RIP: Horizon

The initial announcement has now gone out to SirsiDynix customers — the all singing, all dancing Horizon 8.0 will not receive a general release.
Instead SirsiDynix will concentrate their efforts on developing another new system (codenamed “Rome”) which will be built on the Unicorn architecture.

12 thoughts on “RIP: Horizon”

  1. I guess all our jokes about Horizon 8 being aptly named because although we could see it, it was forever just beyond our reach ended up being a little to close to the truth.

  2. It’s been interesting following the heated discussions on the Horizon-L mailing list. Amidst all the anger and frustration, quite a few libraries are seriously talking about a move to Evergreen.
    If you get chance to, it’s worth comparing the wording of the two emails that were released (one to Dynix/Horizon sites, the other to Unicorn sites).
    I must admit I’ve been disappointed by some of the comments made by a few Unicorn bloggers. Rome has come at a very high price to many Dynix and Horizon sites (especially those further down the 8.0 migration than we were at Huddersfield) and is far from being the “best of both worlds” that SD are claiming.

  3. With all the talk about the news, why don’t libraries really give other vendors a chance to show what they do. There are a couple of vendors that have tried to listen to the public libraries and encourage insights yet they are often discarded because they are not SirsiDynix.

  4. Kim — I think every single Dynix/Horizon site out there is already thinking or already doing that!
    We went out to full tender just over a year ago and fully evaluated all of the major systems available in the UK. For our needs, Horizon 8.0 was quite simply the best system for us…

  5. Dave –
    Horizon 8.0 was not the best system out there for you… the unfulfilled promise of Horizon 8.0 was better than the unfulfilled promises from the other vendors you evaluated. That and the (somewhat modern, yet over-integrated) UI which did demo well.
    The cool thing about open source projects (like Evergreen) is that you pretty much know what you get and no more. Unfortunately, it will take MUCH work for that particular bit of code to begin to meet the needs of even a moderately-sized library.
    NOW, if a number of libraries could form a consortium whose sole reason for existing is to actively enhance and make Evergreen become what Horizon was promised to be… THAT, my good man would be an incredible benefit for all libraries. The key is not so much maximizing member investment in terms of saving money, but maximizing member investment in enhancing and supporting the system. A software co-op could be a beautiful thing.

  6. Another Dave,
    What do you base this statement on?
    Unfortunately, it will take MUCH work for that particular bit of code to begin to meet the needs of even a moderately-sized library.

  7. From
    * From a library perspective, what does Evergreen do? What modules or components are available?
    Evergreen currently has modules for circulation, cataloging, web catalog, and statistical reporting. Evergreen also supports the SIP2 protocol for self-check and Internet/computer access control.
    * What does it not do?
    Evergreen’s Acquisitions and Serials modules are currently under joint development with the University of Windsor. Other features on our roadmap include telephony and credit card support.
    * When will it do these things?
    We’re hoping for early to mid-year 2008 for the Acquisitions and Serials modules.
    Based on their CVS check-in logs there are 3, perhaps 4 active developers on the project. Don’t get me wrong, they are on the right path, but it’s a long path indeed.
    SirsiDynix (and others) are not threatened by Evergreen because while it seems very romantic to drop your vendor and jump on the open source bandwagon, that move takes incredible resources (and more fortitude than most libraries can muster). Such a decision moves many of the concerns that were the responsibility of the vendor directly in-house. No longer can you blame the vendor when explaining the system shortcomings to your board of directors. Even if Evergreen had half of the PROVEN functionality of Unicorn the training/staffing is not something that most libraries want (or can) to take-on.
    Also, Evergreen has (to my understanding) not been TRULY scaled nor load tested (this was the real Achille’s heel of Horizon). Not to mention another Sirsi project killed before it could be really good… Taos! That being said, few new library products undergo TRUE load and scale testing. That part of the project can take many weeks into months to address appropriately. Some projects are simply luckier than others in this area. Sink or swim.
    PINES will, with luck, take the Evergreen project slowly and nuture it like a farmer sowing seed and patiently waiting for the yield. Alternately, an injection of resource backed by realistic expectations and a will to “do it right at the right rate” could make the tree grow faster.
    Vista can not afford the time as their investors expect certain results within certain (relatively short) timescales. Bummer that – for libraries AND their patrons.
    I do believe that the remaining core of leadership at SirsiDynix is comitted to doing the right things for their customers and more importantly, for their customers’ patrons. Sadly, they are not 100% sure of what that right thing is and thusly approach the design and development of solutions with less than 100% comitted efforts. This could change, I believe, but I seriously doubt it.
    Back to the beginning… if you KNOW what the right thing to do is, dedicate some resources to the only place where that input can truly become a reality. That project is Evergreen. Just be willing to water it, to nuture it, and to WAIT for the bloom.
    Because trust me… you WILL be hearing things like “Rome wasn’t built in a day.” from the good folks remaining at SirsiDynix.

  8. Evergreen runs a consortium that the architecture of the previous commercial system was completely unable to handle. The comments above about scalability are simply not supported by the evidence, Evergreen’s levels of throughput have been solid *in production* for a common circulation system that serves 250+ libraries. It has succeeded, using commodity hardware, at a fraction of the cost and many times the stability of its predecessor. There are metrics for Open Source projects and organizations like NASA put OSS applications on mission-critical systems. The satisfaction of Evergreen’s current users alone is a strong indication that this path is worth exploring, and whatever was “proven” in the previous system in terms of functionality, no one seems to be missing it much in the State of Georgia right now.

Comments are closed.