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

RE: [agile-usability] Customer isn't a UI Designer

Expand Messages
  • Dan Rawsthorne
    Our newest version of the Use Case course combines usability discussions and use cases in an agile environment. We use personas as validation mechanisms as
    Message 1 of 27 , Sep 21, 2004
    • 0 Attachment

      Our newest version of the Use Case course combines usability discussions and use cases in an agile environment. We use personas as validation mechanisms as well as to help derive requirements for the use cases. It will be a small section of the book I’m working on. I look forward to seeing what you come up with.

      Dan Rawsthorne, PhD, Sr. Consultant
      www.netobjectives.com
      DrDan@...
      office: 425-269-8628

      Net Objectives' vision is effective software development without suffering. Our mission is to assist software development teams in accomplishing this through a combination of training and mentoring.

      -----Original Message-----
      From: Edmund Schweppe [mailto:schweppe@...]
      Sent: Tuesday, September 21, 2004 6:12 PM
      To: agile-usability@yahoogroups.com
      Subject: Re: [agile-usability] Customer isn't a UI Designer

       

      David Anderson wrote:

      > To conclude where I started - it seems that no one (that we
      > collectively know of) has yet written about Scenarios and Personas in
      > conjunction with agile development.

      I'm currently attending SD East 2004 in Boston [1], and Larry
      Constantine gave an interesting class on trying to mix usage-centered
      design with Agile development. One story he told was of the McKesson HED
      project (a medical management system which had to get through FDA
      certification). For the McKesson project, a usability team first went
      through a full-bore usage-centered design process, building task cases,
      abstract and figurative prototypes, and the rest of the whole nine
      yards. Then, the *usabilility team* took on the XP Customer role,
      feeding stories to the developers in short iterations and giving them
      constant feedback.

      > I asked the question because I work for Microsoft nowadays and my
      > team has just released a draft of the forthcoming MSF Agile process.
      > MSF Agile does use Scenarios and Personas where XP uses User Stories.
      > Our team will be publishing material on how to create and use
      > Scenarios and Personas in agile development and I wanted to insure
      > that we referenced any prior art when we went about that.

      I'll look forward to seeing what you folks come up with.

      [1] That's Boston, Massachusetts, USA; not Boston, Lincolnshire, UK.
      Just so you know ...

      --
      Edmund Schweppe -- schweppe@... -- http://schweppe.home.tiac.net
      The opinions expressed herein are at best coincidentally related to
      those of any past, present or future employer.



      ______________________________________________________________________
      This email has been scanned by the MessageLabs Email Security System.
      For more information please visit http://www.messagelabs.com/email
      ______________________________________________________________________
    • Dan Rawsthorne
      I notice that Jeff Patton is delivering a tutorial at OOPSLA on this topic, see http://www.oopsla.org/2004/ShowEvent.do?id=105 Dan ;-) Dan Rawsthorne, PhD,
      Message 2 of 27 , Sep 22, 2004
      • 0 Attachment

        I notice that Jeff Patton is delivering a tutorial at OOPSLA on this topic, see http://www.oopsla.org/2004/ShowEvent.do?id=105

         

        Dan  ;-)

        Dan Rawsthorne, PhD, Sr. Consultant
        www.netobjectives.com
        DrDan@...
        office: 425-269-8628

        Net Objectives' vision is effective software development without suffering. Our mission is to assist software development teams in accomplishing this through a combination of training and mentoring.

        -----Original Message-----
        From: Edmund Schweppe [mailto:schweppe@...]
        Sent: Tuesday, September 21, 2004 6:12 PM
        To: agile-usability@yahoogroups.com
        Subject: Re: [agile-usability] Customer isn't a UI Designer

         

        David Anderson wrote:

        > To conclude where I started - it seems that no one (that we
        > collectively know of) has yet written about Scenarios and Personas in
        > conjunction with agile development.

        I'm currently attending SD East 2004 in Boston [1], and Larry
        Constantine gave an interesting class on trying to mix usage-centered
        design with Agile development. One story he told was of the McKesson HED
        project (a medical management system which had to get through FDA
        certification). For the McKesson project, a usability team first went
        through a full-bore usage-centered design process, building task cases,
        abstract and figurative prototypes, and the rest of the whole nine
        yards. Then, the *usabilility team* took on the XP Customer role,
        feeding stories to the developers in short iterations and giving them
        constant feedback.

        > I asked the question because I work for Microsoft nowadays and my
        > team has just released a draft of the forthcoming MSF Agile process.
        > MSF Agile does use Scenarios and Personas where XP uses User Stories.
        > Our team will be publishing material on how to create and use
        > Scenarios and Personas in agile development and I wanted to insure
        > that we referenced any prior art when we went about that.

        I'll look forward to seeing what you folks come up with.

        [1] That's Boston, Massachusetts, USA; not Boston, Lincolnshire, UK.
        Just so you know ...

        --
        Edmund Schweppe -- schweppe@... -- http://schweppe.home.tiac.net
        The opinions expressed herein are at best coincidentally related to
        those of any past, present or future employer.



        ______________________________________________________________________
        This email has been scanned by the MessageLabs Email Security System.
        For more information please visit http://www.messagelabs.com/email
        ______________________________________________________________________
      • Petteri Hiisilä
        Hi David & everybody else, A great final for a great thread. Thanks. To celebrate this event, I ll attach Ron Jeffries signature in the bottom of my mail! I
        Message 3 of 27 , Sep 22, 2004
        • 0 Attachment
          Hi David & everybody else,

          A great final for a great thread. Thanks. To celebrate this event, I'll attach Ron Jeffries' signature in the bottom of my mail! I hope we now agree that all these "new" methodologies can produce incredibly successful results, compared to any "old" approach. (Assuming you know what you're doing, of course...)

          --> Just imagine what we could do, if we could combine them?

          The Cooper HQ (Sue Cooper & Kim Goodwin) promised in their mailing list that they are working on an "official" book about personas and goal-directed scenarios ("user stories"), but the progress is slow. The book project is in the a.s.a.p. lane anyway.

          In the meantime you Agile Experts could experiment with the idea in your own projects, and write your own book, how to use the knowledge about users in the agile development process? About the real effects and the side-effects of putting more effort in user research into an agile process. What's good about it, what's bad about it? Can we benefit from the goods, and live with the bads? After all, if you're going to make an omelet, you're going to break some eggs.

          The details of persona creation is already available through the Cooper U Interaction Design Practicum in just four days. I took it. Jeff has taken it. A few hundred other people have taken it. It's great. Developers should take it too, and managers. Not just the "UI designers". The stuff they teach there is fundamental, very useful down-to-earth information.

          If you want to know how to really boost your understanding about your real users' real goals and needs, I warmly recommend that course. After that you will know what your users need, and what's even more important: what they don't need! No more useless features to code! Simpler, but more usable products! To me it has been the best move I've ever done in my career. And I have done a lot of moves in my career.

          Piece and love,
          Petteri

          --
          Petteri Hiisilä
          Palveluarkkitehti / Interaction Designer /
          Alma Media Interactive Oy / NWS /
          +358505050123 / petteri.hiisila@...

          "Here is Edward Bear, coming downstairs now, bump, bump, bump, on the back
          of his head. It is, as far as he knows, the only way of coming downstairs,
          but sometimes he feels that there really is another way, if only he could
          stop bumping for a moment and think of it. And then he feels that perhaps
          there isn't." (A. A. Milne; copied from Ron Jeffries <http://www.xprogramming.com/> )

          David Anderson wrote:
          Having kicked off the thread on scenarios and personas in agile
          methods, I thought I'd try to sum some of it up.

          To the guy who suggested I "go hire someone" - I am someone - maybe
          I can hire myself? ;-) - see http://www.uidesign.net/ for details.

          On the extreme notion that up front design is unnecessary, I would
          first like to quote the great agile philosopher Philip Bradley (a
          member of the original FDD team) "Being the UI Architect is like
          being the England football manager - there are 11 million guys in
          the pub on a Friday night that think they can do the job better than
          you!"

          Jakob Nielsen put it a little more scientifically when he called
          it "'throwing mud at wall' usability."

          It's clear to me that there is value in persona and scenario design
          as a proxy for the customer. Customer involvement can be hard to
          schedule and could be a constraint and bottleneck when a good
          interaction designer can make progress without it.

          What's important that this activity doesn't become too big before
          the user gets involved.

          So there could be two approaches - create a few scenarios for a few
          personas and go code them - then test that code with users. I'd much
          prefer the idea that this testing is done by a trained usability
          professional and isn't just some "acceptance" test performed by the
          dev team with a sample customer.

          A second approach might be to paper prototype and go test those with
          real users. One aim of this would be to validate the persona and
          scenario definition as accurate. Another alternative might simply be
          to have a persona and scenario review with a focus group.

          Either way, we are replacing - just code it and hands on customer
          involvement - with expertise. But that might just be better. My
          resident customer if I have one may be a lousy UI designer and
          unrepresentative of my wider user base.

          To conclude where I started - it seems that no one (that we
          collectively know of) has yet written about Scenarios and Personas
          in conjunction with agile development. I asked the question because
          I work for Microsoft nowadays and my team has just released a draft
          of the forthcoming MSF Agile process. MSF Agile does use Scenarios
          and Personas where XP uses User Stories. Our team will be publishing
          material on how to create and use Scenarios and Personas in agile
          development and I wanted to insure that we referenced any prior art
          when we went about that.

          I'll post links to the next draft of the MSF Agile material on
          uidesign.net when the time comes.

          Thanks,

          David




          -- 
          Petteri Hiisilä
          Palveluarkkitehti / Interaction Designer /
          Alma Media Interactive Oy / NWS /
          +358505050123 / petteri.hiisila@...
          
          "I was told there's a miracle for each day that I try"
           - John Petrucci
          
          
        • David Anderson
          Thanks! I know Larry and his work. I also value it but we ve gone down the concrete specific example route with Personas and Scenarios and we ll see how it
          Message 4 of 27 , Sep 22, 2004
          • 0 Attachment
            Thanks! I know Larry and his work. I also value it but we've gone
            down the concrete specific example route with Personas and Scenarios
            and we'll see how it works out.

            :-) I'm in Redmond WA so I assumed Boston MA !

            --- In agile-usability@yahoogroups.com, Edmund Schweppe
            <schweppe@i...> wrote:
            > David Anderson wrote:
            >
            > > To conclude where I started - it seems that no one (that we
            > > collectively know of) has yet written about Scenarios and
            Personas in
            > > conjunction with agile development.
            >
            > I'm currently attending SD East 2004 in Boston [1], and Larry
            > Constantine gave an interesting class on trying to mix usage-
            centered
            > design with Agile development. One story he told was of the
            McKesson HED
            > project (a medical management system which had to get through FDA
            > certification). For the McKesson project, a usability team first
            went
            > through a full-bore usage-centered design process, building task
            cases,
            > abstract and figurative prototypes, and the rest of the whole nine
            > yards. Then, the *usabilility team* took on the XP Customer role,
            > feeding stories to the developers in short iterations and giving
            them
            > constant feedback.
            >
            > > I asked the question because I work for Microsoft nowadays and my
            > > team has just released a draft of the forthcoming MSF Agile
            process.
            > > MSF Agile does use Scenarios and Personas where XP uses User
            Stories.
            > > Our team will be publishing material on how to create and use
            > > Scenarios and Personas in agile development and I wanted to
            insure
            > > that we referenced any prior art when we went about that.
            >
            > I'll look forward to seeing what you folks come up with.
            >
            > [1] That's Boston, Massachusetts, USA; not Boston, Lincolnshire,
            UK.
            > Just so you know ...
            >
            > --
            > Edmund Schweppe -- schweppe@i... -- http://schweppe.home.tiac.net
            > The opinions expressed herein are at best coincidentally related to
            > those of any past, present or future employer.
          • David Anderson
            I m interested to know how you reconcile the abstract nature of use cases and actors against the concrete nature of personas. The persona/scenario movements
            Message 5 of 27 , Sep 22, 2004
            • 0 Attachment
              I'm interested to know how you reconcile the abstract nature of use
              cases and actors against the concrete nature of personas. The
              persona/scenario movements actually started in response to the
              abstract nature of use cases and in another world of marketing,
              demographics.

              David

              --- In agile-usability@yahoogroups.com, "Dan Rawsthorne"
              <DrDan@N...> wrote:
              > Our newest version of the Use Case course combines usability
              discussions
              > and use cases in an agile environment. We use personas as
              validation
              > mechanisms as well as to help derive requirements for the use
              cases. It
              > will be a small section of the book I'm working on. I look forward
              to
              > seeing what you come up with.
              >
              > Dan Rawsthorne, PhD, Sr. Consultant
              > www.netobjectives.com
              > DrDan@N...
              > office: 425-269-8628
              >
              > Net Objectives' vision is effective software development without
              > suffering. Our mission is to assist software development teams in
              > accomplishing this through a combination of training and
              mentoring.
              >
              > -----Original Message-----
              > From: Edmund Schweppe [mailto:schweppe@i...]
              > Sent: Tuesday, September 21, 2004 6:12 PM
              > To: agile-usability@yahoogroups.com
              > Subject: Re: [agile-usability] Customer isn't a UI Designer
              >
              >
              >
              > David Anderson wrote:
              >
              > > To conclude where I started - it seems that no one (that we
              > > collectively know of) has yet written about Scenarios and
              Personas in
              > > conjunction with agile development.
              >
              > I'm currently attending SD East 2004 in Boston [1], and Larry
              > Constantine gave an interesting class on trying to mix usage-
              centered
              > design with Agile development. One story he told was of the
              McKesson HED
              > project (a medical management system which had to get through FDA
              > certification). For the McKesson project, a usability team first
              went
              > through a full-bore usage-centered design process, building task
              cases,
              > abstract and figurative prototypes, and the rest of the whole nine
              > yards. Then, the *usabilility team* took on the XP Customer role,
              > feeding stories to the developers in short iterations and giving
              them
              > constant feedback.
              >
              > > I asked the question because I work for Microsoft nowadays and my
              > > team has just released a draft of the forthcoming MSF Agile
              process.
              > > MSF Agile does use Scenarios and Personas where XP uses User
              Stories.
              > > Our team will be publishing material on how to create and use
              > > Scenarios and Personas in agile development and I wanted to
              insure
              > > that we referenced any prior art when we went about that.
              >
              > I'll look forward to seeing what you folks come up with.
              >
              > [1] That's Boston, Massachusetts, USA; not Boston, Lincolnshire,
              UK.
              > Just so you know ...
              >
              > --
              > Edmund Schweppe -- schweppe@i... -- http://schweppe.home.tiac.net
              > The opinions expressed herein are at best coincidentally related to
              > those of any past, present or future employer.
              >
              >
              >
              >
              >
              >
              > Yahoo! Groups Sponsor
              >
              >
              >
              > ADVERTISEMENT
              >
              >
              <http://us.ard.yahoo.com/SIG=129i8fblm/M=295196.4901138.6071305.30011
              76/
              >
              D=groups/S=1705007709:HM/EXP=1095902342/A=2128215/R=0/SIG=10se96mf6/*
              htt
              > p:/companion.yahoo.com> click here
              >
              >
              >
              > <http://us.adserver.yahoo.com/l?
              M=295196.4901138.6071305.3001176/D=group
              > s/S=:HM/A=2128215/rand=328382759>
              >
              >
              >
              > _____
              >
              > Yahoo! Groups Links
              >
              > * To visit your group on the web, go to:
              > http://groups.yahoo.com/group/agile-usability/
              >
              > * To unsubscribe from this group, send an email to:
              > agile-usability-unsubscribe@yahoogroups.com
              > <mailto:agile-usability-unsubscribe@yahoogroups.com?
              subject=Unsubscribe>
              >
              >
              > * Your use of Yahoo! Groups is subject to the Yahoo! Terms of
              > Service <http://docs.yahoo.com/info/terms/> .
              >
              >
              >
              _____________________________________________________________________
              _
              > This email has been scanned by the MessageLabs Email Security
              System.
              > For more information please visit http://www.messagelabs.com/email
              >
              _____________________________________________________________________
              _
            • Jeff Patton
              ... I can pipe in with a little experience here - or at least an opinion. These days I m finding it s fairly important to have lots of ways of looking at the
              Message 6 of 27 , Sep 22, 2004
              • 0 Attachment
                --- In agile-usability@yahoogroups.com, "David Anderson"
                <netherby_uk@y...> wrote:
                > I'm interested to know how you reconcile the abstract nature of use
                > cases and actors against the concrete nature of personas. The
                > persona/scenario movements actually started in response to the
                > abstract nature of use cases and in another world of marketing,
                > demographics.

                I can pipe in with a little experience here - or at least an
                opinion. These days I'm finding it's fairly important to have lots
                of ways of looking at the design of the software. As I've advertised
                in other posts, I primarily lean on Constantine & Lockwoods U-CD. As
                a result I'm usually swimming in fairly abstract artifacts: user
                roles - which aren't at all personas, tasks or essential use cases -
                which are a specialized form of use case, and interaction contexts -
                which aren't like an artifact I've seen elsewhere. All these
                artificats are pretty strait forward to create and make excellent
                vehicles to carry forward into development. U-CD tasks become my XP
                stories. Interaction contexts are used to build a navigation map and
                explain to everyone workflow of the software at a high level, and
                roles give specific guidance on how the user interface for tasks
                should be designed and validated.

                That all sounds great, but at times I and the teams I work with start
                to swim in abstraction. When we start talking about roles doing
                tasks in a context, we sometimes forget about the real people using
                software to do work. At times folks will suggest a role needs to do
                a certain task and something doesn't sound right. At those times
                I'll say "tell me a story about how that would work in real life."
                The person I'm talking with often starts by using a role name, and
                I'll cut them short and ask them to give that role a real name -
                David or Dan for instance. I'll suggest that knowing what we know
                about the role can we imagine a real person filling that role and
                tell a very concrete story about why they'd perform a particular task
                and how they'd go about it.

                Inventing a persona on the fly and writing a scenario for that
                persona helps us work out the kinks in our abstractions. The
                abstractions help us better break things up for development, help us
                better find and build the objects in our software. For me the
                techniqes compliment each other.

                I've also found a provisional persona and scenarios helpful in
                explaining design to stakeholders, testers, or others who might have
                trouble with the abstraction. I've found scenarios helpful when I'm
                dealing with software where the person using the system can do many
                things at any time. Documenting that in a use case or task case is a
                pain. Just describing all the possible interactions is unclear. the
                concrete story/scenario shows how all the possible interactions can
                be used together to perform real work, meet real user goals.

                Having spent a little time with the Cooper folks and learned from
                them how they create personas I'm mindful of the rigor they put into
                them. I suspect Microsoft puts equal rigor into theirs. Knowing
                what I know about personas, I can't call the "thing" I make a persona
                in good conscience. But I do find more a more concrete
                representation of the user built from whatever moddest facts and
                assumptions I have is helpful.

                Thanks David & Dan for posting. I'm enjoying all the smart people
                posting on this list!

                -Jeff
              • Chris Pehura
                I usually use artifacts that map personas to actors. One time, because of the artifact s audience, personas equated to actors. For me, artifacts are a tool for
                Message 7 of 27 , Sep 22, 2004
                • 0 Attachment
                  I usually use artifacts that map personas to actors. One time, because of the artifact's audience, personas equated to actors.
                  For me, artifacts are a tool for communication, and their usage and shape is determined by those using them to communicate.
                  Sometimes I mix in the "little people" methodology too. What ever is effective.
                  -----Original Message-----
                  From: Jeff Patton [mailto:jpatton@...]
                  Sent: Wednesday, September 22, 2004 9:06 PM
                  To: agile-usability@yahoogroups.com
                  Subject: [agile-usability] Re: Customer isn't a UI Designer

                  --- In agile-usability@yahoogroups.com, "David Anderson"
                  <netherby_uk@y...> wrote:
                  > I'm interested to know how you reconcile the abstract nature of use
                  > cases and actors against the concrete nature of personas. The
                  > persona/scenario movements actually started in response to the
                  > abstract nature of use cases and in another world of marketing,
                  > demographics.

                  I can pipe in with a little experience here - or at least an
                  opinion.  These days I'm finding it's fairly important to have lots
                  of ways of looking at the design of the software.  As I've advertised
                  in other posts, I primarily lean on Constantine & Lockwoods U-CD.  As
                  a result I'm usually swimming in fairly abstract artifacts: user
                  roles - which aren't at all personas, tasks or essential use cases -
                  which are a specialized form of use case, and interaction contexts -
                  which aren't like an artifact I've seen elsewhere.  All these
                  artificats are pretty strait forward to create and make excellent
                  vehicles to carry forward into development.  U-CD tasks become my XP
                  stories.  Interaction contexts are used to build a navigation map and
                  explain to everyone workflow of the software at a high level, and
                  roles give specific guidance on how the user interface for tasks
                  should be designed and validated.

                  That all sounds great, but at times I and the teams I work with start
                  to swim in abstraction.  When we start talking about roles doing
                  tasks in a context, we sometimes forget about the real people using
                  software to do work.  At times folks will suggest a role needs to do
                  a certain task and something doesn't sound right.  At those times
                  I'll say "tell me a story about how that would work in real life." 
                  The person I'm talking with often starts by using a role name, and
                  I'll cut them short and ask them to give that role a real name -
                  David or Dan for instance.  I'll suggest that knowing what we know
                  about the role can we imagine a real person filling that role and
                  tell a very concrete story about why they'd perform a particular task
                  and how they'd go about it. 

                  Inventing a persona on the fly and writing a scenario for that
                  persona helps us work out the kinks in our abstractions.  The
                  abstractions help us better break things up for development, help us
                  better find and build the objects in our software.  For me the
                  techniqes compliment each other.

                  I've also found a provisional persona and scenarios helpful in
                  explaining design to stakeholders, testers, or others who might have
                  trouble with the abstraction.  I've found scenarios helpful when I'm
                  dealing with software where the person using the system can do many
                  things at any time.  Documenting that in a use case or task case is a
                  pain.  Just describing all the possible interactions is unclear.  the
                  concrete story/scenario shows how all the possible interactions can
                  be used together to perform real work, meet real user goals.  

                  Having spent a little time with the Cooper folks and learned from
                  them how they create personas I'm mindful of the rigor they put into
                  them.  I suspect Microsoft puts equal rigor into theirs.  Knowing
                  what I know about personas, I can't call the "thing" I make a persona
                  in good conscience.  But I do find more a more concrete
                  representation of the user built from whatever moddest facts and
                  assumptions I have is helpful.

                  Thanks David & Dan for posting.  I'm enjoying all the smart people
                  posting on this list!   

                  -Jeff



                • Jeff Patton
                  ... Here s what I understand about personas: They aren t creative writing. They re built out of concrete data gathered from interviews or other sources. The
                  Message 8 of 27 , Sep 22, 2004
                  • 0 Attachment
                    --- In agile-usability@yahoogroups.com, "David Anderson"
                    <netherby_uk@y...> wrote:
                    > Having kicked off the thread on scenarios and personas in agile
                    > methods, I thought I'd try to sum some of it up.
                    ...

                    > It's clear to me that there is value in persona and scenario design
                    > as a proxy for the customer. Customer involvement can be hard to
                    > schedule and could be a constraint and bottleneck when a good
                    > interaction designer can make progress without it.
                    >
                    > What's important that this activity doesn't become too big before
                    > the user gets involved.

                    Here's what I understand about personas: They aren't creative
                    writing. They're built out of concrete data gathered from interviews
                    or other sources. The more data gathered, the more likely the
                    persona is to represent a constituency the software will serve. When
                    assembled every fact or quote in the persona was ripped out of an
                    interview - nothing's fabricated. Although they may look like a
                    single person, they're a composite of lots of people. they're a
                    stereotype of a customer/user we think we'll be writing for.

                    Knowing what I think I know, good personas are time consuming to
                    create. No offense to the very smart Cooper folks, but they seem a
                    bit heavy for the projects I'm involved in. It seems hard to build
                    good personas quickly.

                    ...
                    > A second approach might be to paper prototype and go test those
                    with
                    > real users. One aim of this would be to validate the persona and
                    > scenario definition as accurate.

                    That sentance stopped me a little. Since a persona is a composite of
                    common traits, a persona isn't really any one person. As a result,
                    any real person is likely to vary from the persona. I'd worry that
                    testing software with the intent of validating a persona would,
                    unless you tested with lots of people, result in invalidating almost
                    any persona. But, I'm not speaking from experience here. I think it
                    would take a sharp person with a clear understanding of the data
                    behind the persona to recognize that a particular individuals
                    contradictory opinions or behavior does or doesn't invalidate a
                    persona. I suspect the same may be less true of scenarios...

                    Looking back at this discussion on personas I recall reading a post
                    in another group a couple years ago [I searched and couldn't find
                    it]. But in that post someone described how they used personas in
                    their company: As they made design choices they updated the persona
                    to reflect those design choices. For example: if the persona needed
                    to have certain knowledge - say about the difference between a csv
                    file and an excel spreadsheet, they'd modify the persona to indicate
                    the person had that understanding. They'd keep refactoring the
                    persona in response to design changes until the persona started to
                    look impossible. Then they'd push the persona back into shape, and
                    with it the design of the software. At first read I thought that
                    sounded OK - since the persona they used described the sort of
                    concrete person it would take to use their software. But, as I
                    learned more about personas, it now seems a mis-use - or at least
                    some alternative use of the form. Basically, their refactored
                    persona isn't directly connected to the market they're trying to
                    build the software for.

                    Lastly, in search for the post I described above, I found this:
                    http://www.cooper.com/newsletters/2001_07/perfecting_your_personas.htm
                    It may be worth looking at too.

                    Thanks,

                    -Jeff
                  • David John Anderson
                    So here is a thought... Could agile personas be creative writing? i.e. the designer, or a cross-functional team, create personas from their experience and
                    Message 9 of 27 , Sep 25, 2004
                    • 0 Attachment
                      So here is a thought...

                      Could "agile personas" be creative writing? i.e. the designer, or a
                      cross-functional team, create personas from their experience and
                      knowledge of the market but in the first instance they are creative
                      writing - based on subjective opinion rather than objective data. It's
                      faster and cheaper to do, but requires validation later.

                      What do you think?

                      David

                      PS And BTW, I've tried this in the past with some success. using the
                      cross-functional team approach to devise the persona definitions.

                      --- In agile-usability@yahoogroups.com, "Jeff Patton" <jpatton@a...>
                      wrote:
                      >
                      > Here's what I understand about personas: They aren't creative
                      > writing. They're built out of concrete data gathered from interviews
                      > or other sources. The more data gathered, the more likely the
                      > persona is to represent a constituency the software will serve. When
                      > assembled every fact or quote in the persona was ripped out of an
                      > interview - nothing's fabricated. Although they may look like a
                      > single person, they're a composite of lots of people. they're a
                      > stereotype of a customer/user we think we'll be writing for.
                      >
                    • Miriam Walker
                      ... My worry about using creative writing is that we would descend into opinion based design - which we can do pretty easily anyway without personas. I see
                      Message 10 of 27 , Sep 26, 2004
                      • 0 Attachment
                        > From: David John Anderson
                        > Could "agile personas" be creative writing? i.e. the designer, or a
                        > cross-functional team, create personas from their experience and
                        > knowledge of the market but in the first instance they are creative
                        > writing - based on subjective opinion rather than objective data. It's
                        > faster and cheaper to do, but requires validation later.

                        My worry about using "creative writing" is that we would descend into
                        opinion based design - which we can do pretty easily anyway without
                        personas. I see personas as a useful tool for grounding a team who
                        thinks of themselves as users too.
                        Miriam

                        Miriam Walker
                        Engineer-Scientist, Strategy & Research
                      • Adam Carter
                        ... I think to some degree all personas are creative writing, or at least they are creatively created based on a collection of user (data) that you have
                        Message 11 of 27 , Sep 26, 2004
                        • 0 Attachment
                          David John Anderson wrote:
                          > So here is a thought...
                          >
                          > Could "agile personas" be creative writing? i.e. the designer, or a
                          > cross-functional team, create personas from their experience and
                          > knowledge of the market but in the first instance they are creative
                          > writing - based on subjective opinion rather than objective data. It's
                          > faster and cheaper to do, but requires validation later.
                          >
                          > What do you think?
                          >
                          > David
                          >
                          > PS And BTW, I've tried this in the past with some success. using the
                          > cross-functional team approach to devise the persona definitions.

                          I think to some degree all personas are creative writing, or at least
                          they are creatively created based on a collection of user (data) that
                          you have recently studied.

                          However, I have this niggling question in the back of my mind when it
                          comes to creating personas based off of what one person thinks that the
                          genera user may be. How does the designer know that the users that have
                          been fictionally created are that simular to the users that they are
                          creating for right now?

                          If Civil Engineers just trusted that the materials they where going to
                          use in a building was safe because they used the same materials in a
                          previous (and simular) building … would you feel safe?

                          Maybe I am just seeing this as “make up the data today, and change it
                          later (if we can)”. O_o

                          Adam
                        • Joshua Seiden
                          ... writing, or at least ... user (data) that ... my mind when it ... person ... the users ... users that they are ... they where ... materials in a ... safe?
                          Message 12 of 27 , Sep 27, 2004
                          • 0 Attachment
                            >
                            > I think to some degree all personas are creative
                            writing, or at least
                            > they are creatively created based on a collection of
                            user (data) that
                            > you have recently studied.
                            >
                            > However, I have this niggling question in the back of
                            my mind when it
                            > comes to creating personas based off of what one
                            person
                            > thinks that the
                            > genera user may be. How does the designer know that
                            the users
                            > that have
                            > been fictionally created are that simular to the
                            users that they are
                            > creating for right now?
                            >
                            > If Civil Engineers just trusted that the materials
                            they where
                            > going to
                            > use in a building was safe because they used the same
                            materials in a
                            > previous (and simular) building ... would you feel
                            safe?
                            >

                            I don't think that this is an apt analogy. Personas
                            model the context of use, not the construction
                            material. To feed this distinction back into your
                            analogy, it would be more apt to imagine civil
                            engineers modeling the Golden Gate by saying, "it's
                            windy and foggy," and thus determining that a bridge
                            build there had better be strong and impervious to
                            rust.

                            It's accurate, and better than nothing, but clearly not
                            sufficient.

                            Your niggling question may then really be one of
                            tolerance. Having a model of the context of use is
                            better than not having one. That's (nearly) a given. So
                            the question becomes, "How accurate a model does one
                            need?" That's a business decision, a safety decision,
                            and ethical decision.

                            It also implies a question about how much we can rely
                            on models--especially qualitative models like personas.

                            Personas based on good research are certainly more
                            accurate than personas based on a bunch of folks
                            sitting around talking. But how accurate can they be?
                            How useful for prediction? In which spheres? I love the
                            persona method, but I recognize it has limits. While
                            personas could be useful to teams thinking about
                            product safety issues, for example, I would want to
                            rely exclusively on this type of qualitative model to
                            make such calls. I'd want to use other models, like
                            crash test dummies. [Insert your own customer jokes
                            here.]

                            JS
                          • Joshua Seiden
                            ... qualitative ... Should read: I would *not* want to rely exclusively on this type of qualitative model to make such calls.
                            Message 13 of 27 , Sep 27, 2004
                            • 0 Attachment
                              Correction:

                              > I
                              > would want to rely exclusively on this type of
                              qualitative
                              > model to make such calls.

                              Should read:

                              I would *not* want to rely exclusively on this type of
                              qualitative model to make such calls.
                            • Jeff Patton
                              ... It s ... I d use that as a technique - but I d want those using it to be pretty clear of the short term tradeoffs they were making. As Josh said in his
                              Message 14 of 27 , Sep 27, 2004
                              • 0 Attachment
                                --- In agile-usability@yahoogroups.com, "David John Anderson"
                                <netherby_uk@y...> wrote:
                                > So here is a thought...
                                >
                                > Could "agile personas" be creative writing? i.e. the designer, or a
                                > cross-functional team, create personas from their experience and
                                > knowledge of the market but in the first instance they are creative
                                > writing - based on subjective opinion rather than objective data.
                                It's
                                > faster and cheaper to do, but requires validation later.
                                >
                                > What do you think?

                                I'd use that as a technique - but I'd want those using it to be
                                pretty clear of the short term tradeoffs they were making. As Josh
                                said in his post, any model of the person using the product and the
                                context of use is better than no model. The exercise of writing a
                                persona gets people's heads in the right space. My only worry is
                                that we might start to fall in love with our own creations - to
                                defend our creative writing decisions. Or worse yet, to believe the
                                software we've designed and written is good because it will work for
                                the persona we've fabricated. Lots of opporunity for self-delusion -
                                but I guess that's what software design always resorts to anyway.
                                ;-)

                                As you said, we'd need to take steps to validate that persona later.
                                We'd need to do this in some subjective way. Instead of finding
                                examples of people that match our persona because we want proof it's
                                right, we need to look at all the folks we expect our software to
                                serve, and harvest personas from that data. Then reconcile the
                                personas created using these two approaches. This smacks of doing
                                things twice - but so what? If we've written code designed to
                                change, and if doing the agile persona allowed us to get started
                                sooner, and we're open to changing the software in response to that
                                change, I think it's a good strategy.

                                In responding to this and thinking about it, the takeaway for me is
                                that we're ending up with a similar workproduct via two different
                                approaches - one more rigorous than the other. There might be some
                                useful parallels in agile development. An agile persona might be
                                like a "spike" in XP terms. A bit of code written to learn about
                                some technology or confirm an architectural approach might work. The
                                spike allows us to betters do technical design and estimation, to
                                know if a development approach is feasible. At the end of the day
                                the spiked code should be thrown out and replaced with code written
                                with more rigorous attention to unit tests and needs required in the
                                current development work being done. If I'm using Larry
                                Constantine's term correctly, it's a consumable, not a deliverable.

                                Just thinking out loud.

                                Thanks,

                                -Jeff

                                >
                                > David
                                >
                                > PS And BTW, I've tried this in the past with some success. using the
                                > cross-functional team approach to devise the persona definitions.
                                >
                                > --- In agile-usability@yahoogroups.com, "Jeff Patton" <jpatton@a...>
                                > wrote:
                                > >
                                > > Here's what I understand about personas: They aren't creative
                                > > writing. They're built out of concrete data gathered from
                                interviews
                                > > or other sources. The more data gathered, the more likely the
                                > > persona is to represent a constituency the software will serve.
                                When
                                > > assembled every fact or quote in the persona was ripped out of an
                                > > interview - nothing's fabricated. Although they may look like a
                                > > single person, they're a composite of lots of people. they're a
                                > > stereotype of a customer/user we think we'll be writing for.
                                > >
                              • Dave Cronin
                                I think the second you start calling personas creative writing is the second they will have zero effect as a tool to help make product definition and design
                                Message 15 of 27 , Sep 27, 2004
                                • 0 Attachment
                                  I think the second you start calling personas "creative writing" is the
                                  second they will have zero effect as a tool to help make product
                                  definition and design DECISIONS.

                                  While they will help you maintain some focus as you brainstorm or knock
                                  around ideas, one of the biggest values of personas are their ability to
                                  help a group of people with diverse agendas make hard decisions about
                                  what a product ought to do. Without a basis in research, I don't believe
                                  you'll be able to get anyone to stick to the decisions. (And while I
                                  understand you agilers are comfortable with changing direction
                                  midstream, revisiting every decision is a great recipe for feature
                                  creep, thrash and a never-ending revision cycle.)

                                  Also, I totally agree with Josh's point about the limited scope of
                                  personas. They are fantastic for focusing design activities and making
                                  larger scale decisions. When you get down to nuts and bolts, it is often
                                  useful to augment them with real-life use cases, usability testing,
                                  customer feedback and/or quantitative studies.

                                  -dave
                                • David John Anderson
                                  Joshua, I think your getting at the essence of the agility question - when is an agile method appropriate and when not? Clearly there are systems where its
                                  Message 16 of 27 , Sep 27, 2004
                                  • 0 Attachment
                                    Joshua,

                                    I think your getting at the essence of the agility question - when
                                    is an agile method appropriate and when not?

                                    Clearly there are systems where its critical to be thorough and
                                    detailed and have an audit trail of reviews. Civil engineering is a
                                    classic example. [I know the US model is slightly different but...]
                                    The UK model for civil engineering requires three companies - the
                                    design engineers, the build engineers and the consulting engineers.
                                    The first designs, the 2nd and 3rd review the design. The 2nd builds
                                    the design and the 3rd reviews the work done by the 2nd. These ver
                                    un-agile mechanism were put in place precisely because safety is
                                    essential.

                                    So, if there were a defined set of domains for when a typical agile
                                    software development method - with its tacit knowledge and lack of
                                    gates, reviews and signoffs, and its culture of trust and
                                    empowerment - were appropriate would that also be a world where
                                    creative writing personas (with subsequent validation through
                                    testing) were acceptable? Or are we still in place where personas
                                    must be based on accurate extensive market research data?

                                    David



                                    --- In agile-usability@yahoogroups.com, "Joshua Seiden"
                                    <joshseiden@y...> wrote:

                                    >
                                    > Your niggling question may then really be one of
                                    > tolerance. Having a model of the context of use is
                                    > better than not having one. That's (nearly) a given. So
                                    > the question becomes, "How accurate a model does one
                                    > need?" That's a business decision, a safety decision,
                                    > and ethical decision.
                                    >
                                    > It also implies a question about how much we can rely
                                    > on models--especially qualitative models like personas.
                                    >
                                  • Joshua Seiden
                                    ... typical agile ... knowledge and lack of ... and ... world where ... through ... where personas ... data? ... Well, a couple of points. First, Dave Cronin
                                    Message 17 of 27 , Sep 27, 2004
                                    • 0 Attachment
                                      >
                                      > So, if there were a defined set of domains for when a
                                      typical agile
                                      > software development method - with its tacit
                                      knowledge and lack of
                                      > gates, reviews and signoffs, and its culture of trust
                                      and
                                      > empowerment - were appropriate would that also be a
                                      world where
                                      > creative writing personas (with subsequent validation
                                      through
                                      > testing) were acceptable? Or are we still in place
                                      where personas
                                      > must be based on accurate extensive market research
                                      data?
                                      >

                                      Well, a couple of points. First, Dave Cronin alluded to
                                      the political nature of design decision-making. In my
                                      experience, personas are political tools nearly as
                                      often as they are design tools. In this function,
                                      "creative writing personas" are next to useless.

                                      Your question includes the pleasant notion of a culture
                                      of trust, though. With such a culture, the political
                                      need is diminished, and creative-writing or no-research
                                      personas are more useful. The tolerance-of-error issue
                                      remains, however.

                                      Second, you refer (as have some others in this thread,
                                      I believe) to the idea of going back and validating
                                      personas through testing. I don't think that you can do
                                      that, and I'm not sure you would want to. Where I have
                                      used "creative writing personas" in the past (the
                                      phrase we preferred was "provisional personas") what
                                      needed to be validated was the design, not the
                                      personas. (I like Jeff's reference to the notion of a
                                      consumable. I'd not heard that before--it seems an apt
                                      description of a process tool like a persona.)

                                      The risk of provisional personas is that they start to
                                      take on a life of their own. When you go back and start
                                      doing personas for real, the provisional personas hang
                                      around like the neighbors who just won't leave. It has
                                      a complicating effect on the process, unless you can
                                      swap in a new team with fresh ideas.

                                      Which brings me to my question: why in the world would
                                      you *want* to do "agile personas?" For years, software
                                      people have been maligning the method as too
                                      lightweight to be of any use. Why suddenly is it so
                                      burdensome? Does an agile method mean we dispense with
                                      *all* up-front, research, modeling, and planning?

                                      JS
                                    • Adam Carter
                                      ... I don’t think it is that far from the truth of the matter. If you base the creation of an artefact on fiction, then you are running risks. Looking at the
                                      Message 18 of 27 , Sep 27, 2004
                                      • 0 Attachment
                                        Joshua Seiden wrote:

                                        > I don't think that this is an apt analogy. Personas
                                        > model the context of use, not the construction
                                        > material. To feed this distinction back into your
                                        > analogy, it would be more apt to imagine civil
                                        > engineers modeling the Golden Gate by saying, "it's
                                        > windy and foggy," and thus determining that a bridge
                                        > build there had better be strong and impervious to
                                        > rust.

                                        I don’t think it is that far from the truth of the matter. If you base
                                        the creation of an artefact on fiction, then you are running risks.

                                        Looking at the harbour and seeing that it is foggy and windy might make
                                        you think "it has to be strong and resistant to rust" but you couldn’t
                                        then just go out and grab materials based off of those assumptions.
                                        While these assumptions probably direct the focus of your initial
                                        research, but they are not the end of it (which was my point).

                                        It doesn’t matter what you are building, or what you are building with.


                                        Adam
                                      • Joshua Seiden
                                        Adam, I don t disagree with your point at all. I was just trying to be more precise with the analogy, that s all. I like the example, in fact. (So perhaps I
                                        Message 19 of 27 , Sep 27, 2004
                                        • 0 Attachment
                                          Adam,

                                          I don't disagree with your point at all. I was just
                                          trying to be more precise with the analogy, that's all.
                                          I like the example, in fact. (So perhaps I shouldn't
                                          have called the analogy inapt. My apologies for any
                                          offense given.)

                                          People were asking very specific questions about the
                                          persona technique. I was trying to frame the
                                          discussion. The point that I was trying to make is that
                                          personas are models, and we should examine the notion
                                          of modeling if we want to understand the limits of the
                                          technique.

                                          JS


                                          > -----Original Message-----
                                          > From: Adam Carter [mailto:a.carter@...]
                                          > Sent: Monday, September 27, 2004 8:20 PM
                                          > To: agile-usability@yahoogroups.com
                                          > Subject: Re: [agile-usability] Agile Personas was:
                                          Customer
                                          > isn't a UI Designer
                                          >
                                          >
                                          > Joshua Seiden wrote:
                                          >
                                          > > I don't think that this is an apt analogy. Personas
                                          > > model the context of use, not the construction
                                          > > material. To feed this distinction back into your
                                          > > analogy, it would be more apt to imagine civil
                                          > > engineers modeling the Golden Gate by saying, "it's
                                          > > windy and foggy," and thus determining that a
                                          bridge
                                          > > build there had better be strong and impervious to
                                          > > rust.
                                          >
                                          > I don't think it is that far from the truth of the
                                          matter. If
                                          > you base
                                          > the creation of an artefact on fiction, then you are
                                          running risks.
                                          >
                                          > Looking at the harbour and seeing that it is foggy
                                          and windy
                                          > might make
                                          > you think "it has to be strong and resistant to rust"
                                          but you
                                          > couldn't
                                          > then just go out and grab materials based off of
                                          those assumptions.
                                          > While these assumptions probably direct the focus of
                                          your initial
                                          > research, but they are not the end of it (which was
                                          my point).
                                          >
                                          > It doesn't matter what you are building, or what you
                                          are
                                          > building with.
                                          >
                                          >
                                          > Adam
                                          >
                                          >
                                          >
                                          > ------------------------ Yahoo! Groups Sponsor
                                          > --------------------~-->
                                          > $9.95 domain names from Yahoo!. Register anything.
                                          > http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA>
                                          /dpFolB/TM
                                          >
                                          >
                                          >
                                          -------------------------------------------------------
                                          -------
                                          > ------~->
                                          >
                                          >
                                          > Yahoo! Groups Links
                                          >
                                          >
                                          >
                                          >
                                          >
                                          >
                                        • Chris Pehura
                                          Because of my engineering background, I see models as a way to quickly understand behavior, interaction and contraints. They only need to be 70 to 80% right.
                                          Message 20 of 27 , Sep 28, 2004
                                          • 0 Attachment
                                            Because of my engineering background, I see models as a way to quickly understand behavior, interaction and contraints. They only need to be 70 to 80% right.
                                            What I don't like is the tendency for a quick one to one mapping from model to implementation. A model in the engineering sense gives you a quick summary of all the permutations of solutions. This is lost in that mapping. True software engineering follows very different rules than "physics based" engineering.
                                            Still, using a "model" and treating it as "design" really rubs me the wrong way.
                                            -----Original Message-----
                                            From: Joshua Seiden [mailto:joshseiden@...]
                                            Sent: Monday, September 27, 2004 9:58 PM
                                            To: agile-usability@yahoogroups.com
                                            Subject: RE: [agile-usability] Agile Personas was: Customer isn't a UI Designer

                                            Adam,

                                            I don't disagree with your point at all. I was just
                                            trying to be more precise with the analogy, that's all.
                                            I like the example, in fact. (So perhaps I shouldn't
                                            have called the analogy inapt. My apologies for any
                                            offense given.)

                                            People were asking very specific questions about the
                                            persona technique. I was trying to frame the
                                            discussion. The point that I was trying to make is that
                                            personas are models, and we should examine the notion
                                            of modeling if we want to understand the limits of the
                                            technique.

                                            JS


                                            > -----Original Message-----
                                            > From: Adam Carter [mailto:a.carter@...]
                                            > Sent: Monday, September 27, 2004 8:20 PM
                                            > To: agile-usability@yahoogroups.com
                                            > Subject: Re: [agile-usability] Agile Personas was:
                                            Customer
                                            > isn't a UI Designer
                                            >
                                            >
                                            > Joshua Seiden wrote:
                                            >
                                            > > I don't think that this is an apt analogy. Personas
                                            > > model the context of use, not the construction
                                            > > material. To feed this distinction back into your
                                            > > analogy, it would be more apt to imagine civil
                                            > > engineers modeling the Golden Gate by saying, "it's
                                            > > windy and foggy," and thus determining that a
                                            bridge
                                            > > build there had better be strong and impervious to
                                            > > rust.
                                            >
                                            > I don't think it is that far from the truth of the
                                            matter. If
                                            > you base
                                            > the creation of an artefact on fiction, then you are
                                            running risks.
                                            >
                                            > Looking at the harbour and seeing that it is foggy
                                            and windy
                                            > might make
                                            > you think "it has to be strong and resistant to rust"
                                            but you
                                            > couldn't
                                            > then just go out and grab materials based off of
                                            those assumptions.
                                            > While these assumptions probably direct the focus of
                                            your initial
                                            > research, but they are not the end of it (which was
                                            my point).
                                            >
                                            > It doesn't matter what you are building, or what you
                                            are
                                            > building with.
                                            >
                                            >
                                            > Adam
                                            >
                                            >
                                            >
                                            > ------------------------ Yahoo! Groups Sponsor
                                            > --------------------~-->
                                            > $9.95 domain names from Yahoo!. Register anything.
                                            > http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA>
                                            /dpFolB/TM
                                            >
                                            >
                                            >
                                            -------------------------------------------------------
                                            -------
                                            > ------~->
                                            >

                                            > Yahoo! Groups Links
                                            >
                                            >
                                            >

                                            >
                                            >



                                          • David J Anderson
                                            I hadn t heard that software people had been maligning the method as too lightweight. Could it be that these software people had a vested interest in another
                                            Message 21 of 27 , Sep 28, 2004
                                            • 0 Attachment
                                              I hadn't heard that "software people had been maligning the method
                                              as too lightweight." Could it be that these software people had a
                                              vested interest in another method - even if that vested interest was
                                              just training and tradition?

                                              Let me voice my concern, to aid understanding...

                                              I worry that many organizations that have not yet fully adopted an
                                              agile approach even though they could use one, do not have the
                                              skillsets to do personas properly i.e. they have no one on staff who
                                              could gather the field data. Nor do they have the political will to
                                              acquire those skills.

                                              So I am interested in understanding the value of "creative writing"
                                              personas. How close can they get you?

                                              I'm not sure that I buy the political objection when set in an agile
                                              world. In a dysfunctional big corporate world, I totally buy it.
                                              I've spent most of the last 10 years in IBM, Sprint, Motorola and
                                              now Microsoft. Believe me, I get it ;-) However, in an agile world,
                                              we work with cross-functional teams in joint session. Even within
                                              these beheamoth corporations, it is possible to build team trust and
                                              respect and achieve consensus on decisions even between cross-
                                              functional teams.

                                              I've seen people form a consensus around creatively written personas
                                              because they were all involved. And I've seen them accept that the
                                              persona technique is better than demographic abstractions.

                                              So, can we identify/quantify the quality improvement when a persona
                                              is a composite from objectively obtained data versus a creative
                                              composition from subjective data and opinion?

                                              Or are we splitting hairs, because either method must surely be
                                              better than Use Case Actors - the current "norm" for most software
                                              people. [BTW. Are the persona detractors normally practicing use
                                              case modeling?]

                                              David


                                              --- In agile-usability@yahoogroups.com, "Joshua Seiden"
                                              > Which brings me to my question: why in the world would
                                              > you *want* to do "agile personas?" For years, software
                                              > people have been maligning the method as too
                                              > lightweight to be of any use. Why suddenly is it so
                                              > burdensome? Does an agile method mean we dispense with
                                              > *all* up-front, research, modeling, and planning?
                                            • Petteri Hiisilä
                                              ... Personas bring you multiple benefits. With creative writing you are able to get a few of them, but at the same you might get yourself into more trouble
                                              Message 22 of 27 , Sep 28, 2004
                                              • 0 Attachment
                                                > Could "agile personas" be creative writing? i.e. the designer, or a
                                                > cross-functional team, create personas from their experience and
                                                > knowledge of the market but in the first instance they are creative
                                                > writing - based on subjective opinion rather than objective data. It's
                                                > faster and cheaper to do, but requires validation later.

                                                Personas bring you multiple benefits. With creative writing you are able
                                                to get a few of them, but at the same you might get yourself into more
                                                trouble than you began with. Be careful!

                                                These are the common design problems, and how creatively written
                                                personas might work with them:

                                                First, self-referential design. PROS: Creative writing enables you to
                                                create a clear, single target to design. You can measure, how
                                                effectively your design solves the persona's problems. Those problems
                                                are likely to be different than yours. CONS: This might be the wrong
                                                target! It's not smart to snipe down the wrong person, even if you hit
                                                right between the eyes!

                                                Second, the elastic user. PROS: Your target doesn't move anymore. No
                                                need to generalize the real needs. You can go straight to the point.
                                                Scenarios ("use cases", "stories") can be precise. You get a realistic
                                                look at the persona's skill levels. Less feature debates. No guesswork.
                                                CONS: You might be blind to some essential behavior and motives of the
                                                real users. You just didn't guess right. Some important funcionality
                                                doesn't find its way into the design.

                                                Third, edge cases. PROS: With personas you can tackle dozens of "what
                                                if" questions in no time. It's quite obvious, whether your persona will
                                                or will not need certain features. It's possible to see, what features
                                                are very important, and which are not too important. What you should
                                                implement at once, and what whan wait. What must be behind a big button
                                                and what should be hidden inside menus. CONS: You cannot trust your
                                                created personas to make these judgements. And even if you trust your
                                                writing, other people might not. You're still likely to end up having
                                                long edge-case discussions and you'll still make wrong decisions.

                                                The bottom line: while creative writing gives you a clear design target,
                                                others may not believe you that this is the right target, and it indeed
                                                can be the wrong target!

                                                The less research you do, the more risk of failure you have. Personas
                                                can shoot back at you!

                                                Success is not based on luck, except by accident.

                                                If you're going to do some creative writing, I suggest that you at least
                                                read Alan Cooper's The Inmates Are Running the Asylum (1999) before you
                                                begin. That's for your own sake.

                                                http://www.amazon.com/exec/obidos/tg/detail/-/0672316498/qid=1096386102/sr=1-2/ref=sr_1_2/102-9625995-1285746?v=glance&s=books

                                                I hope this helps.

                                                Best,
                                                Petteri

                                                --
                                                Petteri Hiisilä
                                                Palveluarkkitehti / Interaction Designer /
                                                Alma Media Interactive Oy / NWS /
                                                +358505050123 / petteri.hiisila@...

                                                "I was told there's a miracle for each day that I try"
                                                - John Petrucci
                                              • Joshua Seiden
                                                Chris, In my post, I wrote Personas model the context of use, not the construction material. I could have also written that they do not model the solution.
                                                Message 23 of 27 , Sep 28, 2004
                                                • 0 Attachment
                                                  Message
                                                  Chris,
                                                   
                                                  In my post, I wrote "Personas model the context of use, not the construction material. " I could have also written that they do not model the solution. Interaction designers use many means to model solutions, but never personas.
                                                   
                                                  Just to be clear, the phrase "context of use" has a very specific meaning to the user-centered-design community. it means the situation in which a solution will live, and includes (among other things, and depending on who you ask) the user, the environment, and the problems to be overcome.
                                                   
                                                  I regard "context of use" as roughly analogous to the problem statement, and design as analogous to the solution statement. And while I do believe that you can model both sides of that coin, I certainly would not assert that model=design.
                                                   
                                                  JS
                                                   
                                                  -----Original Message-----
                                                  From: Chris Pehura [mailto:chris@...]
                                                  Sent: Tuesday, September 28, 2004 9:59 AM
                                                  To: agile-usability@yahoogroups.com
                                                  Subject: RE: [agile-usability] Agile Personas was: Customer isn't a UI Designer

                                                  Because of my engineering background, I see models as a way to quickly understand behavior, interaction and contraints. They only need to be 70 to 80% right.
                                                  What I don't like is the tendency for a quick one to one mapping from model to implementation. A model in the engineering sense gives you a quick summary of all the permutations of solutions. This is lost in that mapping. True software engineering follows very different rules than "physics based" engineering.
                                                  Still, using a "model" and treating it as "design" really rubs me the wrong way.
                                                  -----Original Message-----
                                                  From: Joshua Seiden [mailto:joshseiden@...]
                                                  Sent: Monday, September 27, 2004 9:58 PM
                                                  To: agile-usability@yahoogroups.com
                                                  Subject: RE: [agile-usability] Agile Personas was: Customer isn't a UI Designer

                                                  Adam,

                                                  I don't disagree with your point at all. I was just
                                                  trying to be more precise with the analogy, that's all.
                                                  I like the example, in fact. (So perhaps I shouldn't
                                                  have called the analogy inapt. My apologies for any
                                                  offense given.)

                                                  People were asking very specific questions about the
                                                  persona technique. I was trying to frame the
                                                  discussion. The point that I was trying to make is that
                                                  personas are models, and we should examine the notion
                                                  of modeling if we want to understand the limits of the
                                                  technique.

                                                  JS


                                                  > -----Original Message-----
                                                  > From: Adam Carter [mailto:a.carter@...]
                                                  > Sent: Monday, September 27, 2004 8:20 PM
                                                  > To: agile-usability@yahoogroups.com
                                                  > Subject: Re: [agile-usability] Agile Personas was:
                                                  Customer
                                                  > isn't a UI Designer
                                                  >
                                                  >
                                                  > Joshua Seiden wrote:
                                                  >
                                                  > > I don't think that this is an apt analogy. Personas
                                                  > > model the context of use, not the construction
                                                  > > material. To feed this distinction back into your
                                                  > > analogy, it would be more apt to imagine civil
                                                  > > engineers modeling the Golden Gate by saying, "it's
                                                  > > windy and foggy," and thus determining that a
                                                  bridge
                                                  > > build there had better be strong and impervious to
                                                  > > rust.
                                                  >
                                                  > I don't think it is that far from the truth of the
                                                  matter. If
                                                  > you base
                                                  > the creation of an artefact on fiction, then you are
                                                  running risks.
                                                  >
                                                  > Looking at the harbour and seeing that it is foggy
                                                  and windy
                                                  > might make
                                                  > you think "it has to be strong and resistant to rust"
                                                  but you
                                                  > couldn't
                                                  > then just go out and grab materials based off of
                                                  those assumptions.
                                                  > While these assumptions probably direct the focus of
                                                  your initial
                                                  > research, but they are not the end of it (which was
                                                  my point).
                                                  >
                                                  > It doesn't matter what you are building, or what you
                                                  are
                                                  > building with.
                                                  >
                                                  >
                                                  > Adam
                                                  >
                                                  >
                                                  >
                                                  > ------------------------ Yahoo! Groups Sponsor
                                                  > --------------------~-->
                                                  > $9.95 domain names from Yahoo!. Register anything.
                                                  > http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA>
                                                  /dpFolB/TM
                                                  >
                                                  >
                                                  >
                                                  -------------------------------------------------------
                                                  -------
                                                  > ------~->
                                                  >

                                                  > Yahoo! Groups Links
                                                  >
                                                  >
                                                  >

                                                  >
                                                  >




                                                • Joshua Seiden
                                                  ... fully adopted an ... have the ... one on staff who ... political will to ... OK. I understand your concern. There are probably two distinct skill sets
                                                  Message 24 of 27 , Sep 28, 2004
                                                  • 0 Attachment
                                                    > Let me voice my concern, to aid understanding...
                                                    >
                                                    > I worry that many organizations that have not yet
                                                    fully adopted an
                                                    > agile approach even though they could use one, do not
                                                    have the
                                                    > skillsets to do personas properly i.e. they have no
                                                    one on staff who
                                                    > could gather the field data. Nor do they have the
                                                    political will to
                                                    > acquire those skills.
                                                    >

                                                    OK. I understand your concern.

                                                    There are probably two distinct skill sets here: the
                                                    research skills and the modeling skills. I find that
                                                    the modeling skills are more arcane. In my experience,
                                                    it's easy to find people who conduct field research
                                                    well; it's more difficult to find people who know how
                                                    to create personas that are useful and actionable for
                                                    interaction design.

                                                    "Creative writing" personas, as we're defining them
                                                    here are personas that are modeled, but not researched,
                                                    right? Or at least the research comes from "tacit
                                                    knowledge?"

                                                    If I were running this team, what I would do is bring
                                                    in a person skilled in persona modeling to facilitate a
                                                    one/two-day session to create personas. This person
                                                    would bring the modeling skills to the table, would
                                                    extract knowledge from the team, would alert the team
                                                    to holes and risks, would identify opportunities for
                                                    high value future research, and would allow the use of
                                                    the technique without requiring a research expenditure.

                                                    This would minimize the research spend, optimize the
                                                    modeling spend, and unlock the value of the existing
                                                    team. (What an elevator pitch!)

                                                    > So I am interested in understanding the value of
                                                    "creative writing"
                                                    > personas. How close can they get you?


                                                    Difficult to answer, except on a case by case basis.
                                                    Good personas give a team confidence in their
                                                    decisions, and the ability to move forward and build on
                                                    those decisions. Bad personas make everything squishy,
                                                    slow, and subject to rework. I wish I could be more
                                                    precise than this.


                                                    > I've seen people form a consensus around creatively
                                                    written personas
                                                    > because they were all involved. And I've seen them
                                                    accept that the
                                                    > persona technique is better than demographic
                                                    abstractions.

                                                    Yes. Good culture helps. The second point has to do
                                                    with the power of narrative, I think.

                                                    > So, can we identify/quantify the quality improvement
                                                    when a persona
                                                    > is a composite from objectively obtained data versus
                                                    a creative
                                                    > composition from subjective data and opinion?
                                                    >

                                                    I know intuitively from my years as a design consultant
                                                    that good research is priceless. I would like to be
                                                    able to articulate this to folks who haven't
                                                    experienced this. It would be interesting to do a study
                                                    comparing satisfaction/quality measures for products
                                                    created using different types of research methods. Has
                                                    anyone done such a study?

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