[Web4lib] RE: Open source support models

Tue Jul 15 15:26:45 EDT 2008

I agree with everything David Walker just said. I would however add that
as part on considering using a specific Open Source product, and
important thing to look at is how smoothly the upgrade process runs. 

In selecting a CMS for our Library, one of the Open Source products we
seriously looked at was Drupal. In the end, however, my primary reason
for going against Drupal was the lackluster upgrade process. Because
much of the functionality of Drupal is contained in modules written by
third party contributors, choosing a particular module for key
functionality for your web site ultimately limits how soon you can
upgrade to the newest version of the core Drupal system, as the API that
the modules use changes over time, and with each successive release of
Drupal, each module must be re-evaluated, and often rewritten, in order
to maintain that functionality. For that reason, there are a lot of very
old Drupal installations out there that do not get to benefit from the
new releases, because the web site the system manages relies on one or
more modules that have not been updated to the newer Drupal core.
Clearly this is not ideal. I do think that hiring a support consultant
could have helped us select modules that would likely continue to be
updated, but I don't think it would have entirely mitigated the problem.

This is not the case for all, or even most, Open Source products, but in
the case of our CMS decision, the commercial option we were looking at
provided a much more reliable path for updates, and broader options for
extending functionality, and that played a role in why we chose to pay
for the software.

With that said, we do use open source software. Even though the CMS we
purchased is commercial, the technology stack that powers it is all open
source (Linux, MySQL, Tomcat, Apache 2, etc.). 

And that brings me back to some of the initial things that started this
thread. The nice thing about Open Source is that it is not this big
monolithic thing that you have to buy-in to. In the case of our CMS, we
are running commercial software on top of Open Source. Alternately, you
could Open Source on top of commercial software. Apache / PHP / MySQL
all have Windows binaries that are easy to install. We could have run
Drupal on Windows. 

Basically, my point is this. You do not need to relearn everything you
know about computers to use Open Source technology, or abstain from
using commercial solutions when appropriate. You simply have to unlearn
the idea that something that didn't pay for doesn't have value.

- David

---
David Cloutman <dcloutman at co.marin.ca.us>
Electronic Services Librarian
Marin County Free Library 

-----Original Message-----
From: web4lib-bounces at webjunction.org
[mailto:web4lib-bounces at webjunction.org] On Behalf Of Walker, David
Sent: Tuesday, July 15, 2008 11:03 AM
To: web4lib at webjunction.org
Subject: RE: [Web4lib] RE: Open source support models

Hi Caitlin,

> If there is no clear benefit in terms
> of money (since instead you end up
> paying for support anyway, or losing
> that money in terms of time) what is
> the advantage of choosing Open Source
> over commercial software?

Maybe I can clarify a little here.

With commercial software, you end-up paying for two things: (1) a
license just so you can use the software in the first place, and (2)
support.

With open source software, you only need pay for the support (if you
choose to have support).  The software is free.

So there is a significant up-front cost savings, since you don't have to
buy a license.  Once you calculate, too, that you will receive future
enhancements and modules for the software for free, the savings of open
source over time become even more significant.

--Dave

==================
David Walker
Library Web Services Manager
California State University
http://xerxes.calstate.edu
________________________________________
From: web4lib-bounces at webjunction.org [web4lib-bounces at webjunction.org]
On Behalf Of Caitlin Nelson [cnelson at hawaii.edu]
Sent: Monday, July 14, 2008 6:34 PM
To: Thomas Dowling; web4lib at webjunction.org
Subject: [Web4lib] RE: Open source support models

Hi All,

Sorry if my emails seem naive or unstructured - I never meant to make a
point-by-point argument or say anything stunningly original, I was just
voicing a thought on a recent experience...  If you're still interested
here
are some more thoughts:

Thomas Dowling wrote:
> I don't understand: are you saying there are commercial alternatives
> where the cost of software plus the cost of support is affordable in
> your situation, but free software plus Liblime support is not
affordable?

Just to clarify: no, I wasn't saying that - I was just complaining :)
Of
course everything costs money (or time, or both) in the end.  It might
make
a difference, though, if you have IT staff who are more familiar with
one or
the other - our in-house IT staff does not support Linux at all.

Marcie Pierson wrote:
>Perhaps what she is saying is that, larger systems often have a person
or
entire staff
>dedicated to IT/ILS, whether they use open source or not. Small systems
usually have to
>have one person become the IT/ILS guru on their own, and while open
source
may have
>online groups that are happy to assist with troubleshooting and such,
that
library person still
>has to do a lot of self-training. Regardless of cost/no cost, on top of
normal library duties, a
>staff member working in a small library has to also become an expert in
some piece(s) of
>software without the benefit of support from another department (or
person).

This is more along the lines of what I was intending to say, thanks
Marcie!

Essentially here's the situation: our current "ILS" is a couple of
wicker
baskets, some quarter-sheets of paper (check-out slips), and an aging
Excel
list of inventory.  Our annual budgets is around $3500 (in a good year).
ANY system plus support is going to cost us money - probably more than
we
can afford.  We have no IT person for the library, though we do have
someone
for the entire school; I am the only librarian.  What I was hoping was
that
an Open Source ILS would work for us, given that there was a chance I
could
install it and get it running myself for free (in terms of just money).

What I ended up figuring out was that the time investment and learning
curve
for installation and setup are huge obstacles to achieving this goal.
Like
I said in my previous post: I'm decent enough with computers and I like
learning how to do new kinds of work.  But I was not familiar enough
with
command-line work, Linux environment, PERL, etc to be able to
troubleshoot
or do anything that wasn't already written in the installation guide.  I
also got fed up with installation guides and error messages that said
something like "Have your systems administrator check this" - I AM the
systems administrator in this case!

On the plus side, the VMWare images were incredibly helpful (thanks Dan)
both for Koha and Evergreen, and a great starting place for us.  It
takes
off the burden of having to do a complete install from scratch and I can
jump straight in to working with the actual ILS.  And, having gone
through
the work of installing from scratch, I am now a more capable
administrator
(becoming more so by the hour, I hope).

I thought I and my library would be a perfect combination for using an
Open
Source ILS, but in the end I was surprised at how unaccessible it turned
out
to be.  Maybe it was only a surprise to me (I should have read this list
more carefully)!  But it did leave me thinking about who exactly is
supposed
to be able to use this kind of software.  Some final thoughts for you
all:

- Who is the market for Open Source ILSs like Koha or Evergreen?
- If there is no clear benefit in terms of money (since instead you end
up
paying for support anyway, or losing that money in terms of time) what
is
the advantage of choosing Open Source over commercial software?
- What solution is there for small libraries who cannot afford support
for
either Open Source or commerial systems?
- It seems like there is a fairly steep tech gap between moving from a
double-click installation process on Windows to the installation process
for
something like Koha on Linux - is there a way to overcome this or is it
something people will just have to "deal with"?

Thanks,

Caitlin Nelson
Interim Librarian
TransPacific Hawaii College
Honolulu, HI
_______________________________________________
Web4lib mailing list
Web4lib at webjunction.org
http://lists.webjunction.org/web4lib/

_______________________________________________
Web4lib mailing list
Web4lib at webjunction.org
http://lists.webjunction.org/web4lib/

Email Disclaimer: http://www.co.marin.ca.us/nav/misc/EmailDisclaimer.cfm