When The Email Goes Dark…

Update regarding the state of emails: I’ve not heard official words from IT on the state of play of the email server, however my account (which was out yesterday) is now back in action. I’m guessing this is a good thing.

Today, along with 1/3 of the other staff and students at Lincoln, I’ve been devoid of emails. This is down to a problem with one of the three email stores at the University, and includes staff accounts beginning with letters A, B, M-O and V-Z along with a third of student accounts (pretty much at random). People are working on fixing it. You can keep up to date with it on Get Satisfaction.

Oddly enough this has let me spend a couple of hours working on stuff without being distracted by people asking silly questions. Instead I’ve been looking at the user interface tweaks necessary to encompass some changes to the student halls network access controller, and thinking more about the dream of a common design and components for web services.

Put simply myself (along with my partner in crime, Alex) have been throwing ideas backwards and forwards for a couple of weeks now on the subject of a single coherent way into all of Lincoln’s web services, inventively dubbed my.lincoln. The idea is of a single website which collects and collates everything you might need to know from the myriad of services as well as letting you fine-tune how they work for you.

The current 'gateway' style used by Online Services.
The current ‘gateway’ style used by Online Services.

As a part of this (once I’d beaten another kink out of how Vista behaves with PFMPC ((Print From My PC))) I began mucking around with some CSS, aiming to throw together a layout based on something Alex mocked up. The old ‘gateway’ style in use in several places is actually quite messy behind the scenes, is extraordinarily narrow, and doesn’t provide much flexibility. You can put buttons at the top, and then a load of text.

There is also a ‘new gateway’ style which I knocked out for PFMPC which fundamentally looks the same (or at least very similar) but which is completely standards compliant with the exception of some little bits of CSS. However, this still has the problems of being narrow, a bit dull, and lacking in anything which makes you go “wow, this is a great, well designed web service”.

Which is why Alex and myself decided a change was needed. Something wider, faster, cleaner, smarter, more flexible, more appealing, ready for Web 2.0, ready for single-sign-on, accessible, standards-compliant and ready for use in every browser we could think of (including Lynx, and even including IE6). We’ve nicknamed it the “Common Web Design” in the vague hopes that the name will explain what it should be used for and people can latch on to the idea.

The very first version of the CWD.
The very first version of the CWD.

Along with this will come a set of guidelines on how to write content for the CWD so that everything clicks together nicely. The whole thing is specifically designed to be portable between services (perhaps using the c.lincoln.ac.uk storage location for CSS, images and JavaScript). More importantly I feel that the CWD is a deliberate disconnect from the old look and feel. Things using the design won’t be a re-hash of the old systems with the same quirks, they will be ground-up redesigns with goals of ease-of-use and interoperability explicitly in mind.

This is very much a work in progress and probably won’t ever be seen in the wild, but we can hope. Ideally I’d like to get the design finished and roll it out for PFMPC and LUNA to help spread the message that ‘things are changing’, but since I’ve mostly done this in my own time and off my own back I may surprise people.

Unifying the Look

During the saga of getting Print from My PC working, I had to build some pages to help people set the whole thing up. Predictably, this included clobbering some HTML and CSS around (and for the record, I still hate ASP with a burning passion).

Online Services already has an online ‘look’ which is visible on the Gateway, which seems to be used wherever possible. The trouble is, the entire layout and design is based on some very old HTML and CSS (and an inexplicable reliance on JavaScript, which I’ll ignore for now). Each individual subset of the online services provided by the University has a subtly different stylesheet and a different way of doing things, so for Print for My PC I decided to mostly scrap the existing code and start from scratch.

The result is visually almost identical, although in a few places it sports crisper lines and cleaner finishes. However, behind the scenes the CSS is smaller, faster, slicker, more standards compliant, provides better support for assistive technologies, makes greater use of flexible positioning and so on. There’s a ‘standard’ stylesheet to provide the unified look, and then an individual ‘tweaks’ stylesheet for some PFMPC specific colour adjustments. Finally, PFMPC makes use of some common JavaScript scripts to handle the nice lightbox effect for popup images.

Why am I blogging about this? Well, firstly this is a blog which in part aims to let you look behind the scenes at what’s happening. Secondly and most importantly, however, the code is specifically designed to be incredibly portable between various services. I envision there being yet another subdomain (sorry!) such as common.lincoln.ac.uk, which exists purely to store objects shared by all services. The stylesheets, the images, the scripts. The result once properly implemented would be ultimately beneficial – it would improve caching, reducing hits to services. It would enforce consistent appearance. Applying a code or styling fix would need to be done once and would be replicated to all services using it. Finally and most importantly, it would get people into the mindset of “data should only exist once” which is a key part of the Web 2.0 way.

Still, mindless code changes for such a minor improvement would be senseless. Which is why I’m going to suggest the following: create common.lincoln.ac.uk, add the resources to it, and when new services are created or services are updated make them use the common resources. It’ll be slow, but it’ll be worth it.