resources
  • organisational issues, bureaucracy, politics
  • appropriate usage of emerging technologies (notably Twitter), and best practice use of social media tools for student engagement

Unsurprisingly, the same issues cropped up frequently during the excellent two-day EdWeb 2009 workshop hosted by the gents at Headscape, about which I plan to blog over the coming days.The first session, entitled ‘Becoming a user-centric institution’, was particularly pertinent for City University because of the plethora of major digital projects that we are committed to, namely website redevelopment, Portal and CRM.

User testing should be a no-brainer, but Paul Boag argued (justifiably) that the HEI sector has a terrible track record for developing websites and applications with user needs as the number one priority, and we see evidence of this everywhere. Oftentimes, ‘user acceptance testing’ is no more than a process to verify that an application or process ‘works’;

“we think that most users can complete the form, and the back-end business processes are all working, so everything’s fine….“

What’s missing here is a focus on user objectives rather than business objectives, and an appreciation that user testing provides a valuable opportunity to enhance the user experience; to quote usability guru Jared Spool, “its real forte is in telling you where the interface causes frustration”.

Aside from the specifics of how to conduct a user testing session, the main thrust of this session was that none of the ‘traditional’ excuses for avoiding user testing (no time, no money, no understanding of the benefits, no expertise, etc) are relevant or insurmountable, and that a we should be adopting a ‘little and often’ approach to user testing throughout the development cycle i.e.:

  • during requirements gathering (test the existing site or application)
  • when developing Information Architecture, taxonomy
  • design stage
  • build stage
  • post-launch

This makes a great deal of sense, and I think we all came away with a renewed desire to educate our institutions on the benefits of user-centered design.