Title:
Move of to new technology
Description:
Project 93 – “Future Direction for the main Informatics web…” has gone beyond its original remit, which was (briefly) to investigate possible routes we could go down to improve the current web site and better tools to support research web site needs. This project retrospectively covers what actual steps have now already been taken following that initial investigation, and what still needs to be done.
Background:
At the time project 93 was started (2008?), the University Website Project (to move all University pages to a new content management system called Polopoly) was in its early stages, and was not (at that time) suitable for our needs. There were a few technical things we needed, that it did not support. So project 93 concluded that actually our current technology for managing was not that bad, and given a few tweaks it could soldier on for a bit longer. However it was accepted that we needed our own WCMS provision to support the web requirements for research groups. At this point project 93 should have stopped and new project(s) started to track the progress of the new WCMS and the improvements to the existing technology, this didn’t happen.
What happened was that we came down to a short list of Drupal or Plone as the tool to use to host our own WCMS service, and Plone was chosen as the most mature/stable of the two candidates. And we quickly had an initial Plone service up and running. This effort was recorded against project 93.
A while later the School was going through a rebranding exercise (commercialisation), and the University was rolling out a new look to its fledgling Polopoly pages, and it was felt that we should provide either/both of these new looks to the WCMS content. So more effort was spent producing a new Informatics “skin” – our so called “Institute In A Box” - for Plone, and this too was recorded against the original project 93. Again, way out of project 93s original remit.
Towards the end of 2009, we were approached by the University Web Project people to encourage us to put at least a minimal amount of content into Polopoly, and we agreed that we could/should. Again the time spent on this has, and is, being recorded against project 93. We’re at a state now that in a few weeks time we’d hope to be able to go live with the initial content in Polopoly.
Plan:
In this project we plan to record the work already done in transitioning Informatics web content to the new technology (both Plone and Polopoly). And what still needs to be done. For example we need help from the RAT unit to aid with generation of some new DB feeds that we can proxy into Polopoly.
As so much time has passed since we originally looked at moving to Polopoly (and ruling it out), the Polopoly service now meets some of our original requirements, and is more flexible that it once was. The UWP team are willing, and encourage, external collaboration to keep improving the service. Once we do go live with our initial content we’ll want to review how that went and if we (to make more use of central services), should continue to migrate more content.
Again going back to our choice of Plone, as so much time has passed, I think we should at least consider the option of supporting another WCMS. Last year I did a quick survey of what other schools around the University were using, there is a small clustering of Drupal and MySource Matrix users (which there wasn’t previously) and we are the only people using Plone. Within Informatics ourselves, some research people (AIAI) are running their own Drupal sites. For these reasons (and the steep learning curve that we’ve still to climbed for Plone), I think we should at least consider supporting another WCMS.
For ww.inf content that doesn’t move to Polopoly, we need to state where it will stay, and how that data will be managed, served, and made resilient against hardware failure. We have an outstanding action to detail our plans on off site duplication. This project would detail those plans.
Deliverables:
Robust infrastructure for content remaining within Informatics.
For content on Polopoly we will need to maintain local technical documentation on what we are providing in the way of DB content that is being proxied to Polopoly.
Assuming we make the switch of the first batch of content to Polopoly, we will provide information on the (no doubt) complicated redirects forwarding old URLs to new Polopoly content.
For Polopoly content authors, local guidance on the basics of how it works (IS run proper training, and provide documentation and support).
For Plone, local guides/training on how best to use the Institute In A Box web site template.
Dependencies:
Access to Informatics DB to produce the data we want to proxy into Polopoly, or give RAT unit details of what we need, and ask them to produce the conduits/data.
Support from UWP to give required access to Polopoly content providers.
Risks:
Currently we have started moving some content in to Plone, mainly ISS/IGS content. Should we decide that Polopoly is now suitable for this content, we will have to reimplement that content. There would not be a practical migration path to do things automatically from Plone (or probably any other WCMS) into Polopoly.