ITAC Meeting Minutes

April 12, 2012, 4-5:30

Allen Board room



Announcements - 5 min

Computer Store Review Summary (Angel Wingate, Timothy Walsh) - 25 min

IT Perspective - Executive Vice President (Dr. Tallman Trask) - 40 min

Fall 2012 Computer Store Offerings (Brian Burtram) - 20 min


Alvy Lebeck called the meeting to order.

The theme for today is the computer store.

Tracy made a couple of announcements;

April 27th is the CIT showcase.

April 18th - We've been getting the word out about the changes to our address management program and how you register computers. This starts in just about a week. Be aware.

Bob reported that we've had a series of low level noise about wifi performance issues, primarily on West Campus. We had an issue back in January that we uncovered and resolved. There were no formal complaints at the helpdesk. We started getting alarms about two months ago regarding CPU usage. Cisco recommended blocking multicasting.  We did an emergency change last night.  The CPU on switches got much better, but we were still getting packet drops, which affects things like video. We worked through the night.  Things are working well right now, no dropped packets. We also extended the pool of addresses that are used for the wifi for West Campus.  The page refreshes went away. Now we are looking to see what these changes may have affected. Right now it looks to have affected Sakai. We had about ten complaints with people getting to Sakai, but they have been resolved. Ultimately this will be resolved by future upgrades to wifi network on campus. The current gear is sized for laptop use, not for all the other devices in use these days. Please pass along any problems. We sent out an email last night & got great responses.

Alvy announced that the minutes from March 1st were available with the agenda. If there are no objections they will be posted to the website.


Computer Store Review Summary (Angel Wingate, Timothy Walsh)

Tim Walsh works in Central Finance and works very closely with the Duke Stores. Tim started by saying that the impact of the recent financial downturn has significantly affected the financial model for the Computer Store. We've also gone through some institutional changes.  Additionally Jim Rigney retired, which provided a chance for change. The most pressing thing has been space pressures in the Bryan Center. We were looking at viability of a computer store in the Bryan Center.  All this set the foundation for this study.

We brought in reps from students, store and OIT.  We drafted Angel Wingate to provide fresh set of eyes to the process. She did an exceptional job leading this process. She did an exceptional job & we're excited to share the results.

Angel reviewed the current state.  We have a store in the basement of the Bryan Center. We have a warranty and repair for faculty, staff and students. We have a warehouse and distribution center on Hillsborough Road. We do consulting services for people who are trying to determine what computer to buy for their particular needs. We have an on-line order processing system.

The process for this evaluation involved a steering committee and lots of data. Brian Burtram and Debbie Deyulia were on the workgroup and can attest that we looked at a lot of data. Michael Ansel was the student representative. He got a lot of feedback from students. We did external benchmarking, but didn't get much response from other schools.

Some of the most telling data dealt with sales volume, transaction volumes and back-to-school information.

Most volume, about $14 million, does not go through the storefront.

The trends show a drop in sales, followed the announcement about extending the useful life of machines and cutbacks in budgets.

It was also pointed out that computers have gotten cheaper in absolute dollars over the past four years.

Tim pointed out that margins were shrunk deliberately as a result of the change in standards process.

Angel resumed, stating that we've gone through severe negotiations with vendors.

A question was raised about pre 2007 sales.  Tim answered that we were on a three year cycle for computers and sales were pretty steady in the $18-20 million range.

Tim noted that standards are in still place. We don't anticipate any further downturn, but we don't anticipate the $15 million baseline to improve significantly.

Angel said that one place that we are looking to recover and grow is in back-to-school sales. This has dropped significantly and we're working to understand why. One key issue we'd like to know is when students get their laptop before arriving at Duke.

It was suggested that they were getting them for graduation. Angel noted that we don't know that. - We're thinking that & if this is true then we have some opportunity to affect some changes with the way we provide information. We're also looking at making changes in the price transparency.

Right now we're seeing a shift in computers that are purchase here. Presently, 72% are Apple, - in 2005 it was 14%.

Tracy added that it varies from year to year, but around 95-98% of students say that they brought a computer to campus with them. A similar percentage of students indicate that this computer is a laptop. Mac has been the computer of choice for more than 50% for freshman for the last couple of years... Most recently more than 70% have been Macs.

Angel showed that in-store transactions indicate that we don't have a lot of foot traffic that results in sales. We had store employees do a tic-mark tracking. There are lots of lookers, not many buyers. Less than half of visitors during this two week test made a purchase. We aren't sure why.

The result of the workgroups findings was a set of recommendations

  • Strengthen the program and become nimble
  • Re-establish partnership with tech and business stakeholders in the Duke community
  • Lots of opportunity to improve our on-line ordering process - We will be looking for help on the IT side for this.
  • Back-to-School - We want to add value and to know what students need and provide that. We want to increase electronic communications with links. We previously relied on glossy brochures which were helpful, but didn't necessarily produce orders. Pricing displayed by Duke was always higher than competitors because there were components built in that weren't transparent . Reasons for higher cost were not obvious - things like extra year of warranty and careless and reckless warranty. We're working on the best way to let people quickly understand the value of these additional coverages.

Alvy noted that this is really an insurance policy. Should we decouple the policy from the sale of the device?

Angel replied that we are looking into the reasons we included these to see if they are still valid. We need to be sure that what we're offering is in line with the support that we can provide.

A question was raised as to whether it makes best sense to communicate directly with the parents. It was pointed out that sometimes this is done as a gift to the student.

Angel answered that we're looking at this.

A student representative offered that he picked his own laptop. A lot of students in Pratt have picked their own, more than in Trinity. It's not uncommon to hear incoming students ask what kind of laptop is needed at Duke. It's not always a student decision, it's often is a joint decision.

The last recommendation deals with the storefront. We will continue to evaluate the needs and assess the impact of the changes that Tim mentioned.

The leadership role for the stores has been filled by Angel Wingate as senior manager.

Wayne Miller wanted to note the value of the store as a showroom. We wouldn't want to lose that.

A question was raised about the repair shop plans. Does it make money?

Angel and Brian replied that it came out loud and clear that this is a viable and beneficial service and should be continued. It is financially viable. It's certainly not a big loser.


IT Perspective - Executive Vice President (Dr. Tallman Trask III)

Dr. Trask started by saying that he'd like to come back every spring for an update.

Dr. Trask noted that we think we're going to have to make some fairly substantial IT investments in the next couple of years. Unfortunately they are boring and mundane things.

The first is parking. We appear to have more parking spaces per customer than any other institution our size. We have more parking spaces than UCLA, but we don't know how to manage them. Our current philosophy on parking is that whenever we can't figure out where some can park we go and build another garage. We're going to look at taking parking into a technological world. Currently we can sort of tell how many people enter a parking lot & that's about it. Because of this we can't direct people to available parking. We believe that even at our busiest that there are about 500 spaces available, but we have no way to identify them. We are currently running a little RFID experiment at Law School. Municipalities have incorporated this sort of intelligence and reported a 20-30% increase in utilization. Thirty percent increase for us is the equivalent of about 7,000 parking spaces. This would avoid a $130,000,000 capital savings versus adding 7,000 spaces.  We currently have a hodge-podge of tools and technologies to address this parking issue. One initial thing we want to do is to throw OIT into the middle of the parking problem.

A question was raised asking if this is for visitors... or all of Duke. Dr. Trask said it's for everybody.

A follow up question was whether this meant that we'd pay based on how far away we are from our destination.

Dr. Track explained that it's not just technology that's a problem. Parking is currently losing money.  We have an arcane system of remote and proximate parking. We have lots of ungated lots that we can't monitor.

There is an agreement to launch the solution in the summer of '13. It will take at least a year to come up with a solution.  We don't intend to solve this problem every year. We will figure out a plan and then implement it over a 4-6 year period.

A second boring thing is the Duke Card. The Duke Card is coming to the end of its useful life. There is a lot of potential capability that could be in these cards. So, we're going to start looking at what that might look like.  In some ways this ties to the parking issue because smarter cards could be smarter in the parking system.

Other than that Dr. Trask is worried about the email consortium. It seems like we're getting to a good place, but the likely outcome will surprise a lot of people

We are working toward late July to bring up the alumni development system. This project is currently on track.

Additionally we continue to talk with basic science departments about what we have to do in these areas. We also are working with high-end computing. Dr. Trask has been watching procurement and what we buy at Duke that is spread out over the campus and we seem to be doing ok.

Dr. Track worries most about investment capital. The things that we want to do cost money. People think that 2008 came and went and we're back where we started. In fact our assets that generate revenue into the operating budget are still down. The income equivalent has been addressed in the operating budget, but we haven't left much room to make big investment choices. We won't see a whole lot of capital projects.  We will need to figure out away to pay for projects that we need.

A question was raised as to whether the capital campaign might change this picture?

Dr. Track hopes so. However, the other problem that is hurting us financially is change in policy. Additionally the impact of 2008 on the families of Duke students has caused financial aid to go from $80 million to $125 million in the last 5 years.

The truth is that it's not awful, it's just not as good as it used to be. We just have to make tougher choices to make.  Tradeoffs are much clearer than 4-5 years ago.

A committee member asked what kind of yield are our investments currently generate.

Dr. Trask said that we assume 8.5% yield.  Up until 2008 it was in fact running north of 16%. Now for a  5-10 year pattern it's a little under 10%. Losing 27% in one year does really bad things to time series of data. Fortunately inflation is down, too. We just aren't making the same investment capital returns.

A question was asked about the Duke China initiative. Is it helping?

Dr. Trask answered that we don't really know. We are hoping that it covers expenses. The financial models that we've run do work. The problem is that it has no financial base so it is basically a tuition based institution. There isn't the foundation of an endowment. It's difficult to know with certainty at this point what the tuition rate will be or how many students will be there. The expectation is that we can scale the project and start out reasonably small. We have lots of capacity. The campus is north of a half million square feet.

A question was raised about the dorm capacity. Dr. Trask said that there's a dorm and a conference space. There's lots of room space with backstop of the nearby Fairmont Hotel.

A question was raised about the efficiency of our IT operations relative to peers. Are there areas we need to address?

Dr. Trask thinks that our IT costs remain on the low end.  Our budget is small compared to Harvard or Yale or Stanford, but in the ballpark with some others. We've done more with less relative to our peer institutions. All of our peers have seen large projects crash.

A committee member asked what the Sakai project cost?

Tracy answered that the cost was $100-150,000 of transitional money to pay for two licenses and people to help with the transition. It was very small.  Once we finish the transition and are in status quo, the costs to operate Sakia will not be much different from the cost to operate Blackboard.

Susan Gerbeth-Jones said that the Nicholas School is using Telepresence and recording to include remote audiences.  Do we see this as a trend?

Dr. Trask replied that this is mostly in professional schools and for remote locations where a return is there.  He doesn't see a lot in core Arts & Sciences, though agrees that it is starting to build.

A student representative asked about changes for the Duke Card. How does this factor in with other things? How much is this costing?

Dr. Trask answered that this has been in use for a long time.  In 1970's Duke created Auxiliary Enterprises. Auxiliary Enterprises never had any money, so while Joe Piet was brilliant at running the organization he would do anything if you paid. We have a lot of things to undo about the card.  We have places where card readers are protecting rooms where the cost of the protection is more than the value of the things being protected. Dr. Trask doesn't think it's going to cost a whole lot of money to fix the problem because we're spending too much now on unnecessary services. Electronic access should be built into the building cost. There may be a one-time hit to make the transition. We assume that the new card will cost more than the current card if we want it to do more. It will affect many things.  We will also need to make a decision about the off-campus question.

There was a question about dual factor authentication. Would the new card do this?

Tracy answered that we are in process of doing investigation into the requirements part of this project.  It is building access, financial transactions, identification, parking and more. The current contract goes until December of next year. We are aiming towards that time cycle.

Dr. Track added that we are encouraging people to not only think about what we need right now, but also what we anticipate we'll need 5 years from now.

A question was raised about how our financial situation stands relative to peer institutions.

Dr. Trask responded that almost all big, private, research universities lost 23-28 % of face value. We were average. What saved us was that we have a small endowment. Our endowment contributed to operating budget 15-17%. Large endowments got clobbered because they were so dependent on endowment income. Only two D1 schools didn't subsidize football in 2008, and in 2009 at least one had to subsidize athletics because of the economic change.

We've done some things to change how we value our assets that make more sense. We are hopeful that we will get back to zero soon.

Alvy said that software as a service, like email, looks good. Are there any risks that you perceive with cloud-based software?

Dr. Trask observed that at some universities these are more interesting because they don't have any new administrative software to speak of and so systems like Workday are promising to give it all to them at extremely low cost. Tracy interjected that Workday is the cloud enterprise system for higher education written by the group that started Peoplesoft. Dr. Trask continued, saying that technically it works, but it makes him really nervous because he doesn't know the exit strategy.   We have everything we need locally. One of the best decisions we ever made was not to separate the health and university system on the financial platform.  The health system would kill Workay because of their complicated payroll and ledger system.

Dr. Trask thinks that email is a good experiment if we can get contractual obligations now that we understand the Microsoft incentives.

Alvy asked what the Microsoft incentives are?

Dr. Trask said that Google is moving into their product space where Microsoft makes most of their money. This is the IBM strategy from the 1980's. - If you get all smart people at all the good universities to use IBM machines, then 20 years later they will buy those products.  They are trying to build brand loyalty and they know that in the teen market that they are being steamrolled by Google. This is a high level decision to address this issue. The people who go to the kinds of schools that are in this consortium are the people who grow up to buy the most stuff.

Dr. Trask also pointed out that this will be an opt-in system. If you don't go you don't have to go. However, if you don't go you won't have any guarantees. The ending deal was Google's unwillingness to deal with federal privacy laws.

Fall 2012 Computer Store Offerings (Brian Burtram)

Brian started the presentation by showing changes to the program for the upcoming year.

We have seen continued decline in sales to incoming students. Efforts have been made to get some sales back. We are looking at providing transparency in pricing by showing the price of the product and also showing the additional components that contribute to the total price. In the past we've just shown the total price and listed all what is included. This should make it easier for buyers to make more realistic comparisons.  We're also making it obvious when they are looking at a consumer machine compared to a business machine.  We also plan to split out the vendor warranty cost. We also buy vendor extended warranty for the additional, fourth year for pcs. In addition, we are also going to removing bundled items like surge protectors and cables as a requirement. We did talk about the internal "crash & burn" warranty, but decided not to change that much this year.

Robert Wolpert asked if we self insure for this.

This was confirmed.

Brian then showed a slide with Apple offerings. The two Mac models are essentially the same as what we offered last year. The basic Macbook will have new name - not "Value."  We are going to change the name to something more attractive. On the Apple side, screen size has driven the cost and determined whether a unit is "Value", "Mainstream", or "Performance."  Additional items will not be optional, but we will show all itemized costs so that buyers can compare.

Brian showed details of models, costs and options for each level in the Mac line. He indicated that one goal was to get a model that we can sell for under $1000. We still weren't able to do this with Apple.

Robert Wolpert asked what we do when new models come out.

Brian answered that this is why we are the changing communications model. We will send out more general mailing that will point to the website. The website will allow us to make changes more often.

Brian then showed the Lenovo offerings. These offerings show prices with everything rolled into one price. We were able to get a Lenovo offering under $1000. We will show all costs on these as well. Last year we did offer the mobility X220. We didn't feature it very prominently and we didn't' get a lot of interest. This year we may feature it a little more prominently to try to generate more interest.

Brian then showed the Dell offerings.  We are looking at getting consumer models so that people can compare with Best Buy etc. In the past we haven't been able to do warranty repair work on these models. We've worked it out this year so that we will be able to do reimbursable warranty work on consumer units that we sell. We hope this will capture some outside sales, and service.  Brian pointed out that consumer models are comparable in capability. They just aren't as durable.

Ed Gomes pointed out that components change more quickly for consumer models.

A student representative pointed out that it would be valuable to give descriptions that point out typical use cases for particular models. Descriptions of mainstream models weren't very descriptive as to how students would use them. Information provided didn't seem to match up to what a student needed to know.

A student representative noted that if Duke had offered consumer models he probably would have bought at Duke. Adding consumer models will allow students to compare more directly.

The meeting adjourned at 5:29