[WEB4LIB] Re: open source and librarianship

[Edward Wigg]

At 08:31 PM 1/29/2000 -0800, Jeremy Frumkin
<jfrumkin at bird.library.arizona.edu> wrote:
>
>Your perception is my perception, and a lot of other people's. The current
>economic model that most library software vendors follow promotes
>proprietary mechanisms, not open standards. The few current standards that
>are out there (MARC, Z39.50, perhaps Dublin Core and RDF, although these
>aren't yet widely implemented) aren't enough.
>....

>
>Exactly! Modularity is key to building the next generation of ILS. 
>....

The economic model that automation vendors follow is largely created by the
market, if you want modularity you have to persuade enough libraries to
abandon the current notion of an ILS; given the way most libraries generate
RFPs this might be tough.

RFPs usually require the all singing all dancing system, with its parts
tightly coordinated and all working seamlessly together (what they GET is a
often a different matter :-). External interfaces may be included in the
specification but they tend to fall into the z39.50/ILL/EDI mold and are
essentially add-ons. Achieving modularity presumes that you can either buy
_all_ the parts of such a system right away (and that developers have
produced them on spec.), or that some libraries are willing to buy some
modular parts now and hope that the rest of the pieces can be filled in
later (initially a NON integrated system).

In the current environment why would a vendor create a complete modular
system? It would probably seem to them like willingly wearing a 'kick me'
sign! Over time they can probably be slowly moved in the right direction,
but only if a significant number of their clients (especially big new
accounts) require it; it's going to be tough sledding to get them there. 

In addition to changing the pattern of development you also need to change
their minds about licensing and maintenance. I don't know how all vendors
do it, but I know the policy of at least one is that if you buy one part of
their system (say the serials program) then drop it, if you change your
mind and want it back you have to pay all the maintenance for the period
that you didn't use it, which quickly mounts up to more than the original
purchase price. This sort of practise, while understandable from their
perspective, does nothing to encourage modularity.

We should remember the roots of the open source movement: the part that is
a crusade (fsf/gnu),  the hobbyist and groups of interested people, amateur
and professional,  with spare time (the Linux kernel), and the people in
academia who have managed get paid or earn credit while producing software
and who were then generous enough not to try and hoard the result. The fact
that you can go out and get a CD ROM and install an open source UNIX look
alike is the end result of fifteen years, or more, of largely piecemeal
effort. The commercial open sourcers are a much later development who are
unlikely to have been here were it not for the others having created the
environment first. Library automation software is such a tight world that I
don't think its current environment is likely to encourage existing vendors
to change, and the barriers to entry for new open source vendors are high.

There are close parallels between open source and libraries, even down to a
similar confusion about the varying meanings of the word free. As the FSF
people say: "you should think of 'free speech', not 'free beer.'" 

Edward