4:00 – 4:05 – Announcements

Ken Rogerson welcomed all to this last meeting of the summer and there were no announcements.

 

4:05 – 4:30 – DukeMobile Changes, Hugh Thomas

 

What it is: The DukeMobile app, available for most mobile devices, puts important information about Duke at your fingertips. However, its next release will transition the app to a map-centered approach. Hugh will preview the upcoming changes and explain the data driving the decisions, as well as what they mean for users.

 

Why it’s relevant: After several iterations of DukeMobile, this next version represents a significant shift – changing the face of the app to focus on one priority use case, based on user goals and feedback. We will discuss this new direction and how it will address user needs.

 

At the ITAC meeting on 3/22/2018, the presentation on DukeMobile featured the following:

  • New Look and Major UX redesign including both iOS & Android native versions
  • New Map / location context with advanced search capabilities and Campus navigation feature
  • User customizable content including simpler navigation to user specific content and secure account login option
  • Planned release during Summer of 2018

 

The new Duke Mobile App to be launched this month showcases a new look & feel with 100% features preserved from the previous version and a new interactive campus map with clickable building details and images.  It includes campus and local navigation with multimodal directions for (Walk, Bike, Drive, Transit options) and sophisticated searchable building data using GraphQL platform and server backend with Solr search capability.

 

Future updates and the current wish list includes user specific experience with personalized and richer content, Shib authentication & biometric login.  Also, incorporating Duke Card support & Flex account management.  The additional map features will be transit improvements (Durham & Triangle) and also

Indoor floor plans (protected via Shib).  Meta search features will include “Food”, “parking”, “menus”, “bathrooms” etc.  “Tiles” will be incorporated to get higher level information such as travel approvals etc.

 

Today we are only able to search based on buildings but will be able to expand to multiple services in the future.  The server itself is using a technological GraphQL which is a sophisticated search with a lot of plumbing that’s not utilized fully today but serves as future flexibility.

 

When using the app, in the home screen you will see a bubble about 5 miles around Duke Central but if you’re within that bubble the app will locate you and draw your center but if you’re beyond the 5 miles region, the center will be around Duke.

 

The entire campus is now mapped out and the each one of the buildings is a clickable entity.  It’s interactive and the building details come up when you click on buildings.  The Search bar allows you to search and move around the map and see the metadata around them - people, places, events, things etc.  The UI design is taken from Google Maps and Apple Maps applications.  It’s a conglomeration of data in xml format from both the Facilities Management organization that provided all the layouts and foot prints of the buildings as well as the CampusBird application maps.duke.edu that provided the metadata, building descriptions, and photographs.

 

At the moment searches are not trained but the plan is to train the results to be predictable on the number of hits.  This has been in test for about a month, runs on iOS, and the next release will include bug fixes and floor plans.  This is not part of Duke Mobile today but will be in the future.

 

Questions and Comments:

Q1.  What is the ETA?

A1.  There will be quarterly releases.

 

Q2.  Is there any way to add a PayPal or Venmo app to pay for things not through Duke Flex?

A2.  The Duke Card will become part of the app and will be associated with the flex account.

 

Q3. Is there any public information for consumption prior to the release that you can share with the community?

A3.  We are looking at a couple of weeks before releasing that information.

 

Q4.  How streamline is maintaining the parallel iPhone and Android versions of these?

A4.  Since there are only 3 people in the mobile group, we continually build and update the same features on both platforms.

 

Q5. Are the menu items still external browsers?

A5.  Yes, it’s a web based app so the majority are external.

 

Q6.  Which of these are native apps?

A6.  Giving, Safety, Maps, Social Media, Events etc.

 

Q7.  You mentioned that the App uses Shib authentication as well as biometric data.  Are there any plans to store the biometric data?

A7.  Yes, we are tapping into whatever biometric data Apple or Android maintains.

 

Q8. You’re data is coming from various sources and people tend to change things overtime.  Do you expect it to be a challenge to keep things up-to-date?

A8.  The notion of GraphQL platform allows us to adapt different types of data so we don’t expect major challenges.

 

Q9. Facilities gives us an excel spreadsheet and if they decide to add a column, do we then have to massage the data?

A9.  We’ve taken what we need in terms of a foot print and now we need a centroid of every building.  Next step is to enhance the dataset.  We’ve done a lot of the uplift for now and in theory it will be ongoing incremental improvements.

 

Q10. Do you have any plans to be able to customize at the School level?

A10. There is a steering committee that manages the new features and these decisions. Our goal is to keep improving this overtime.

 

Q11. What do the building colors represent?

A11.  We will be using the same pantomimes as they are in blue book including the hospital buildings.  It differentiates between Duke owned facilities vs. rented.  As soon as the backend has updated data, the app will sync.

 

Q12.  Do you really hope to someday have indoor maps of complicated buildings like the libraries and the hospital?

A12.  Yes, high traffic and popular public spaces and that vision would be based on the data we get from FMD.

 

 

4:30 – 4:45 – Drones at Duke, Chip Bobbert, Evan Levine

 

What it is: OIT’s Academic & Media Technologies group has been heavily involved in the expansion of drone usage at Duke. This presentation will walk us through the history of Duke’s drone program, as well as where it’s headed and areas for further growth. We’ll also review training opportunities and practical applications in academics and research.

 

Why it’s relevant: Drones are an expanding market that is changing the way we do things – from inspecting infrastructure and producing videos, to delivering goods and scanning art. These systems are increasingly intelligent, making use of advanced concepts such as computer vision, full “driverless automation,” augmented reality, and cloud connectivity. Developing a plan for scaled management and smart growth will provide community exposure opportunities, use case development, and training programs; and it is an essential approach to leveraging the new technology.

 

Drone History

  • Drones aren’t new, they are recently consumerized but have a long history
  • Strong connections with the model airplane industry
  • Significant drops in price and improvements to technology have enabled new “flyers”
  • FAA has recently developed the first standards for drone licensing
  • Drones are no longer toys, many commercial applications for light systems are emerging

 

Drone History @ Duke

  • Research into law and practice
  • Working group formed to establish safe standards specific to Duke beyond FAA or state requirements
  • drones.duke.edu website established to inform the campus community of new standards
  • “Flight Request” process was developed
  • “Safe-to-fly” areas were established

 

Current Progress

  • Continue to refine drones.duke.edu
  • Drone Pilot Program - Drone Checkout Kits
  • Access to commercial drone pilot(s) as its required
  • Roots Training Classes This Fall to learn use cases
  • Incorporation of Drones in OIT and other campus showcase events
  • Symbiosis with current programs as with the Maker space and VR programs

 

Use cases and Research @Duke

  • Humans & Autonomy Lab
  • Marine Science (MaRRS) - Marine Robotics & Remote Sensing for mapping beach erosion, whale sizes etc.
  • https://sites.nicholas.duke.edu/uas/

 

Future Progress

  • Data Collection
  • Adding a registry of certified drone pilots
  • Drone “cage” where drones can fly around in a controlled space
  • Drone Research
  • Interface with networks and applications
  • Drone specific events (racing, VR/AR, “practice flying”)
  • Possible expansion of Drone Fleet?
  • Expanded Training on certification track with a 20 hour time commitment
  • Increased flying locations
  • Streamlined process for approvals
  • Scaled tech, drone flight logging and tracking system as a Cloud equivalent of the Black Box.

 

Questions and Comments:

Q.  How far behind the curve are we with these policies?

A.  Right now Public safety will stop you if they see you flying a drone.  Also, the Duke Drone Club has started which will comply with the safety guidelines.

 

Q.  If someone checks out a drone, do you provide them with a certificate to start flying it?

A.  Simply checking out a drone, does not allow them to start operating it without the security guidelines.

 

Comment:  We do get real request from visitors with Drones in the car trunks wanting to take pictures of the Chapel.

 

Q.  How many are available to check out?

A.  We have 2 of the smaller kits that have collision avoidance and best suited for researchers such as to scan an archaeological dig site.

 

Q. How enforceable are these policies?

A.  Completely enforceable under NC Law.  The police will stop them.  The National Defense Authorization Act for 2018 recently approved by Congress contains language that restores the requirement for hobbyists to register drones over .55 lbs.

 

Comment: The cost of drones are fairly low and the newer drones will have facial recognition.

 

 

 

 

4:45 – 5:05 – Ivy+ Updates, Evan Levine, Chris Meyer

 

What it is: Representatives from top-tier schools meet annually to discuss and share information in various areas. Topics include overall university directions, budgets, projects, online learning tools, and daily operations.

 

Why it’s relevant: Sharing experiences and discussing challenges with our peers helps provide a collaborative environment where ideas are formed and problems are solved. Evan and Chris will share their experiences at the 2018 conferences for the Labs and Administrative Systems groups.

 

Evan hosted the Ivy+ Classroom technologies group which largely turned into an academic technologies conversation.  There were discussions on how to keep the two Ivy+ groups that exist on successful and different tracks and not merge into the same thing.  The conference was well attended and was quite enjoyable.  Some of the takeaways were the following:

  • Our lab strategy has been to reduce desktop computing labs that are general purpose and repurpose staff and facilities to support Maker Spaces, which is unique among our Ivy+ peers.    Other Ivy+ Maker Space efforts appear to be much smaller or funded in different ways. 
  • For E-Print, we are by far the most generous in the pricing and free printing.
  • Trinity participated in the Classroom technologies and people really like the classroom in a box, a $12,000 investment that gives you the ability to walk into a room and have all the tools such as camera, microphone etc.  Keeping dongles secured is still a problem.
  • Panapto is still used for recording lectures at many Ivy+ schools.
  • Some schools started using Alexa to create accessible AV controls in rooms.  Voice is the next frontier and the notion of Classrooms being smarter in terms of knowing all the presets and everything just comes on as needed even when the room changes.
  • People were very impressed with the broad integration of technologies at Duke.

 

Questions and Comments:

Q1.  How unique is our Co-Lab?

A1.  Fairly unique.  The difference is that all schools have a Maker Space but ours is run by IT and made available to anyone affiliated with Duke as opposed to a specific academic program.  We are seeing positive results because of this access.

 

Chris attended the Conference for Administrative Systems Group last April and reported the following highlights:

  • There were good discussions around Application Programming Interface management systems.  Several schools are continuing with the Community edition of MuleSoft which is more cost effective as opposed to the Enterprise edition.  The same is true for Salesforce. 
  • Some schools are exploring the G-Suite collaboration tools and also refreshing their Legacy Alumni systems. 
  • During lunch with the host school CIO, talks focused on initiatives for the upcoming year, some of which included collaboration, video conferencing, directory strategies, upgrading legacy and aging systems, Quality and Usability, and finally User Experience.
  • One school has taken the lead to experiment with Dev Ops where they break up system silos into project teams and portfolios to improve efficiencies in handoffs.
  • Cornell talked about their student experience initiative to better connect students to resource.

 

5:05 – 5:30 – DNS Sinkholing at Home and on Campus, Niko Bailey

 

What it is: “DNS Sinkholing” is a technique to block internet traffic to malicious or otherwise undesirable sites by altering the default way data is routed. “Pi-Hole” is a free, Linux-based, network-level, blocking application that can run on very basic and inexpensive hardware, such as a Raspberry Pi. It functions using DNS sinkholing to block known ad domains, as well as other malicious domains, with the option to import custom feeds of undesirable sites as well. Pi-Hole doesn’t scale well on larger networks; it is intended for smaller home or private networks. Duke has recently implemented DNS sinkholing on a broader scale using various intelligence feeds and blocking domains by way of our Intrusion Prevention System.

 

Why it’s relevant: DNS sinkholing is a much more effective alternative to IP address blocking, as many malicious domains change IP address frequently to keep their malware effective. Here at Duke, this service is already in place and having a positive effect on your day-to-day internet use.

 

The internet is largely funded by target advertising.  When you visit a website, it assigns a tracking cookie to your session which follows you throughout your session and build a profile on your browsing habits and servers you targeted ads.

 

DNS is a database of Domain Names of Servers and IP addresses.  A DNS Sinkhole is a DNS server that is essentially setup to return a false response based on records that it maintains of the malicious domains or ad domains.  Sinkhole is a good technique for add blocking.

 

But ... Why DNS Sinkhole?  “If you’re not paying for the product, you are the Product”.

Duke’s ITSO started blocking by domains instead of the IP addresses because IP addresses can dynamically change.  The following have been achieved by DNS Sinkholing:

  • Removed 79 million ads for redirects to malware – uses dynamic IPs
  • Removed 48 million ads for unwanted software installs
  • Suspended 7,000 AdWords users for scams.
  • Blacklisted 700,000 mobile apps for violating ad policies.
  • Blocked 12,000 websites with content scraped from legitimate websites

 

Malware from ads is also used for generating bit coins while the page is being viewed.

A Raspberry Pi costs about $35 which can be used with pi-hole OS to setup a sinkhole.

The default set of credentials that come with the device should be changed.

 

Helpful links included the following:

  • https://pi-hole.net/
  • https://www.reddit.com/r/pihole/
  • https://github.com/pi-hole
  • https://dietpi.com/
  • https://www.raspberrypi.org/downloads/ (Other OS’s for the Raspberry Pi)

 

Questions and Comments:

Q.  How bad is it really, if you don’t know that it’s happening?

A.  It all comes down to how we value our privacy.