Sunday, June 13, 2010

Koha 3.2 : reflections on going live

Its been a long long time since I wrote for this blog, way to long, and I'm taking this cold wet Sunday afternoon to look back over the last few months.

Koha 2.2.9 > 3.2
Back in September 2008 we made the decision to upgrade from Koha 2.2.9 to Koha 3.0. We realised very quickly that this was not going to be a mere upgrade, but effectively an entirely new system. The biggest change was the shift from our RDA-like biblio-group-item arrangement to a biblio-item arrangement. This change was essential to make the MARC work nicely and MARC is the key to all the groovy stuff that made 3.0 so attractive.

After investing a couple of months getting to grips with what the change to 3.0 would mean the decision was made to wait for 3.2 since the upgrade was going to be so huge it was better to wait a few months more and go the whole 9 yards. Looking back, this was a very good decision and meant we went through 1 delayed upgrade n0t 2 in 12 months.

Our Goal
We decided early on that our Koha had to be developed from the patron's view not the staff's and the OPAC would double as our website homepage too. We also wanted our Kete Horowhenua digital content to sit alongside Koha search results. We asked Katipo to write the html required to customize our Koha using the built in system preferences in Koha , and also help us to create the 'inside' pages which included setting up a Trust Kete.

I also needed to delve deeply into the system preferences and tools and make informed decisions about the myriad of system settings. This was a significant and important step. When a library chooses open source software its different to choosing a proprietary system. The responsibility lies with the library to make the software do what you want it to do, and of course if it doesn't then you can make changes or develop enhancements so it will do what you need it to do.

Lets do it!
So the Koha upgrade was shifted to a back burner and a year passed, and then at the end of March 2010 the decision was made that we would go live in 8 weeks - ready or not. That right there is the reason I havn't posted for a while :) We couldn't wait for the official 3.2 release as our funding was good for 2009/2010 only. If we didn't spent the upgrade budget by the end of June 2010 we'd lose it.

That decision to go live in 8 weeks was another good decision looking back. It forced us to get on with job, make decisions, assign priorities and just do it. We focussed all our efforts for a very concentrated, intense period of time. We had to decide what had to work on day 1, and what could be left for day 2, day 3 etc. 3.2 is still in development and the code was being enhanced constantly throughout the 8 weeks leading up to Go Live, and in fact continues still.

This takes a bit of getting used to if you aren't used to the rapid pace of development that happens with Koha where things can be fixed from hour to hour. Our ex Head of Libraries Rosalie Blake was dragged out of retirement and asked to 'learn' 3.2, prepare tutorials and then train all 25 staff .... identifying, reporting and fixing bugs throughout.

Go Live
We had a soft launch 2 days before we officially announced it. We had signs up in the library and all staff wore badges saying we were in training in the hope that people to be kind with us - and they were - which is good because we really were making it up as we went. One problem which tripped us up was the slowness of 3.2 compared to 2.2.9 but changing browsers from Firefox to Chrome saw an immediate improvement.

I have no regrets for our seat of our pants approach. There is nothing like the first day of a Go Live to test software. Very little didn't work well enough because we had tested all circulation and accounting aspects pretty thoroughly. But some stuff wasn't perfect on day 1. This meant that all staff were involved in identifying bugs; not just saying 'this is broken' but "this is what happened and here is how you can replicate it". Some of the stuff we would never have found in a testing environment and I completely endorse the advice of going live when its close enough not when its perfect -- coz it never will be. Just make sure you have your staff are on board and stay calm, and that your Tech-Guru is on hand to fix things as they crop up.

Our OPAC is the jewel in the Koha crown and staff are so proud of it and get such a buzz out of the public's response to it. We have added so much value to the patron's experience and when things settle down we will hold sessions showcasing the new features. We spent a bit of dosh buying in supplementary content so we could launch with a stunning OPAC. Library Thing for Libraries, Syndetics and Amazon all add value to our catalogue, plus a 3.4 enhancement which draws on our 10 year store of issues history to provide a 'people who read this also read these" service.

To maximise the value of our OPAC we really needed to issue passwords to all our library borrowers - and of course showcase the new functions that were now available. So 3 days after Go Live we started reregistering all our borrowers. That was 2 weeks ago. And what a fortnight. But also what a fantastic opportunity it has turned out to be. We have had 4 staff working flat out issuing new library cards to patrons and in the process we get to point a few things that they might find useful, based on clues picked up in the 2 or 3 minutes it takes to check the Patron's data and assign user logins and passwords. The power of conversation: don't underestimate it as a marketing tool.

What next
2 weeks post launch and we are pretty much there. A few wrinkles in acquisitions to iron out and a few enhancements but thats about it really. Overall this has been a remarkably stress free upgrade. I'm glad it was hard and fast, 8 weeks of concentrated effort really, and the next task is to maximise the PR value of the new system and use it get out and connect to as many different pockets of our community as possible.

Final Words
  • Identify what your priorities with Koha are. Our primary goal was a fantastic OPAC because that's that makes Koha 3.2 a great LMS compared to the others. We also needed a solid acquisitions module with a clear audit trail,
  • Think about how to 'present' your collection to library patrons in the OPAC,
  • Get totally familiar with all the system preferences and how different combinations can be used because Koha is very powerful and flexible and highly customisable,
  • Just do it: go live when its good enough and tie the loose ends up as they appear,
  • market, market market because 3.2 is fantastic.

No comments: