Loading ...
Sorry, an error occurred while loading the content.

Re: [XP] Re: Certification on eXtreme Programming

Expand Messages
  • Olof Bjarnason
    ... I think one good indicator of skill is the list of apps completed so far . Show-casing what you ve done previously. [Rant: Certification in general just
    Message 1 of 94 , May 3 11:22 PM
    • 0 Attachment
      2009/5/4 Larry Brunelle <brunelle@...>:
      > Ron Jeffries wrote:
      >> Hello, Larry.
      >> I'm pretty sure I mentioned something about scaling in my posting
      >> ...
      > Went back and looked, but didn't seem to see it . . .
      > apologies if I still missed it. More comments below.
      >> On Sunday, May 3, 2009, at 3:25:13 PM, you wrote:
      >>>> What kind of social or other process could there be that would
      >>>> address the legitimate need to know whether someone was fair, good,
      >>>> or excellent at something. Or whether we ourselves were?
      > [snip]
      >> ... My question has to do with the broader
      >> question of how we might screen our own skills, or those of others.
      >>> It might be relevant to discuss what is the "something".
      >>> Is it a skill? Or is it the ability to acquire the
      >>> necessary skill? Is the most relevant skill, say, a
      >>> programming language? Or is it the skill to recognize
      >>> unnecessary complexity and to know what to do about it?
      >>> Or maybe even the skill to share skills with others
      >>> effectively?
      >> I would think there would be many skills of interest to hiring
      >> entities, including but not limited to things like:
      >> Test-Driven Development
      >> Exploratory Testing
      >> Design (itself a big topic)
      >> Refactoring
      >> There might be topics like Story Splitting, simple design, and on
      >> and on.
      >> I would suggest that we could grab a group of people, assess some
      >> others, and pretty much agree on what things they were good at, and
      >> what they were not ... at least on some dimensions.
      >> If we couldn't ... then there is no hope for any kind of
      >> certification. But if we could ... that suggest to me that there is
      >> something there to be measured.
      >> How could we do that in ways we could live with?
      > Hm. As far as I know, the primary drivers for commercial
      > certifications so far have been
      > a) something we hope has predictive validity about
      > whatever is being certified, and
      > b) some representation that can be made to third-parties,
      > generally in respect of our ability to sell something,
      > some quality of which the certification nominally guarantees.
      > In other words, absent such a driver, probably no one is
      > going to pay for the creation and maintenance of the
      > certification mechanism.
      > If we were to look at those "certification" mechanisms that
      > many of us might agree are among the most valid and useful,
      > such as commercial pilots' and physicians' licenses, I think
      > we'd find that much of the meaning of the credential (per se)
      > is that the individual was compelled to complete a course of
      > study AND what amounts to a significant period of "apprenticeship"
      > with a large one-on-one component. In other words, SOMEone
      > (probably a NUMBER of someones) has had direct contact with
      > the individual's work over a period of time. I think
      > architects and LPEs undergo cognate periods of scrutiny
      > by those with significant experience.
      > No one would argue that these approaches are perfect, nor
      > that they are not costly. They are intensive and time-
      > consuming. Curiously, they arguably can be said to scale.
      > For software, hm, and also for self-evaluation: what
      > dimensions are REALLY important? Is today's skill what
      > counts, or is it the ability to perceive business issues
      > and drivers and quickly to craft a means to address them
      > effectively?
      > Could a TEAM be more properly certified than an individual?
      > What would really be most useful to the likely users of
      > the certification? Would this be like: "what stories
      > would apply to its use?" Since developers are necessarily
      > among the "users" - what stories must this contemplated
      > certification implement well to satisfy developers?

      I think one good indicator of skill is the "list of apps completed so
      far". Show-casing what you've done previously.

      [Rant: Certification in general just does not work today in the s/w
      industry. I don't trust certified ppl, in fact I look for people that
      are not certified because a certification would imply a belief in
      todays certification systems, which does not work and thus degrades
      the value of the applicant in my eyes.

      Now the certification Ron and Chet does might be different, I don't
      really know the certification process (I live in Sweden and haven't
      bump in to anyone certified by them).]


    • Adrian Howard
      ... Not sure that it is mind - I m often wrong :) Adrian
      Message 94 of 94 , May 8 7:55 AM
      • 0 Attachment
        On 6 May 2009, at 04:08, Ron Jeffries wrote:

        > Hello, Adrian. On Tuesday, May 5, 2009, at 11:37:58 AM, you wrote:
        >> That said, my perception is that the odds of getting a bad candidate
        >> at interview increase if they have certifications listed on their
        >> resume.
        >> My post hoc rationalisation for this experiences is not so much
        >> related to the certification programme itself (although many of them
        >> have severe problems), but to the sort of folk they can attract.
        >> The less creative, by-the-book, list-following developer seems to see
        >> certification programmes as a way to prove their competence. The more
        >> creative, problem-oriented, passionate developers seem to spend their
        >> free chunk of career development time building something cool, or
        >> learning erlang, or contributing to open source projects or...
        > Interesting model. I can see how it might be.

        Not sure that it is mind - I'm often wrong :)

      Your message has been successfully submitted and would be delivered to recipients shortly.