Marco Dorantes' WebLog

"Computer science is no more about computers than astronomy is about telescopes" -Edsger W. Dijkstra

MSF v4.0 and agile development

MSF v4.0 and agile development

  • Comments 6

Microsoft Solutions Framework next version 4 explicitly includes elements from Agile Methods for software development [1], some of these elements will have built-in support [2] in next version of Visual Studio .NET

 

Behind this evolution are modern observations about software development singularities which imply a particular management to create goods made of bits, in contrast to traditional management to build goods made of atoms [3].

 

What is plain wrong and misleading with this article [2] is the characterization between an agile and a formal process:

 

Formal processes are a set of very specialized mathematical modeling techniques that are applicable to systems design, heavily based on propositional calculus or first-order logic mathematics with tools like VDM or Z language. See http://www.afm.sbu.ac.uk/

 

On the other hand, agile processes are based on empirical control, self-adaptation and emergent behavior.

 

The article seems to contrast these quite orthogonal things and doing so just add confusion and misleading information.

 

Perhaps and most likely, what the author is trying to say with “formal processes” is “processes following established forms and conventions”, but even if this is the case, a less misleading phrase could be “traditional processes” instead of “formal”.

 

A more real difference can be stated as “a plan-driven” (traditional) vs. “planning-driven” (agile) processes.

 

[1]

Manifesto for Agile Software Development

http://www.agilemanifesto.org/

http://www.agilealliance.org/

 

[2]

Visual Studio 2005 Team System: Microsoft Solutions Framework

http://msdn.microsoft.com/vstudio/teamsystem/msf/default.aspx

http://msdn.microsoft.com/library/en-us/dnvsent/html/vsts-msf.asp?frame=true

 

[3]

No Silver Bullet Revisited

http://virtualschool.edu/cox/pub/NoSilverBulletRevisted/

 

  • Marco,
    I think you're quibbling about the term "formal". We make no attempt to refer to "formal methods" in the sense of Z or VDM. We are certainly aware of these, but they are largely irrelevant to the bulk of software development.

    Maybe we have picked the wrong term for "MSF Formal". We have considered "MSF Complete" and "MSF Plan-Driven", but both of these seem to imply that MSF Agile is neither complete nor plan-driven, which implication we do not want to make.

    We'd welcome your suggestions for a more appropriate name. Thanks,
    Sam Guckenheimer
  • Thank you Sam for the clarification

    MSF Traditional or MSF Plan-Driven could very well describe in a name what we have known till version 3.0

    Now, if MSF 4.0 wants to catch up agile essence in its name, consider:

    MSF Continuous Planning
    MSF Active Planning
    Dynamic MSF
    Evolutional MSF
    Co-evolutional MSF, see:
    Wicked problems, righteous solutions
    http://blogs.msdn.com/marcod/archive/2004/06/12/154131.aspx

    Progressive MSF
    Microsoft Solution Discovery Framework
    MSF Discovery Edition
    Human-oriented MSF
    People-driven MSF

    Best regards,
    Marco
  • My recommended name for MSF 4.0 is
    ( SMOOTH ) Software Methodology Orthogonally Opposed To Heinousness
Page 1 of 1 (6 items)
Leave a Comment
  • Please add 6 and 2 and type the answer here:
  • Post