This digest was created in real-time during the meeting, based on the speaker’s presentation(s) and comments from the audience. The content should not be viewed as an official transcript of the meeting, but only as an interpretation by a single individual. Lapses, grammatical errors, and typing mistakes may not have been corrected. Questions about content should be directed to the originator. The digest has been made available for purposes of scholarship, posted by David Ing.
INCOSE International Workshop 2014, Torrance, California
Review articles are at:
SDLC looks like waterfall, but saying it’s agile
Executives think about dollars and time as linear
- Engineers think iteratively
- Phase-gate processes can stifle innovation
Two things important
Get governance and operational delivery as separate
- Governance: concept, plan, develop, manufacture, deploy
- Operational behaviour in envision cycle and explore cycles, iterative
Governance as primarily finance and on risk
- Waterfall has been around so long, so that governance and product development cycles are synchronized
Governance cycle shouldn’t mirror waterfall activities, as impacts risk and assessments
- If want to minimize biggest risk areas first, could be requirements, but could also be design
- Could build simulation model
- Operation as envision and explore cycles
- Could have multiple operational delivery cycles under the same governance
Have used this approach for a hardware company, as well as a bank
Comments: Have done work with companies that contract multiple companies and then in phase 2 pick the one who has been doing the best work.
Comment: Map to government customers, DoD?
No, I stay away from DoD. Get a call once per year, and am usually too busy. Have shared ideas with DoD people, but haven’t engaged in contracts with them.
Comment: DoD customer has moved towards this, affordability analysis over decision-gates.
Comment: This message is as much to the agile community as the systems engineering community.
Comment: For people to get continued investment, need to be concerned with governance.
Change from initial agile work, to more recent larger projects.
- Have been in projects where agile people have said ridiculous things about we can’t tell you what you’re going to get, or how much it costs.
- Architecture in software or hardware, it’s a paradox in agile. Amount of up-front architecture has to be done on a product-by-product basis.
- Saying architecture will evolve, could result in data schema in driving the system, but don’t want database management system to drive.
Comment: From DoD perspective, diagram shows a lot of exploration, versus deployment. Paradox of governance and knowing when we’re going to get done. Should there be commitments to envision cycle, before explore cycle, before deploy cycle.
Explore cycle means having all things that are required for continuous delivery, e.g. test, deployments.
- In software, are moving towards continuous delivery, hwich means can deliver new features to a customer on even a day baiss.
- Saleforce.com has 100,000 automated tests, and can deploy rapidly.
- In hardware development, would have to approach differently.
- In software, can stage these over 3 to 6 months period, or could do continously.
- Explore cycle still has outcome of something that is deployable.