We didn't set up any formal or quasi-formal test sites for SchoolTool
Calendar, because it was clear early in the year that getting it out
in time for use this school year was going to be a close run thing. 
At this point, people are testing and using SchoolTool, and while some
people may find it doesn't yet suit their needs, I don't think anyone
is sitting around feeling burnt or profoundly let down by the status
of the application, as they might have if we'd spent a lot of time
convincing them to commit to a big SchoolTool rollout with their
faculty in August as part of a big formal partnership.

We can't "play it safe" like this forever, however.  As I've discussed
on the list in the past, we have been semi-formally planning at least
one test site, at High Tech High in San Diego, CA.  Beyond that, there
are still two options:

* create as generally useful a user-configured application we have and
send it out into the ether for people to use as they will;

* focus explicitly on a small number of additional test sites, say,
four in the Northern hemisphere and four in the Southern, and focus on
a relatively customized application that will work for them, providing
a high level of on-line support, then taking the lessons from those
sites to create a truly finished, documented, generalized and locally
customizable production SIS for spring 2007.

I had been leaning toward the first, but now I'm thinking the second
course of action is better.  The parts of the system I'm worried about
at this point are the "meta-systems," like demographics, reporting and
import/export.  Ultimately, for SchoolTool to be used internationally,
these pieces need to be extremely flexible.  We talked a lot about
these systems in Vilnius and had a lot of trouble getting beyond
handwaving.  We just don't have enough experience with the direct use
of these components to create highly flexible meta-components around
them, at least without a high probability of things spinning out of
control.

To put it another way, in a pinch, Marius or Albert or Stephan could
probably write a usable basic demographic system for a given high
school in about two days.  A flexible meta-system to allow admins to
create demographic setups for their school would take a few weeks at
least, but more importantly, would have more weird edge cases and
unanticipated behaviors that would likely cause problems for months.

I'll talk more in the future about what functionality will be ready
for non-test schools by next fall.  In the meantime, I'm drafting an
announcement seeking test sites that I'll submit here for feedback in
coming days.

In the meantime, how does all that sound?

--Tom
_______________________________________________
Schooltool mailing list
[email protected]
http://lists.schooltool.org/mailman/listinfo/schooltool

Reply via email to