Dear Yann,
With respect to the integration question you raise in point two, you are
correct at one level but I believe that there is another way to look at this
issue. Library ILS software is
vertically integrated but only within
a particular vendor
application. The turnkey aspect of
such systems seems the most
convenient solution for the
staff user who needs a complete toolkit to
perform a range of functions. But as someone who has moved MARC data
between a number of utilities and local systems, this has been mixed
blessing in three
respects.
1. Sometimes the vendor does
not provide all of the components of the system and we have to patch in another
solution to get needed functionality. Suddenly, the operation
is no longer seamless.
2. The tools the vendor
provides are not "best of breed' and you find yourself stuck with a serials
control or media booking or some other subsystem that has less than optimal
functionality. The selection process of ILS software is often a matter of
compromise in selecting the feature set that most closely, but seldom in all
ways, meets ones requirements.
3. Vendors implement solutions
that are not standards-based or that are not able to reflect standards upon data
output. This can leave you in a bind when you want to share data or
migrate to a new application. My colleagues have many stories
to share on this subject during our recent implementation of a new
ILS.
The lack of full vertical integration that
you see in EAD is, in fact, the norm in web delivery of information.
HTML editors, database software, and web servers all must be synchronized, even
if the details of that process are hidden from the various participants in the
operation. Most applications seem to require the bolting of some, if
not all, of the components together. This approach lets
you employ the best and most standardized solution for each part. One
trades control for a certain level of complexity.
Michael