This session consisted of two different
presentations. The first presentation was presented by long-time
Computers in Libraries veteran Marshall Breeding, Director of
Innovative Technologies and Research at Vanderbilt University. Mr.
Breeding provided a good, balanced overview of the library automation
system scene looking at the differences between the proprietary and
the open source offerings and looking at the satisfaction that
different libraries have with the different systems.
First, Mr. Breeding provided some
context pointing out the move (at least among academic libraries)
away from conventional Integrated Library Systems (ILS) to broader
Library Services Platforms as well as several of the reasons and
consequences of this kind of move.
Next he showed some of the results of
the most recent survey he has done of libraries and pointed out
several trends. First of these was that although open source systems
are trending in popularity, it has been a gradual trend, and is not
any kind of huge rush away from proprietary systems.
Interestingly of the libraries
surveyed, only the ones that are currently running open source
systems are really enthusiastic as a whole about open source
solutions. The users of the proprietary systems, by and large, are
not expressing much excitement about the promise of open source
solutions.
Of all of those using systems of any
kind, it's the quality of support that makes the libraries satisfied
or dissatisfied with the systems. The libraries that are happiest
with an open source solution are those that installed it themselves,
and consequently have considerable internal knowledge of the system
as a result. However libraries running the exact same system that
are getting support from a company that they find inadequate to the
task are unhappy with the system.
Mr. Breeding then looked at the profit
models for both open source and proprietary systems pointing out that
generally open source is not a nonprofit endeavor, but a commercial
endeavor. The open source support companies mainly use a different
revenue model, charging for data conversion, installation,
configuration, training, support and hosting. A proprietary system
might cover many of these costs in a license fee and not break them
out as much. Proprietary systems do seem to have a much larger
cohort of full-time developers than open source vendors, although
most open source vendors have a respectable 15 or so full time
developers on staff.
Finally Mr. Breeding pointed out that
from a flexibility and usefulness standpoint, an open API was much
more powerful than just having the system written in open code. If
the system was not written with hooks that can be taken advantage of
and it's open source, for many library purposes it will not be as
open and useful as a proprietary system that does have an open and
available API that can be used for more advanced projects.
Mr. Breeding was followed by Irene
McDermott of Crowel Public Library in San Marino, California. In Ms.
McDermott's presentation she described their rapid transition from a
proprietary system to an open source system (in this case from Dynix
Horizon to Koha).
Her library (a municipal library)
bought Dynix Horizon in 2006 a few months before Sirsi bought Dynix,
and then killed off Horizon. This did not please the library (which
had decided against Sirsi Unicorn) and when, in early 2011 the city
of Crowel decided it really needed to cut the budget, the library was
charged with finding a cheaper alternative before August of that
year. The library decided initially on the more expensive of a
couple options (Polaris), but the city, wanting to save as much money
as possible, told the library to go with the cheapest option, Koha as
supported by LibLime.
Ms. McDermott described the rather
breakneck pace at which they made the transition and then pointed out
several features of the system that she and her staff have found
frustrating or annoying. Overall, however, she seemed satisfied with
how everything turned out in that the library was able to the
transition and keep functioning without having to suffer staff cuts.
In her presentation Ms. McDermott
mentioned that Koha as supported by LibLime had some proprietary
portions that had been developed by LibLime. This was countered as
inaccurate by a representative from LibLime who stated that they were
opening that code. An open source advocate in the audience argued
that this really wasn't open until it was out, starting a brief
argument about what makes something open source, an argument that for
better or worse didn't really reflect well on the open source
community (I say this as someone typing this in an open source
application running an open source operating system).
Although this session wasn't
immediately useful to me, it did provide some insight into the state
of library automation systems at the present time and was definitely
worth my while.
No comments:
Post a Comment