This is the governing body for certification in Disciplined Agile.

Produce a Potentially Consumable Solution

  • 15 Jul 2018 10:54 PM
    Reply # 6382504 on 6317613
    Scott Ambler (Administrator)

    Everyone, great feedback.  As usual I'll go one at a time:

    @Drennan:

    • Yuck!  Very good point though.  Adding that to the trade-offs.
    @Valentin:
    • Managing the work and WIP: Good idea.  I'd rather say Coordinating the work and WIP though.  Also, I suspect that this is completely covered by the new version of Coordinate Activities, which we haven't published yet (likely in August).  So let's hold this thought for now until that goal is out
    • Look-ahead planning: Agreed.  I've updated accordingly.
    • Look-ahead modeling: Agreed, also updating.  I really like your thoughts around eliminating waste.  Sounds like a good blog posting for you.
    • Coordination meeting: Agreed.
    • Continuous documentation: Agreed, added a tradeoff.
    @Aldo:
    • BDD - I've pointed out in the description that it's also known as SBE. I've always felt the BDD vs. ATDD vs. SBE debate falls into the category of "how many angels can dance on the head of a pin?"  Might be interesting over a beer, but in general a distraction for people trying to learn the technique.
    • Refactoring - This is covered separately in the Improve Quality goal.
    • Testing - We're applying the automation pyramid in Accelerate Value Delivery where a lot of V&V stuff is captured.  Not yet using the Testing Quadrant there, although likely should.  Need to think about that a bit more though, but definitely leaning in that direction.  It's good stuff, just not completely sure yet if we want to go to that level of detail in this book.
    • Deliverable documentation - Living documentation is an example of single sourcing info as I pointed out for Improve Quality (and now explain in that goal).

    I will update this excerpt in a few minutes as there are no changes to any of the graphics based on this feedback.
     


  • 04 Jul 2018 6:46 PM
    Reply # 6360759 on 6317613

    Hi All

    There is one Item that requires clarification from the "Develop Software" decision point. It indicates that BDD and ATDD is the same in the detailed description on pp. ; I have met some people that considers these as different items, and some saying is it the same. The same argument applies to SBE (Specification by example); Perhaps the option should be documented as "BDD/ ATDD/ SBE" instead of just "BDD". 

    Would it also not be useful to specifically mention refactoring as a separate option? or even mention it as part of the TDD option: "TDD and Refactoring"?

    Then a note here: The Testing Quadrants and automation pyramid can also be used at this level. I would recommend considering some of the testing choices from the Initiate phase in this section as well as the tools mentioned in there will apply to a lower level as well... 

    Write Deliverable documentation: Gojko Adzic talks about "living documentation" where test scripts are used as the latest documentation. Would it be possible to look at multi-use documentation types? (such as using test scripts for doubling as test scripts AND user documentation)

    Regards

    Aldo



  • 23 Jun 2018 12:22 PM
    Reply # 6335458 on 6317613

    Look Ahead Planning

    IMPORTANT – as a trade-off for <Potential to be wasted effort if the work item is deprioritized or even removed …>

    • Eliminate waste of waiting in case of missing information. Lack of availability
    • Eliminate waste: ordering allow to more efficiently negotiate scope by deferring less important work items

      

    Look Ahead Modeling (Requirements/Solution)  

    VERY IMPORTANT - Imo in this part we need more trade-off guidance because Look Ahead it is the main instrument to eliminate waste!

    “If the work item becomes a low priority and is not implemented the modeling work becomes a waste. The further ahead you model he greater the risk that the requirements will change, and your modeling will be for naught.”

    Looking ahead could be a great instrument to eliminate the waste, but also could create waste if it is not adapted to the context and we do not follow some principles:

    • The amount of modeling info should be inversely proportional to how further ahead we model. We should invest more only in high priority items close to the main development. A too detailed look ahead it is not a look ahead (but Big Up Front …).  
    • Looking ahead will eliminate waste of waiting due to unmanaged dependencies on exchanging information and waiting due to unexpected lack of availability.  These are the proper goals of Looking ahead and not getting detail specifications.  

    Again, Looking Ahead it is one of the main instrument to eliminate waste of waiting and getting a Lean, streamlined process! Looking ahead on development is similar with Looking ahead on driving.

    Waiting is a disastrous kind of waste because will mess-up the next incoming work in cascade.


    Plan The Work, Coordination meeting

    Help to optimize the team collaboration needs and people inter-dependencies in that day. Looking ahead - will avoid waste of waiting for the current day.

    Write Deliverable Documentation, Continuous documentation – Following iteration.

     Ime it is not recommended when we do not validate the working software with automated tests. We cannot end the iteration without testing, and we cannot perform manual testing without documentation.

  • 23 Jun 2018 9:43 AM
    Reply # 6334988 on 6317613

    I propose another decision factor: “Managing the work and the work in progress”. Planning is just one aspect, TDD as development approach is also one aspect, but the first decision to make is how we will split the work. This is one fundamental aspect of Agile & Lean.

    If our development work will be (or need to be) split in days, then we must choose and master appropriated practices for planning, requirements, solution, consumability. That mean decisions on all aspects must serve and make possible the selected work management approach.

    In fact, this factor it is just a translation of the principle “Produce a consumable solution incrementally”: the first effect it is at work splitting level (and reducing WIP).

     Managing work and work in progress

    • Roadmaps
    • Release (with a preference to shorter releases)
    • Release phases (with a preference to make inception and transition shorter)
    • Iterations (with a preference to reduce iteration size and ceremonial)
    • Weeks (for larger Iterations; used by XP)
    • Days/Day
    • Looking ahead – opportunistically looking ahead
    • Just in time – ad-hoc collaboration, and decisions and feedback
    Notes
    • No efficient ordering (just timing size) and no default options; all are evermore valid

    Principles

    • Split the work and make progress to reduce the work in progress
    • Have a pragmatic approach for all the levels of our work rolling wave: product roadmap, release vision, iteration, look ahead, just in time. That will require adequate practices for all levels

     

  • 18 Jun 2018 1:42 AM
    Reply # 6317775 on 6317613
    One Trade-off closest to my heart wrt Coordination meetings/Daily standups:
    • An in-the-open conversation could open the conversation to micro management and derail self-organisation even when stand-up rules exist. (I have seen stand-ups where it is hurried before "management" arrives so everyone can be open and not have to defend themselves)
  • 17 Jun 2018 8:07 PM
    Message # 6317613
    Scott Ambler (Administrator)

    We just posted the excerpt for Produce a Potentially Consumable Solution.

    Please post any feedback that you have for this goal here.

    Thanks in advance!

© 2013-2019 Project Management Institute, Inc.

14 Campus Boulevard

Newtown Square, PA 19073-3299 USA

Powered by Wild Apricot Membership Software