Team 2 Week 12 Blog

Team 2 Week 12 Blog

 

This week our beneficiary discovery continued to focus on deployment and user testing. We conducted A/B tests of program features, including a smart search bar, a calendar style training plan view, and an improved training card design.

 

Key Takeaways

  1. We received good feedback this week on some new prototype feature A/B tests. Military users would like to see an option for a calendar layout, more information on the training card, and the option to use a smart search bar that automatically suggests training based on METL tasks.
  2. We were exposed to new options for web development, including crowdsourcing through “project discover,” a Duke program run by the colab where interested developers can join. We are not sure whether this is a viable or effective option yet however.
  3. We reconnected with some continued contacts and connected with some new contacts who like the program enough to be advocates. These will be key to integration past 3rd group 2nd battalion.

 

Key Takeaways

 

  1. Chief Joe – Company Commander
  • Blueprint wins over search. After approved training plan then they get list of METL. If it takes the   METL list and looks at venue and dates open with alternatives for cost and location then that would be awesome.
  • Calendar wins if it pulls from ATARS and group asc calendars and goes through process above. A one stop shop for dates would be amazing because currently he has five tabs open with five different  calendars.
  • Current process for scheduling is

-Pull school dates from Atars

-Look at training events run by group

-Look at free dates and overlay the open days

-If contract isn’t open then they may need to look for civilian training(cost, location, and rating is important)

-Currently set aside time for individuals to go to advanced schools(planning individual training at advanced schools would be helpful)

-Balance available dates to open advanced schools

 

  1. Will Ye – Student Developer at Co-lab
  • Project Discover is a new program run by the colab where people can post projects and then interested student developers can join
  • Still new so may take some time and involves posting project specifications
  • Alternative is posting to Duke computer science facebook page since that gets a lot of views

 

  1. Michael Faber– Manager, Academic Technologies Duke OIT
  • Check out Project Discover to get web development help
  • Using Project Discover does not cost any money
  • Time scale is variable, depending on the intensity of the project

 

  1. Tony Fernandez– Research Director, EMS Performance Improvement Center
  • To build the back end of an EMS-facing service, you would have to work directly with the state or the national registry.
  • It may be possible to instead work with the accreditation body for EMS continuing education, CECBEMS.

 

  1. Matt Nash– Managing Director for Social Entrepreneurship, Duke Innovation & Entrepreneurship
  • As you look at civilian applications, it’s important to understand the challenge each industry faces: Is the issue that certain trainings are not certified by the certifying body or is it that the different systems not talk to each other?
  • For additional funding, consider entering the Venture Startup Challenge or the D Prize for Social Innovation, consulting the Duke Angel Network, and possibly the Entrepreneurship class.
  • Bureau of Labor Statistics and SHRM may give you a clearer picture of how training and CEUs are broken up, by district, country, industry, etc.

 

  1. Former ODA Commander Aaron
  • Assuming teams buy in, this will save around 8 hours per week.
  • Love and would expand on the running budget tracker. Can the budgets of each team be put together to form a battalion budget?
  • In terms of A/B testing of calendar vs. list style training plan format- would like to see an option for both. Calendar is good for seeing where team is overloaded and where white space is. List style is good for seeing upcoming training events in more detail.

 

  1. Team Sergeant Johnny
  • “I’m really looking forward to using something new, we’ve been doing the same thing for 40 years!”
  • Add to export capabilities- list of participants with contact info and identifying info
  • Tracking ammo should be a priority- amount estimated and expended. Teams already track this but it isn’t organized in one place. New style training card is good for this.

 

  1. SF IT Contact – Brian
  • Approving a new program to go on the SF network requires a CAPER request. It goes through the Group S6 to USSOC G6 to First FC G6.
  • SF systems and rules are semi-independent from Big Army.
  • Brian could not think of any web-based applications that they used. He thought SharePoint but when I probed more closely he realized SharePoint was almost certainly being run on Army servers.
  • He did not have knowledge of security standards, system administration and monitoring requirements, or commonly used IT platforms/standards that are already in use.

 

  1. S8 Jade
  • His role is Force Modernization and Innovation.
  • His job is to work with G6 to support ODAs in meeting resourcing needs that are more unconventional in nature.
  • On innovation-side, he works w/ groups such as Johns Hopkins and MIT to evaluate and ultimately deliver – if possible – proposed systems/hardware to meetin SF needs.
  • He was highly enthusiastic about BulletTrain. He has a contact in DC that has been trying to interest him in an upcoming Hacking For Defense conference.

10. Tim Betts

  • CICD still the way to go for an installed solution but you lose 75% of the benefit. How often would they update? Who administers the push? How does that staff get trained? Staffing and technical costs would balloon.
  • Unless we learn SF has a development infrastructure, setup CICD in the cloud. There just won’t be true continuous delivery. How would we deliver updates? Environment setup and admin requires a SysOps type. On a small team, it is a subtask. Larger team, defined role.
  • Secret vs Non-secret is interesting. Does the restriction play out to the internal user-base or if you are on the system, you are cleared for all of it? If the differential was internal as well, we could do altogether separate systems. That would impair the network effects. Alternatively, we do separate databases where needed – different training plan databases but single course catalog. Insights and Enhancements functions would require workarounds. If you take the whole thing to Secret, does everything flatten? Or are there degrees of secret that leave us in the same situation?
  • Development, testing, and pre-prod environments are generally straightforward. Proper curation of the mock data is issue as always. Bigger problem is early days there is no existing data to model/scrub.

This entry was posted on Thursday, April 4th, 2019 at 3:47 pm and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

Leave a Reply

Your email address will not be published. Required fields are marked *