[K12OSN] OT: Apple Imac lab

[David Hopkins]

Rob, and everyone else that has responded.

Thanks for vote of confidence on this.  I've read all the responses
and I'll try to address them in this response.

> I think your argument above is very valid, and it should probably be your primary argument to the administration.  "I don't want to spend
> money on Macs for a video lab, because I'm afraid we'll end up spending money on Macs in areas where they're not needed".

Yep ... biggest concern.  The discussion about 10% usage for
video/audio is also valid but I learned some more details as well.
There are some new state level standards (Web 2.0)  that dictate that
students will learn video/audio packages (not sure why, it really
feels like the 'must learn MS Office debate a decade ago and the must
learn DBase approach I saw before that in Florida).  So, have to have
the capability, somehow.  IF (and it is a big IF) I can get buy-in to
make this a very restricted special case scenario the lab could
approach.  That requires that there is a line drawn though. Now, the
teacher has already added that she'll want all the packages on the
thin clients on the Macs but ... she doesn't want OpenOffice on the
Macs because 'it doesn't work right'.  Instead she wants a different
package. And we need to order the Type-to-Learn package for the Macs
as well.  So, the 'usage creep' has already started and the lab isn't
even ordered. *SIGH ...

> The solution doesn't have to be "don't buy Macs for the video lab".  The school could instead implement a "standard computer package",
> probably based on LTSP, and require special authorization (and paperwork, expert witnesses, etc) to spend more.  That could help prevent
> the slippery slope that you're concerned about, but it would also not lock out non-LTSP systems.

So far this has been done except that newer staff and adminstrative
staff in particular coming in from the other schools aren't buying
into thin clients quite as much. They acknowledge that the uptime
statistics are phenomenal (less than 4 hours of downtime this year to
this point and part of that was a power failure that took out the
school) but then they say "We can do the same with Windows as well"
which I'll admit can be done with a huge investment of resources. They
also acknowledge the ease of use and ability to sit anywhere and work
is also fantastic.  TeacherTool is very very popular as well.  Does it
work with FC10?  I have to check.

Next topic: To address the central file server, authentication and
such.  This is exactly the scenario I already use and have described
before but I'll relist it here.  I have LDAP/Samba authentication
(Thanks David Trask and Matt!) with a secondary authentication server
as well, file server with channel bonded output to handle the load,
channel bonding on my main backbone between switches (I forget the
actual type off the top of my head but amer.com switches support
linking multiple ports between switches) to further remove bandwidth
issues, heavily loaded servers on a single 24 port GigE switch which
removes switch-to-switch network traffic completely in many cases, 2
Windows Terminal Servers, separate system to run backups (BackUPPC is
EXCELLENT, thanks Les!), 6 LTSP servers running LTSP CentOS based,
separate wireless network with filtering between it and the main
network (firewalled), 200+ thin clients, 90+ printers.  It does just
work.  All of this cost less than the estimated cost of retrofitting a
single classroom to be the new Mac lab when you include the cost of
the new Macs.

> I know you're a volunteer there, so maybe you're not in a position to impose that kind of approval system.  But you should probably suggest
> it to the powers that be.

I'm lucky in this respect. I started volunteering at the school when
the school consisted of the School Director and a secretary working
out of an office at the YMCA during the planning phase.  I handled the
IT since (with support from this list).  I have a lot of 'capital' to
spend based on how things currently work. In fact, this Wednesday
there is supposed to be a meeting as part of a truly independent task
force requested by the new Governor to look at IT in the state. (The
chairwoman isn't even from the state and has no vested interests to
worry about) A representative from the task force  is coming to see
how NCS "does IT" to determine if the model should be looked at as a
possible solution state-wide.  I wrote a letter that has been sent to
this task force (if you'd like a copy, let me know). Again, there are
many people who participate in this list that deserve tremendous
amounts of credit for how successful NCS's IT solution is.  I will
absolutely be giving credit to everyone I can think of at this coming
meeting.  What will be the outcome of the meeting? I honestly don't
know. I do know that if we have to then say "but we can't meet current
state standards" using it, then it won't be as effective.

Budget issues: There is tremendous resistance within parts of the DOE
to going this route. Unfortunately for them, Delaware is looking at a
20% reduction in the state budget next year across the board and the
new Governor is very open minded to new ideas and approaches.  This is
the best chance I've seen in 8 years to really get some attention. And
the timing of the Mac lab couldn't be worse?  But, everything works
out in the end. I'm sure that will be the case here as well.

Again, thanks for all the different viewpoints and suggestions to my
question.  They have been very helpful in letting me get a better
handle on the multiple issues/concerns/approaches that exist.

Sincerely,
Dave Hopkins