[balug-talk] Linux computers for the police! With Link

[Thomas Leavitt]

Greg Herlein wrote:
>> It would indeed be a major task to retool a police department with Linux--
>>     
>
> The cost actually would have little to do with Linux - and a lot more
> to do with the applications any legacy organization has in place.  To
> port/rewrite/modernize those applications for linux is a MAJOR amount
> of work - not to mention the training costs associated with the
> smallest change (for the cops as well as the support staff).
>
> This thread is interesting but ultimately I fear nothing will come of
> it.  The cost of re-tooling an organization has so little to do with
> the OS and so much more to do with the work flow, the custom
> applications, the support infrastructure, and the momentum of
> sloooooooowwwwww changes...  and frankly, all of the current
> infrastructure is already paid for.
>
> Is linux better? Obviously.  Would it make fiduciary sense to change
> the City PD to Linux?  Probably not.
>   
But isn't this all work that is likely to have to be done, anyway? How
compatible with Windows Vista are custom applications written for
Windows 95 likely to be... which isn't really even the question. The
real question is, how likely are applications written for Windows 95,
and likely written nearly a decade ago (or more, possibly), how likely
are they to be appropriate for use in a modernized SFPD IT architecture...

Really, what we're asking here is not Linux or Vista, but what is the
opportunity cost of not modernizing their IT architecture... what could
they be doing, now, that they can't do, because they're handicapped by
having to cope with a 1995 era IT architecture and operating system (and
associated hardware)? What risks are they exposed to, at a tactical, and
operational level? I would presume these machines, for instance, aren't
being allowed to connect to the Internet, given the vast number of
unpatched security holes and issues in Windows 95 and associated
applications... as someone mentioned, what are the ongoing maintenance
costs of being restricted to hardware that supports Windows 95... are
there operational integration issues with the rest of the city's
systems? With other law enforcement agencies? With newer systems inside
the SFPD?

There are questions every organization faces as it evaluates its IT
architecture and asset base, and evaluates whether or not it is time to
upgrade (of course, some organizations "solve" the question by simply
upgrading everything, automatically, every three years - but I don't
think public institutions have that luxury)...

I'm pretty sure that a rational evaluation would lead just about anyone
looking at the situation to conclude that change is appropriate - and
given that, we shouldn't be talking about the costs of converting to
Linux from Windows 95, but the comparative costs and advantages of
converting to Linux vs. Windows XP / Server 2003 or Windows Vista /
"Longhorn"... and Office 2007 vs. OpenOffice 2.x, etc.

But really, that's not the way to approach it either - what really
should be done is a full scale evaluation of needs, goals, etc., and
then an evaluation of what solutions are available on each platform, and
open source vs. proprietary systems, etc. The end result could very well
be that a heterogenous network, mixing Windows and Linux, on the desktop
and server, is appropriate... open source applications on Linux and
Windows, and proprietary applications on Windows (and maybe Linux as
well). Who knows? I doubt any of us are in a position to speak to the
issues facing the SFPD.

What I am certain of that a compelling case could be made for an orderly
transition off a Windows 95 based IT architecture to something
reasonably modern... as long as such a migration is driven by specified
needs and thoughtful process (re)engineering. Simply pulling out their
existing systems, and doing a drop in replacement with Linux or Windows
whatever, might save some money, but would qualify as a huge missed
opportunity, and probably not be terribly economically or
organizationally beneficial (other than perhaps, after the bugs are
ironed out, resulting in less work for the IT department).

Thomas