My NCSU colleague, Tito Sierra, said to me yesterday, “Haven’t the last two years of the library system industry been more interesting that the previous 10?” I could not disagree. And today’s news not only adds credence to that theory, it could make the next 5–10 years continually interesting.
SirisDynix announced today that it would begin developing a “holistic platform” for its automation system, code-named Rome. Okay, what does this mean? First some facts, and then some conjecture and editorial.
SirsiDynix will unveil that we are blending the strengths and best features of Unicorn, Horizon/Corinthian, and other solutions to create a new, versatile technology platform to serve 21st-century libraries and consortia. Code-named “Rome,” this platform goes beyond the traditional integrated library system to encompass the full range of technology building blocks for managing library operations and resources, while providing meaningful user experiences to your information consumers.
—from SirsiDynix message to customers
It means that Horizon 7.3 and Unicorn 3.1 will be the last versions of those products from SirsiDynix. Horizon 8 / Corinthian and Unicorn 3.2 will not be released. That said, Rome will be built (not in a day, I venture) upon the architecture of the Unicorn system. End-of-life (EOL) has not been announced for these two platforms. Only Multilis (June 30, 2007) and DRA Classic (February 8, 2009) have been given EOL status.
I understand this decision from a business standpoint, even if I might not agree with it 100%. Choosing a single platform will save SirsiDynix a lot in R&D costs. Not supporting 14 different OPACs (a little hyperbole). Think of main things like development, sales, implementation, and support. Who wants to multiply all of those thing by two?
The problem, to me, is that all the descriptions of Rome sound a lot like the descriptions of Horizon 8.0. Heck, they sound a little like Taos, for those who want to go back even farther. Rumors abound about the instability of Horizon 8. I cannot speak to those rumors. So, I think the decision about Rome is essentially a “time to market” decision. It’s a choice between a sexy platform that is unstable and a stable platform that is unsexy. I used to be happy when stability won the day; now my own ideals are staring me in the face.
SirsiDynix sounds sincere about its direction. I spoke to several principals, including Martin Taylor, John Dickson, Berit Nelson, and Tom Gates—they seem confident in this road to Rome. I am going to reserve the bulk of my judgment for a few months. But I also wonder how this might be playing out in Rome, Georgia.
As many a blogger and library developer has said, the release of the open source Evergreen System by the Georgia PINES group has the potential to change the library automation landscape. “Salivating” was the word I used with a colleague to describe the open source community’s reaction to this plan. That salivation might not start until the news of Horizon 8’s death sinks in, but I think it will hit them sooner rather than later.
With over 4,000 Horizon libraries and 4,000 Unicorn libraries (libraries, not server sites), this news means the potential for hundreds of migrations over the next 3-6 years. I say as a guess based on SirsiDynix’s policy of actively supporting a given product in its current release and two prior releases.
The Horizon customers will be the most interesting ones to watch. SirsiDynix would argue that they were facing an upgrade and change of (at least staff) interfaces no matter what. That argument could also be made by libraries for choosing another vendor. Since there will undoubtedly be financial incentive to stay with SirsiDynix, this will be a difficult decision for many libraries. But it is also why I predict salivation from the open source providers, who will argue that the pain of migration is equal but that the value of openness is priceless. I am sure that SirsiDynix’s competition is ready to pounce as well.
Final thoughts (for today)
It’s likely that Vista’s purchase price for SirsiDynix was roughly twice their annual revenues, or over $200M (the rumor I heard actually put the figure higher than that). As a $1B equity firm, Vista is investing a sizable chunk of its money in this decision. That’s commitment. We’ll see if it pays off.
[This post originally appeared as part of American Libraries’ Hectic Pace Blog and is archived here.]
In response to Rome, maybe the PINES folks should rebrand Evergreen as “Visigoth”…
It will certainly require full retraining for Horizon customers and, by the sounds of it, hardware purchased for 8.0 may not be suitable for Rome (due to the limited platform and database options – i.e. Oracle).
For all intents and purposes, I’d argue that it is an entirely new product for Horizon sites.
We ranked Unicorn as one of the least suitable ILSs for our needs when we went out to tender recently — having our existing ILS scrapped and then being forced into having a hastily upgraded Unicorn system doesn’t strike me as being “the best of both worlds”! 😉
What makes you think there will be financial incentives to sticking with SD? I would think the opposite, especially for Horizon (rather than Unicorn) customers, a migration to Evergreen will likely be MORE AFFORDABLE than a migration to “Rome”. What am I missing?
My fault for not making this more clear in my first message. My understanding from SirsiDynix is that Rome is an upgrade path from Unicorn and Horizon, not an entirely new product that will require purchasing the software.
Point taken, Dave. And I will admit to not knowing enough about Horizon 8.0 in relation to 7.3 to determine how much of an “entirely new product” it would have represented.
My sympathy (as much as I have left) for vendors lies in balancing the hue and cry for something new and innovative with the demand that it also be free and easy.
“new and innovative with the demand that it also be free and easy.”
On your best day you can pick two of the four – new, innovative, free and/or easy.
On your worst day you can only pick one.
Horizon 8.x was a from-the-ground-up new system, and that is partly what attracted us to it when we selected it three years ago. For that reason, one of my concerns with SD’s latest announcement is the fact they’re building Rome on the architecture of a 20-year old ILS.
I worked for a competing vendor several years ago, and this is the option they chose when they created their second-generation product. As one of the people who tested the software, I can tell you that this solved none of the problems with the old system and compounded the issues on the new one. It was like ripping the body off of a 20-year old Ford Escort and plopping a new Mercedes shell on…it looked nice, but still wouldn’t run properly if your life depended on it! 😉
Granted, this isn’t the same corporation that I worked for…but I honestly can’t see how this won’t cause complications for all of the SirsiDynix customers in the future.
I think most of the staff at Huddersfield who’d seen Horizon 8.0 said it felt familiar. The interface was much improved, there were lots of cool new features, but the core functionality was the same and we felt we’d only need a small amount of retraining. From a purely technical point of view, 7.3 to 8.0 would have been a migration, but from a staff point of view it would have been an upgrade.
It’s also worth sparing a thought for the dozens of libraries who have passed the point of no return in their own migration to 8.0 (both new customers and former Dynix or Horizon sites) — they are the ones who have invested hundreds of staff hours and tens of thousands of dollars in a product that the company repeatedly promised that it would continue to develop.
Apart from one brave member of SirsiDynix staff who stuck his head above the parapet to declare that the announcement was an “exciting opportunity”, none of the reaction on the Horizon-L mailing list has been positive. And “yes”, there’s a lot of discussion about Evergreen going on in there!
They are just dumping the completely written Java based Horizon 8.0 system for the old Sirsi system. It is “stable” because they haven’t added much to it. It’s my understanding that the unicode conversion of the db in their last upgrade was unsuccessful and not implemented by Sirsi libraries. It seems to me I heard a quote from someone about putting lipstick on a pig, and wasn’t that pig this system? Or am I thinking of something else?
I think it is a mistake to think that Rome will be just like Unicorn. As it was told to me in a phone conversation yesterday, “if you understand Unicorn administration, then Rome will look familiar.” Things that will most likely be changing is Unicorn’s policy structure, the API (with an interactive interface), and a recognition that libraries don’t need an entire system, so maybe more standalone modules.
Question: Which of the two platforms allowed easier access to YOUR underlying data?
My experience had always been Horizon, though I have no idea about 8.x.
“It seems to me I heard a quote from someone about putting lipstick on a pig, and wasn’t that pig this system?”
The quote that I saw recently with this reference was from someone at a Horizon library who made the comment in reference to improving their OPAC!
This is unsettling news having just had a demo of Horizon last week as a candidate for our first ILS. We were told that Unicorn and Horizon were on “separate development paths” and that there were “no plans to favor one system over the other,” but that there were some “fusion” products in production. I don’t know that this is really a harder time than any other to select an ILS, but it’s difficult to put credence in what I’m told by a sales rep about system longevity. All I know is that I want a system that works for us and I don’t want to see a forced migration within 6 months of implementation.
I’m sure the decision not to develop Horizon any further was not taken overnight. I just wonder how much was known prior to us signing off our migration from Dynix to Horizon last month… we were certainly at that time “too far gone” to contemplate any other LMS and we are now back to the same position as when we were on Dynix – no development there either but it’s cost us thousands for that position this time! Always had a cautious feeling about the SirsiDynix merge from the start.