[PLUG] linux for elementary classroom/lab

[Daniel Pittman]

Nathan Young <nathan at ncyoung.com> writes:

G'day Nathan.

> Has anyone used linux on a large scale among lower grades in the portland
> area? I have a few specific questions areas of interest.

Nope.  On the other hand, I can pass on some of what was found down here in
Australia when folks tried to do the same sort of thing, some lessons from
which might be valuable to your efforts too.


> Do you know how PPS interacts with linux? I've been told that if the
> district buys computers for the school then they also support those systems
> to a degree, while if systems are donated or bought using PTA funds or any
> other source, the school is on it's own to install/configure/maintain.

One thing to watch out for here: in Australia the license terms that the
schools ended up with (usually through a state-wide contract) for Microsoft
software were based on the number of machines in the place.  So, it often
ended up that they would pay for a Windows volume license even if they bought
in a Mac machine or ran Linux on the system.

It sounds like you have that covered here, but it might be worth
double-checking that PPS will not continue to bill them for "supporting" the
Linux machines, even if they don't lift a finger to actually do anything.

> I've also been told that the support forthcoming is not so incredible that
> it forms a strong incentive to stay with PPS provided systems (in other
> words the bar is pretty low and likely to be pole vaulted by one or two
> involved and knowledgable parent volenteers and a few easy to come by
> corporate donation program grants).

One of the hard lessons here was that this support could easily be beaten by a
small number of motivated, Linux-skilled folks, but that it was *really* hard
to get those Linux folks available on the timetable the school staff needed.

Being able to do a better job was useless if we couldn't deliver it on the
spot, when the machine was not working, during the school day.  You might need
to account for that when you think about what volunteer efforts apply.


One other thing to watch out for: here we ran into a surprising degree of
outright hostility from the folks who did the regular support stuff, even to
adding a new system to the network.

My guess is that they were basically just change-adverse, and saw anything
added as a risk that they might be asked to do more[1] even though they never
touched the machine.

For example, we had semi-regular objections that because the Linux systems
were not running the "standard" anti-virus software they would be a much
greater risk than Windows machines would - and because the department didn't
have the skills to do a security evaluation they couldn't certify anything we
did for use in the schools, because they couldn't know if it was secure or
not.


> ok second question: anyone tried edbuntu.etc?

Nope.

> Or is it better to install a recent distro and LTSP and configure from
> scratch?

This one I can answer, though: absolutely, without question, not.  There is an
enormously huge amount of hidden work in getting this sort of stuff working
smoothly, and you absolutely do *not* want to add that to the already huge
task of getting this system into the school and supporting it.

As a guide, at a previous job we quoted close to a year for producing a
locked-down environment that ran a dozen applications and had *no* persistent
storage other than email, in an LTSP-like environment, for use in a prison
situation.  (That was a fair estimate of actual effort, too.)

In that we were starting with the LTSP equivalent, and experience building
these things.  While there were some bits that would be easier in your case,
other bits would probably be more involved, so I wouldn't be surprised to
see a similar investment of effort required.

[...]

> ok third question in this rambling chain: Anyone done this on one or two
> hours of time investment a week?

I don't know, but based on commercial experience delivering similar
solutions, I wouldn't want to bet that it could be done in that sort of
timeframe.

Well, maybe that much *support* time after release, and when it was bedded
down, but you would have to budget for more support overhead every upgrade.

A couple of hours a week would hardly cover hardware problems in a large scale
environment, by the way - those kids can be ... challenging for the machines. :)

> I've got 6-8th graders, is it insane to start contemplating a student run
> corner of the lab (or just insane to think that would save time)?

Just the later, though at that age I don't know I would trust the judgement of
the kids if you gave them admin rights.  It ... opens the door to potential
abuse that I don't know I would want a hand in.


Anyway, I don't want to discourage you: I think this is an absolutely great
thing, and would love to see it happen.  I just don't want to see you assume
it is easy and find out the hard way just how much work it really is.

Regards,
        Daniel

Footnotes: 
[1]  Generously, I could say they were worried that they didn't have time to
     do everything they already had to do in a day.  More cynically ... well,
     you can probably fill it in. :)

-- 
✣ Daniel Pittman            ✉ daniel at rimspace.net            ☎ +61 401 155 707
               ♽ made with 100 percent post-consumer electrons