Re: [unrev-II] Species survival & open source funding etc. (was Re: Intel's new XML network devices)

From: Jack Park (
Date: Thu May 11 2000 - 07:25:28 PDT

  • Next message: Jack Park: "[unrev-II] An open question on XML/ClientSide/ServerSide/RDF"


    I believe you just described the "waterfall" (a bad characterization, if I
    may say so), as I have. I believe I mentioned that the whole process is
    iterative. You have added emphasis on the release early, release often
    approach. We agree on that.

    What you are describing is much closer to Directed Evolution, perhaps the
    most appropriate way to design and build stuff. In any case, the process
    must preceed prototypes with use cases and requirements, as I have said
    early, and said often.


    From: Eugene Eric Kim <>

    > On Wed, 10 May 2000, Jack Park wrote:
    > > I believe it is a mistake to assume, at this time, that the "waterfall"
    > > methodology is dead for the OHS/DKR project. Extreme programming
    > > the equivalent of a Darwinian hack and slash approach, one that I do not
    > > favor in a project that is as complex and potentially important as this
    > The waterfall methodology does one thing very nicely: it outlines the
    > steps required for developing software. What it does not do is take into
    > account truly iterative development or properly model the process of
    > software creation. Many methodologies start with the waterfall process
    > with modifications to address these problems.
    > I don't want to get into a methodology war, and I'm not advocating total
    > conformance to the extreme programming methodology. Extreme programming
    > requires pair programming, building test modules before code, etc., things
    > that are either not feasible or not applicable to our project.
    > Nor am I advocating a "hack and slash" approach, with total disregard to
    > design, requirements management, testing, etc. What I'd like to see are:
    > - Emphasis on building prototypes.
    > - Small sets of prioritized requirements per release.
    > - Aggressive scheduling -- releasing early and often.
    > I think all of these things are beneficial for open source software
    > development in general, and especially for what we are trying to achieve.
    > -Eugene

    Double your manufacturer's warranty on all computers,
    home appliances, and electronics AND win up to $500
    towards your purchase!

    Community email addresses:
      Post message:
      List owner:

    Shortcut URL to this page:

    This archive was generated by hypermail 2b29 : Thu May 11 2000 - 07:30:52 PDT