A few weeks ago Equinox Software published a blog post I wrote about Evergreen in 2009. My first draft and my final draft were very different. Draft by draft I stripped out the history of how SCLENDS started, not because I didn’t want to tell it but because in the larger Evergreen context it wasn’t what I wanted to say. The very fact that some remained though and that I did start with so much tells me something. It is a story I want to tell and while that post wasn’t the place, this is. Why? Honestly during that first year we did a lot of “make it work and fix it later.” Document? If there’s time. It’s easy to be critical of that approach but we had tight deadlines and if it hadn’t been done the way it was it might never have happened. But now I have a little time to write it and want to do so while my memory is clear, at least of the elements that stand out in 2009.
I’m not going to claim this is a complete history. Beyond the fallibility of memory I doubt I know the whole story and it’s naturally biased towards the events I was present for. SCLENDS was started by many people, library directors, circ managers, systems librarians and more. I worked with most of them but some only tangentially. No single person was present for every conversation and no person could know the whole story. And since I’ve admitted that this will be an incomplete telling I will also offer that I’m going to try to keep it brief. The story begins properly with the development of writing in ancient Mesopotamia and Egypt … just kidding.
In 2008 I was the Systems Librarian in Florence County, South Carolina. The library’s director, Ray McBride, and I had been deeply involved in the process of re-evaluating our technology plan. One thing we were not concerned about was our ILS. We were very happy Horizon users and had assumed that we would upgrade to Horizon 8 when it was released. It had already been delayed but why would we consider other options? Going out for an RFP is a process to be avoided like an invasive unnecessary medical procedure. Plus, we were happy with Horizon, it was user friendly, it fit our needs and was stable. Sure, it had gotten a little long in the tooth but the upgrade would give it the refresh it needed.
Then one day I was reading through my daily mail and there was a correspondence from Sirsi-Dynix. Horizon 8, Rome, was being canceled. Instead they would take the modern code base of their other product and merge it with the user friendliness of Horizon and like tunes being played together it would be Symphony. It was the kind of over the top marketing speak that made it clear they were trying to make users feel positive about news they knew we would be unhappy with. They would have been right about the unhappy part.
Fast forward and we had a meeting. I had compiled a list of possible ILSes we could upgrade to. Polaris was a strong contender. We seriously looked at Symphony, hoping for the potential of an easy migration. There were others we dismissed due to expense or lack of features. There might have been another we considered that I can’t remember now. And I threw Evergreen onto the stack for consideration.
Why did I suggest Evergreen? Florence was an almost pure Windows server environment and this was a radical departure. I didn’t try to convert the Florence environment to Linux despite my preferences because with the staff limitations the library had and applications they had invested in running within a Windows environment, Microsoft made sense. Migrating to a mission critical application on Linux was a big departure. But, when I looked at the growth of open source, what I saw happening in the Evergreen community and my own opinions about the relationship between open source and library philosophies I was of the conviction that we should consider it. Not go to it, just consider it. Frankly, with my time limitations an easy upgrade to Symphony sounded pretty good to me.
We formed a committee of public service staff and administrators. We invited in representatives from companies to talk about their ILSes. Evergreen was open source so I distributed a fact sheet. We had reps from Polaris and SirsiDynix come in. We talked to other libraries. One library referred to recent updates to Symphony in …. unflattering terms and told us they were migrating to Polaris as soon as they could. Others were only slightly kinder. Polaris looked good but didn’t blow us away. A Sirsi representative made it clear that migrating to Symphony would not be like an upgrade and there was Horizon functionality that did not have one for one parity in Symphony.
Discussions were lively but in the end we selected an ILS: Evergreen. At that point Evergreen was about version 1.2 and rough. As we talked about it one theme came up again and again. We believed that whatever shortcomings Evergreen had at that point in mid 2008 that it was the right long term choice for us. We believed that in time it would match and exceed the other options we had to pick from. We also wanted a choice that we felt would last us ten years. I think it was Ray who said later that this would be the last ILS a library would ever need to migrate to. He may well be proven right, only time will tell.
It can be strange what you remember. It was a Thursday afternoon in November that I was having coffee with Ray. We were discussing Evergreen and forming our plans for the migration. One of my concerns was the long term support, especially if I left. We began discussing approaching an external company for support of our servers. That would give me more time to spend in the community and support regardless of staff turnover. As we looked we also began to discuss moving to remote hosting and increasingly liked the idea though it meant moving nearly all technical management external to the library, not something we had traditionally done. However, while we had put a lot of value on internal staff management of technology we also had increasing needs without an increasing budget so going with a remote hosting option made sense.
All of this, especially the budget concerns, was in my head when I threw out another idea. In one sense, this was the start of SCLENDS. What if we invited others a to join us to start a consortium and reduce costs? Ray liked the idea and threw the idea out to the South Carolina library director’s listserv. From there I become a peripheral part of the story until January. During that time in the periphery I was aware that the offer was expressed and interest returned. I was tasked with inviting a vendor who could run servers for us. The clear option was Equinox, having been founded by the original developers and administrators of Evergreen at Georgia PINES. Additionally, they had a lot of experience with startup consortiums so they would understand what we were embarking on.
December passed and January of 2009 arrived. I found myself in the large meeting room at the Florence Library. The interested libraries were arriving. Eleven libraries in total attended that meeting, interested in sharing costs and materials in a new consortium. That meeting brought together not only the directors but systems administrators and circulation managers of the libraries.
Eleven libraries were present and ten of them went on to form SCLENDS. Honestly, that day was a blur of faces and voices. One person whose name I don’t hear mentioned much in connection to SCLENDS is Catherine Buck-Morgan and it should be. Although I don’t know this for fact I suspect she is the one who created the name (had it been left to me I probably would have chosen something tree related). Additionally, she was a critical part of this happening. It may have happened without her involvement, it may not have, I don’t know. I do know it wouldn’t have happened as quickly and the way that it did.
Catherine was the head of IT at the State Library and closely involved with the distribution of LSTA money in the state. I later discovered that she had already written a concept paper for creating a resource sharing consortium in South Carolina. I don’t believe her idea was inherently based on open source but she did cite PINES as an example of what she was thinking of in terms of resource sharing. Her idea hadn’t been circulated outside the State Library but this had dovetailed with it perfectly. She was critical to getting us LSTA funding to kickstart the migrations.
SCLENDS would quickly move over to a self sufficient model independent of LSTA and State Library money but those funds paid for the first two years of hosting and many of the migration expenses over two fiscal years that included our first three waves of libraries. Partial funds also helped one later wave.
Honestly, I thought the idea would be a much tougher sell than it was. Eleven libraries attended that first meeting and I had imagined half would back out. In the end only one, Greenville County, chose not to join SCLENDS, objecting to sharing their videos with other libraries. Most of these discussions happened in January and early February. Then we got to work. In less than five months, driven in large part by a window of opportunity for grant monies, we went from a first meeting to go live.
Wave one went live in late May 2009 and consisted of the State Library itself, the Union County Library and Beaufort County Library System. I later went to the State Library myself for a tenure at the IT Director there where I ironically ended up working with the Union County director, Nancy Rosenwald. We had both taken positions there and had offices next to each other. I really enjoyed working with her both within SCLENDS and at the State Library. She also had good taste in tea. Beaufort had one of the most dramatic go live days when a construction crew cut their fiber line during the first day of go live. The story the local newspaper printed was essentially “Evergreen Fails” instead of “No Internet at Library.” I understand they later printed a retraction in small print in an obscure text box. Ray McBride after a stint as a museum director even took over the library system there proving that it is a very small world. I discovered that Beaufort had been investigating Evergreen in 2008 as well though not as far along nor with plans as definite as our’s in Florence.
Wave 2 was in October of 2009 and included Fairfield County, Dorchester County, Chesterfield County and Calhoun County. Frank Bruno of Dorchester I think I fought with as much as I agreed with. I remember his staff loved him because he supported them. He passed away last year and the world is poorer for losing him. Drusilla Carter left Chesterfield for Virginia where she helped start talks that may have led to their own Evergreen consortium and eventually landed in Conneticut where she is a part of Bibliomation, another Evergreen consortium. Kristen Simensen is still at the Calhoun County library and fighting the good fight. Sarah McMaster of Fairfield retired right around the same time I left South Carolina and her last SCLENDS meeting was, I believe, my last one as well. Aside from personally liking Sarah as a person, professionally, there isn’t a library in the country that would not benefit from having a copy of Sarah on staff.
Finally wave three went live in December and included my own library Florence. Shasta Brewer of the York County library became a close co-worker of mine over those months and became the leader of the early cataloging discussions. Faith Line of Anderson had pervious consortium start up experience and continued to long be a voice that people looked to leadership on the executive board. I believe it was Faith her that suggested the creation of the working groups to aid in the migration that eventually became the main functional staff bodies of the consortium. Even when there were later attempts to expand or redefine them the original ones persisted in being the main ones. In Florence, Ray served as the chair man of the board during the infancy of the consortium and after leaving came back to another SCLENDS library.
And there were others – other staff, other stories and later other libraries which brought yet more staff and stories. SCLENDS grew over the next few years. But those stories belong in other years. I may or may not write about those stories some day but I think they’re better documented so there is probably little need. Did I leave some things out? Sure. The Thanksgiving Day Massacre. The Networked Man Incident. The Impossible Script Mystery. Probably others as well, and they make for fun stories, but aren’t core to the history I think.