Cowboy(s) Out: Oklahoma State and Stolen Software
I am sitting here in a place called
the Carroll
Street Cafe, which sits in
the heart of Cabbagetown,
a factory mill town built in the 19th Century (1885 to be exact) by the
owners of the Fulton Bag & Cotton Mill. It was commonplace back then
for mill owners to provide "shotgun housing" for employees and
their families. Officially, it was part of a mutual trust built between
management and labour. The reality is that it was a way for management
to keep labour under their domain and control. This was the ultimate business
control for management.
Compare that to the debacle taking place at Oklahoma State University (OSU).
As reported by Ed Brill on his personal blog
late last week, OSU CIO Gary Wiggins "resigned" after two of
his staff members that he brought with him from Texas Tech were also forced
to resign because
they had stolen software from Texas Tech
and deployed it on OSU Servers. This is the same staff, hand picked and
brought from Texas Tech by OSU President David Schmidly when he assumed
his new position.
This follows on the heals of a number
of heated discussions surrounding the unilateral decision by Schmidly to
have a mature, stable Notes/Domino network of ~35,000 users ripped out
and replaced with Microsoft Exchange/Outlook, and his hand-picked staff
set that in motion. As an IBM Business Partner, I have to say I carry a
bias in this decision. However, as an Business Controls professional, this
whole sequence galls me as a taxpayer to see public money wasted (in the
federal sector, it is referred to as "fraud, waste and abuse")
and thrown away because of the decision of one person. Dr. Schmidly. While
this is a common occurrence in the private sector, it is less problematic
because they are using their own money and are not accountable to taxpayers.
The recent resignations (and accompanying controversial severance payments)
point out glaring weaknesses, or the total absence, of strong internal
controls at Oklahoma State University, including but not limited to:
- Inadequate separation of duties
- Purchase of software that was not evaluated to determine if it filled
a critical business need
- The ability of two key employees to steal software from another source
and deploy it enterprise-wide without any documentation waivers or licenses
- The liability faced by OSU to now have to pay Texas Tech for software
that duplicated out-of-the box functionality in Notes/Domino, without competition.
- The award of a 35,000 seat contract for e-mail to Microsoft without providing
for adequate competition.
There is also the small but important matter of open records for government
organizations. One of the terminated employees had created an open-source
discussion forum in support of the migration, except that he conveniently
deleted any postings that were critical of the decision. It was not until
I spoke to an external affairs official at OSU and confirmed that these
postings were in fact matters of public record that they were restored
to the forum.
Of course, this does not let Texas Tech off the hook. As reported in the
University Daily:
"A June 25 report from OSU's General Counsel's Office said two employees
''more likely than not'' gained unauthorized access to Texas Tech computers
to
copy programming code used for OSU's online events calendar."
Texas Tech failed to either disable logical access controls for the individuals
when they left the employment of Texas Tech, lacked an adequate firewall
to
protect their network (a common problem for public universities who are
forced to balance between "free speech/academic freedom" and
security), lacked
an effective intrusion detection system, did not have adequate security
on their file servers, and/or failed to have these employees sign non-disclosure
agreements when they left for OSU. Of course, it is possible they left
with code in hand or it was provided to them by a friend/colleague that
remained at Texas
Tech (this is a business control weakness better left for discussion another
day)
Now in fairness to the two terminated employees, I would argue that a publicly-funded
institution has no right to apply intellectual property protections to
software and other items developed and paid for with public monies. I am
likely to lose this argument depending on who I am talking with. (a side
note: the
definitive guide for data rights in federal contracts
used to run over 1,000 pages!).
I will now argue that all actions taken by this hand-picked team and all
approvals signed by Dr. Schmidly should be investigated further by OSU
and that
Texas Tech should launch their own internal review of their internal control
structure.
Your turn...