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

Collocated vs Dispersed...which is better?

Expand Messages
  • Joseph Little
    Hi all, I am looking for evidence.... Here s my hypothesis, assumptions and basic arguments. What facts / experiences do you have to refute or confirm the
    Message 1 of 62 , Jul 30, 2008
    • 0 Attachment
      Hi all,

      I am looking for evidence....

      Here's my hypothesis, assumptions and basic arguments.  What facts / experiences do you have to refute or confirm the hypothesis?

      Hypothesis:
      A team is about 2x more productive doing scrum / agile collocated than dispersed.

      Dispersed = no two people are in the same location.  eg, each person is working at home.

      Assumptions:
      a. Disruption impediments are handled to the degree that disruption is equal in either case.
      b. There is no physical reason why the team cannot collocate (eg, no extreme commutes).
      c. There is no significant cost advantage to the firm in dispersion (or collocation) other than the productivity effect.
      d. A collocated team can get the advantages of a team room of a comfortable size, and also have space nearby for reasonable privacy.
      e. A dispersed team has "the best" dispersed support, except for high quality continuous video-conferencing.  Video conferencing can be done on a shoe-string (web cams) and, on rare occasions, in an ok set-up.
      f. No other factor is different (I can't think of any other differences, but perhaps I am overlooking something...tell me if you think I am).

      Discussion:
      I have a client who believes they have facts that say a dispersed team (in this case, mostly working at home) is "better" than a collocated team. "Better" for them has a different meaning that productive.

      An argument is made that people in large offices get interrupted a lot; more than at home.  I am assuming that, to the degree that that might be true, the SM and the Team reduce that impediment so that the disruption is equal.  I further assume that the cost of dealing with that impediment is small.

      Let me add that...disruptions at home an infinitely less visible to the team and SM, so that they seldom are addressed ("unless the coder tells his wife to go to hell", as someone once put it). 

      I have an additional hypothesis that collocation is also better in all respects (except in a few people's opinion), but let's do that one later.

      I do think that non-collocated teams can work, and sometimes be very productive.  Jeff Sutherland has data to support that.

      I am just saying that no team should be allowed to give up on collocation "just because" (ie, without some powerful reasons), since they are cutting off half their productivity.

      My personal experiences are that collocation is so obviously better for new teams learning Agile, that no data is needed.  But I also want to have data (if you all have some), to convince others. 
      I personally find it extremely difficult to be effective as a coach if the team is not collocated.
      Collocation helps a new team do Forming, Storming, Norming (better, faster).
      My experience is that non-collocated "teams" never really form as a team.  Or, if we call it a team, it has nothing like the same intensity and spirit.
      Xebia does high-quality high-productivity distributed (2 pod) development.  BUT only after collocating each team for 10 weeks.  I note that they do *not* do dispersed development (no two people in the same location). See Jeff Sutherland's paper on this. 

      Evidence:
      The following would be helpful.
      a. studies, papers on studies, books on studies, etc
      b. references to web sites
      c. your personal experiences
      d. references to experts
      e. references to previous threads here or in other lists
      f. other??

      Thanks!
      Joe



      Joseph Little
      Agile coach, MBA, CST
      Kitty Hawk Consulting, Inc.
      704-376-8881 (Charlotte)
      917-887-1669 (cell)
      http://www.kittyhawkconsulting.com/
      http://leanagiletraining.com/
      Blog: "Agile & Business" (Google that)

    • davenicolette
      Mike, There was some discussion recently (here, I think) about how anyone who has a good idea about how to run software projects likes to call whatever they re
      Message 62 of 62 , Aug 1, 2008
      • 0 Attachment
        Mike,

        There was some discussion recently (here, I think) about how anyone
        who has a good idea about how to run software projects likes to call
        whatever they're doing "agile." Puzzling, really.

        I'm not stuck on buzzwords; more interested in effective delivery of
        value, however it's done. It seems that recently people are getting
        confused about what the word "agile" is supposed to mean. It's
        starting to seem as if "agile" is just a broad synonym for "good."

        So, if you're not collocated you're not "agile." Or so people say. Why
        do you care? If collocation isn't appropriate in a given case, then I
        guess "agile" isn't the way to go in that case. That's not a crime.

        Of course, if you've got something to sell these days, it's de rigeur
        to label it "agile." Might not sell very well otherwise, just now. I
        can relate to that. I've got bills to pay, too.

        I will reiterate, and stand by, my earlier statement that it's a red
        herring to claim that anyone has ever seriously proposed having 200+
        people work in the same room. Or is the proper term "strawman
        argument?" See, there's the imprecision of human language, again.

        Dave


        --- In scrumdevelopment@yahoogroups.com, mike.dwyer1@... wrote:
        >
        > Dave
        > Condescending?!? Sorry, I'll cop to frustrated, flip, sarcastic,
        but condescending? Pontifications that if you are not collocated
        your not agile are qualifying comments for that term.
        > Stand you for a round next week.
        > Best to all
        >
        > Sent via BlackBerry by AT&T
        >
        > -----Original Message-----
        > From: "davenicolette" <dnicolet@...>
        >
        > Date: Fri, 01 Aug 2008 14:11:16
        > To: <scrumdevelopment@yahoogroups.com>
        > Subject: [scrumdevelopment] Re: Collocated vs Dispersed...which is
        better?
        >
        >
        > Mike,
        >
        > I appreciate all that you say here. It's a far better statement of the
        > position than the previous post. Less dismissive. You still seem to
        > assume that you're charting new ground when working with larger
        > organizations. I'll grant that the community as a whole hasn't made
        > much headway at that level and I would not say we have crossed the
        > chasm with agile just yet, but people have been dealing with
        > medium-sized enterprise projects (in the range of the low hundreds of
        > participants) already. It's great that your firm (among others) has
        > practical advice and help to offer. I only hope the condescending tone
        > I detected earlier was a figment of my textual imagination.
        >
        > Cheers,
        > Dave
        >
        > --- In scrumdevelopment@yahoogroups.com, "Mike Dwyer"
        > <mike.dwyer1@> wrote:
        > >
        > > Dave
        > >
        > > There are people who think Agile and collocation have no
        boundaries and
        > > when, on other threads, we look for failures there is never a
        > mention of the
        > > scaling limits of teams and programs. The space I occupy in Agile
        > and Scrum
        > > deals with groups greater than 100 and engaged in all aspects of IT
        > > (operations, production, maintenance, and software development) as
        > well as
        > > product development. While 200 people sounds like a lot, on this
        > side of
        > > the chasm it represents a small portion of a large organization.
        > >
        > >
        > >
        > > So what happens when a large (say greater than 1,000 people) group
        doing
        > > software products within a big (say Fortune 500) company starts on
        > the Agile
        > > path? To carry your fish analogy a bit farther - that path does
        > have many
        > > places for 'dead fish smells' to appear and from my perspective
        the red
        > > herrings in the path are caused by holding dear what we are most
        > comfortable
        > > with and not inspecting and adapting to meet the new needs of our
        > customers.
        > >
        > >
        > >
        > > As to the notion of placing all these people in one room - That is
        not a
        > > good idea*, but what is? 30 separate rooms? In one building? on a
        > single
        > > campus, in the same time zone? How have people organized the work?
        > What
        > > about budget? Skills? Skill mix and things like training,
        > implementation,
        > > infrastructure?
        > >
        > >
        > >
        > > This crossing the chasm thing we started to talk about last year
        is not
        > > over, in fact now that more of you are getting to this side of the
        > work, you
        > > might want to hear what we have been doing.
        > >
        > >
        > >
        > >
        > >
        > > *For some interesting reading and thoughts about large groups in small
        > > places, you may find Calhoun's "Behavioral Sink"
        > > (http://en.wikipedia.org/wiki/Behavioral_sink) a good starting
        point.
        > >
        > > Michael F. Dwyer
        > >
        > >
        > >
        > > "Planning constantly peers into the future for indications as to
        where a
        > > solution may emerge."
        > >
        > > "A Plan is a complex situation, adapting to an emerging solution."
        > >
        > > -----Original Message-----
        > > From: scrumdevelopment@yahoogroups.com
        > > [mailto:scrumdevelopment@yahoogroups.com] On Behalf Of davenicolette
        > > Sent: Thursday, July 31, 2008 9:15 PM
        > > To: scrumdevelopment@yahoogroups.com
        > > Subject: [scrumdevelopment] Re: Collocated vs Dispersed...which is
        > better?
        > >
        > >
        > >
        > > Hi Mike,
        > >
        > > >But hey it might be
        > > > interesting to try having 200+ people, broken into 17 communities
        > > and about
        > > > 30 teams all in one room to test the boundaries of collocation.
        > >
        > > The session on the Ambassador Model is certainly relevant and sounds
        > > like it will be very practical and informative. However, to use a red
        > > herring like the statement quoted above as a sales pitch for it is
        > > really unnecessary.
        > >
        > > I've never heard of anyone who thinks scaling agile means having 30
        > > teams all in one room. Of course, each of the 30 teams individually
        > > should be collocated, to the extent practical. That's fundamental. If
        > > you don't understand that, then look me up. ;-)
        > >
        > > Dave
        > >
        > > --- In scrumdevelopment@ <mailto:scrumdevelopment%40yahoogroups.com>
        > > yahoogroups.com, "Mike Dwyer"
        > > <mike.dwyer1@> wrote:
        > > >
        > > > This is all very nice, true and doesn't scale. But hey it might be
        > > > interesting to try having 200+ people, broken into 17 communities
        > > and about
        > > > 30 teams all in one room to test the boundaries of collocation. I
        > think
        > > > that the people who bought WildCard tried it and they might have
        some
        > > > insights, but I don't know where they are. Another idea has come
        > to the
        > > > minds of two of my colleagues who face this problem on a regular
        > basis.
        > > > They have done a couple of standups on the subject and one of them
        > is on
        > > > Tuesday Afternoon. If you get a chance you might want to listen to
        > > Giora
        > > > Morein and George Schlitz at Agile 2008 as they talk about their
        > use of
        > > > ambassadors. http://submissions.
        > > <http://submissions.agile2008.org/node/4718> agile2008.org/node/4718
        > > >
        > > >
        > > >
        > > > If you can't make it or you want to talk shop about what happens
        > > when Agile
        > > > takes on the Enterprise, look us up.
        > > >
        > > > Michael F. Dwyer
        > > >
        > > > BigVisible Solutions
        > > >
        > > >
        > > >
        > > > "Planning constantly peers into the future for indications as to
        > where a
        > > > solution may emerge."
        > > >
        > > > "A Plan is a complex situation, adapting to an emerging solution."
        > > >
        > > > -----Original Message-----
        > > > From: scrumdevelopment@ <mailto:scrumdevelopment%40yahoogroups.com>
        > > yahoogroups.com
        > > > [mailto:scrumdevelopment@
        <mailto:scrumdevelopment%40yahoogroups.com>
        > > yahoogroups.com] On Behalf Of Ron Jeffries
        > > > Sent: Thursday, July 31, 2008 5:19 PM
        > > > To: scrumdevelopment@ <mailto:scrumdevelopment%40yahoogroups.com>
        > > yahoogroups.com
        > > > Subject: Re: [scrumdevelopment] Re: Collocated vs
        Dispersed...which is
        > > > better?
        > > >
        > > >
        > > >
        > > > Hello, Ken. On Thursday, July 31, 2008, at 10:53:32 AM, you wrote:
        > > >
        > > > > Now the real question is, "What's the best choice?". Clearly
        > increased
        > > > > communication is worth a lot and if you're going to have to
        maintain
        > > > > and office and pay everyone the same amount regardless of the
        > choice,
        > > > > I'd go with the increased communication. However, I'd be lying
        if I
        > > > > didn't say dispersed agile didn't have it's attractions to me.
        > > >
        > > > Me too. You don't see me working in an office. But oddly enough,
        > > > when I want to work with Chet, we meet somewhere.
        > > >
        > > > Ron Jeffries
        > > > www.XProgramming.com
        > > > The central "e" in "Jeffries" is silent ... and invisible.
        > > >
        > >
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.