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

Re: [hackers-il] Rate of failure of Israeli software projects?

Expand Messages
  • Omer Zak
    Good question. To have focused discussion, let s limit the scope of non-successful projects to those, which were one or more of the following: 1. Cancelled
    Message 1 of 13 , Aug 22, 2005
      Good question.
      To have focused discussion, let's limit the scope of non-successful
      projects to those, which were one or more of the following:

      1. Cancelled before having been completed, after having been worked on
      for at least 80% of the scheduled elapsed time AND after having spent at
      least 80% of their budget.

      2. Finished at schedule overrun of at least 100% AND budget overrun of
      at least 100%, whether actually used by their target users or not.

      3. Finished, deployed and then the users found that the developed
      project is far from meeting their needs, so they did not really use it.

      This is just an arbitrary definition, which I made up in order to have a
      focused discussion.
      --- Omer

      On Mon, 2005-08-22 at 22:54 +0300, guy keren wrote:
      > On Mon, 22 Aug 2005, Omer Zak wrote:
      >
      > > The other day I saw an item concerning "Agile Development" and it
      > > mentioned, among other things, that very high percentage of software
      > > development projects fail.
      >
      > to what definition of 'fail'?
      >
      > > This was in American environment, with the known politics, power games,
      > > and managementpractices prevalent there.
      > >
      > > I wonder whether things are any different here in Israel.And when
      > > Israeli projects fail, is there any common cause due to the
      > > idiosyncracies of Israeli culture?
      >
      > again - to what definition of 'fail'?
      --
      One does not make peace with enemies. One makes peace with former
      enemies.
      My own blog is at http://www.livejournal.com/users/tddpirate/

      My opinions, as expressed in this E-mail message, are mine alone.
      They do not represent the official policy of any organization with which
      I may be affiliated in any way.
      WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
    • guy keren
      ... to what definition of fail ? ... again - to what definition of fail ? -- guy For world domination - press 1, or dial 0, and please hold, for the
      Message 2 of 13 , Aug 22, 2005
        On Mon, 22 Aug 2005, Omer Zak wrote:

        > The other day I saw an item concerning "Agile Development" and it
        > mentioned, among other things, that very high percentage of software
        > development projects fail.

        to what definition of 'fail'?

        > This was in American environment, with the known politics, power games,
        > and managementpractices prevalent there.
        >
        > I wonder whether things are any different here in Israel.And when
        > Israeli projects fail, is there any common cause due to the
        > idiosyncracies of Israeli culture?

        again - to what definition of 'fail'?

        --
        guy

        "For world domination - press 1,
        or dial 0, and please hold, for the creator." -- nob o. dy
      • guy keren
        ... what about a startup that was shut down? does this fall under this category? ... i think you just described some of microsoft s more successfully-marketed
        Message 3 of 13 , Aug 22, 2005
          On Mon, 22 Aug 2005, Omer Zak wrote:

          > Good question.
          > To have focused discussion, let's limit the scope of non-successful
          > projects to those, which were one or more of the following:
          >
          > 1. Cancelled before having been completed, after having been worked on
          > for at least 80% of the scheduled elapsed time AND after having spent at
          > least 80% of their budget.

          what about a startup that was shut down? does this fall under this
          category?

          > 2. Finished at schedule overrun of at least 100% AND budget overrun of
          > at least 100%, whether actually used by their target users or not.

          i think you just described some of microsoft's more successfully-marketed
          operating systems. you'll have to revise this definition.

          you also described linux kernel 2.4 - it took much much longer to finish
          then initially assumed, and more work ;)

          thus, i don't think this is a good definition for a "project failure".

          > 3. Finished, deployed and then the users found that the developed
          > project is far from meeting their needs, so they did not really use it.

          so we're not talking just about failure from the software-point-of-view.
          we're talking now about failure from a merketing-point-of-view?

          --
          guy

          "For world domination - press 1,
          or dial 0, and please hold, for the creator." -- nob o. dy
        • amos@amos.mailshell.com
          Every serious project should have set goals. Failed should be not reached set goals . Reasoning - if it s just a project of let s start writing something
          Message 4 of 13 , Aug 22, 2005
            Every serious project should have set goals.

            "Failed" should be "not reached set goals".

            Reasoning - if it's just a project of "let's start writing something
            just to play with technology and see where it goes" without ANY
            set objects ("it will allow users to do A, B and C in first revision")
            then it's not in the class of projects discussed (even its "owner"
            can't define his own success or failure).

            If it's a "scratch my itch" project then it already has enough goals
            to start talking about success/failure.

            If it's a startup then it usually have VERY well defined goals
            (a business plan which had to be convincing enough for an
            investor to put his money on, and I'm not talking about Kablanim
            looking to "put some money in this java-java thingy")
            A startup may change its original business plan, but it
            still should have goals of income, revenues, badget, deliverables,
            (usually software or hardware doing a defined function), valuation
            etc.

            How does that sound?

            Cheers,

            --Amos
          • Omer Zak
            If to judge from the resulting insights, Guy Keren again asked good questions. The question which I would like to investigate is whether there are fundamental
            Message 5 of 13 , Aug 22, 2005
              If to judge from the resulting insights, Guy Keren again asked good
              questions.

              The question which I would like to investigate is whether there are
              fundamental traits of today's Israeli software development project
              managers, which cause failures. Accordingly, I am clarifying and
              revising working definitions in the following.

              On Tue, 2005-08-23 at 01:07 +0300, guy keren wrote:
              > On Mon, 22 Aug 2005, Omer Zak wrote:
              >
              > > Good question.
              > > To have focused discussion, let's limit the scope of non-successful
              > > projects to those, which were one or more of the following:
              > >
              > > 1. Cancelled before having been completed, after having been worked on
              > > for at least 80% of the scheduled elapsed time AND after having spent at
              > > least 80% of their budget.
              >
              > what about a startup that was shut down? does this fall under this
              > category?

              If a project was scheduled for completion in 10 months, and after 8.5
              months it was discovered that the market conditions changed in such a
              way to make the project obsolete, then this does not constitute a
              failure of the project manager but of the marketing department, which
              requested the project. The failure is in predicting the market shift.

              So a project carried out by a startup, which was shut down due to
              shifting market conditions would not be inside the investigation's
              scope.

              But a startup, which fails to complete development of its product in
              time belongs to the investigation. Failure can be due to various
              causes, which the project manager should have been able to foresee and
              plan accordingly.

              > > 2. Finished at schedule overrun of at least 100% AND budget overrun of
              > > at least 100%, whether actually used by their target users or not.
              >
              > i think you just described some of microsoft's more successfully-marketed
              > operating systems. you'll have to revise this definition.

              NO. If the projects of developing those operating systems were better
              managed, maybe the users' experience would not have been so bug-ridden.
              Besides, were those projects developed in...ISRAEL?

              > you also described linux kernel 2.4 - it took much much longer to finish
              > then initially assumed, and more work ;)
              >
              > thus, i don't think this is a good definition for a "project failure".

              Linus and his crew changed their development practices after learning
              their lessons from developing kernel 2.4. Now there are no alternating
              periods of stable and development kernels. Therefore I can argue that
              they themselves considered the kernel 2.4 project to be subjected to
              project mismanagement.

              > > 3. Finished, deployed and then the users found that the developed
              > > project is far from meeting their needs, so they did not really use it.
              >
              > so we're not talking just about failure from the software-point-of-view.
              > we're talking now about failure from a merketing-point-of-view?

              I do not know what point of view you are talking about, but I am talking
              about project management point of view. Come over, or else start your
              own thread and investigation (to which I would gladly contribute
              whatever I can contribute).

              The specific case of project, which was completed but not usable, can be
              considered to be a failure of the project manager, who did not have
              enough communication with the users and did not do good enough work to
              elicit their explicit, implicit, unstated and secret requirements; and
              to carry out the project in such way that it meets those requirements.
              --- Omer
              --
              MS-Windows is the Pal-Kal of the PC world.
              My own blog is at http://www.livejournal.com/users/tddpirate/

              My opinions, as expressed in this E-mail message, are mine alone.
              They do not represent the official policy of any organization with which
              I may be affiliated in any way.
              WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
            • amos@amos.mailshell.com
              ... Just because they changed development process doesn t mean that the old process falls under definition of mismanagement . I don t remember the timelines
              Message 6 of 13 , Aug 22, 2005
                On 8/23/05, Omer Zak <omerz.at.actcom.co.il@...> wrote:
                > > you also described linux kernel 2.4 - it took much much longer to finish
                > > then initially assumed, and more work ;)
                > >
                > > thus, i don't think this is a good definition for a "project failure".
                >
                > Linus and his crew changed their development practices after learning
                > their lessons from developing kernel 2.4. Now there are no alternating
                > periods of stable and development kernels. Therefore I can argue that
                > they themselves considered the kernel 2.4 project to be subjected to
                > project mismanagement.

                Just because they changed development process doesn't mean that the
                old process falls under definition of "mismanagement". I don't remember
                the timelines or the decisions made (and I'm sure I'm not aware of lots of
                these), but it could also be that the environment has changed since 2.4
                began development and when it was released (companies joined the
                revolution, people came and left, technologies shifted, market environment
                changed, lots of other things, especially with something so huge as the
                Linux project).

                > The specific case of project, which was completed but not usable, can be
                > considered to be a failure of the project manager, who did not have
                > enough communication with the users and did not do good enough work to
                > elicit their explicit, implicit, unstated and secret requirements; and
                > to carry out the project in such way that it meets those requirements.

                I agree with that.

                > --- Omer

                Cheers,

                --Amos
              • Omer Zak
                ... [... snipped ...] ... If a project was 5% over schedule and 5% over budget but achieved its set goals, then its overall management was not bad enough to
                Message 7 of 13 , Aug 23, 2005
                  On Tue, 2005-08-23 at 08:34 +1000, amos@... wrote:
                  > Every serious project should have set goals.
                  >
                  > "Failed" should be "not reached set goals".

                  [... snipped ...]

                  > How does that sound?

                  If a project was 5% over schedule and 5% over budget but achieved its
                  set goals, then its overall management was not bad enough to serve as an
                  example of bad project management (although there may have been some
                  problematic practices).

                  Also, a project can be well-managed but cancelled midway due to external
                  causes. Since the external causes are not project mismanagement per se,
                  we cannot learn from such a project. Usually, it can be ascribed to
                  marketing failure of predicting the future. But if it was one of 5
                  parallel projects initiated by a very rich and profitable corporation,
                  with the intention that one succeeds and four fail, yet the successful
                  one covers the expenses of all five projects - then the four
                  failing/cancelled projects are not necessarily mismanaged.

                  Now, that we have discussed criteria and clarified them a bit, can we
                  turn to the main subject - that of cultural influences on Israeli
                  project managers, which cause their projects to fail.

                  Are there such adverse cultural influences (the way there are in USA,
                  for example)? Are there any countries, in which project managers
                  consistently excel, thanks to the beneficial influence of their national
                  management culture?
                  --- Omer
                  --
                  MS-Windows is the Pal-Kal of the PC world.
                  My own blog is at http://www.livejournal.com/users/tddpirate/

                  My opinions, as expressed in this E-mail message, are mine alone.
                  They do not represent the official policy of any organization with which
                  I may be affiliated in any way.
                  WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
                • amos@amos.mailshell.com
                  ... You can quantify these 5% in the goals too: Over 5% - failed. Within 5% - goal reached . And so on and so forth for other resources. (As an anacdot, I ve
                  Message 8 of 13 , Aug 23, 2005
                    On 8/23/05, Omer Zak <omerz.at.actcom.co.il@...> wrote:
                    > On Tue, 2005-08-23 at 08:34 +1000, amos@... wrote:
                    > > Every serious project should have set goals.
                    > >
                    > > "Failed" should be "not reached set goals".
                    >
                    > [... snipped ...]
                    >
                    > > How does that sound?
                    >
                    > If a project was 5% over schedule and 5% over budget but achieved its
                    > set goals, then its overall management was not bad enough to serve as an
                    > example of bad project management (although there may have been some
                    > problematic practices).

                    You can quantify these 5% in the goals too: "Over 5% - failed. Within 5% - goal
                    reached". And so on and so forth for other resources.

                    (As an anacdot, I've just checked CRM solutions where a manager could set
                    "predicted time" and "predicted deviation").

                    > Now, that we have discussed criteria and clarified them a bit, can we
                    > turn to the main subject - that of cultural influences on Israeli
                    > project managers, which cause their projects to fail.

                    Where do you expect to find such data? I'd look it up in publications of
                    companies like Deloyt-Tusch(sp) and maybe the IDC, if it's available for
                    the public at all...

                    > Are there such adverse cultural influences (the way there are in USA,
                    > for example)? Are there any countries, in which project managers
                    > consistently excel, thanks to the beneficial influence of their national
                    > management culture?

                    I don't know. I can only tell from my personal experience - the most well
                    managed software development organization I've ever seen is completly
                    run by Israelis (plus a russian-born R&D manager). Don't know how to
                    extrapolate from this or my other experience on the rest of the industry,
                    some are better and some are worse in any country.

                    --Amos
                  • Omer Zak
                    ... Lacking access to such publications, and knowing that the publications do not capture informal anecdotes, I threw the question to the Hackers-IL mailing
                    Message 9 of 13 , Aug 23, 2005
                      On Tue, 2005-08-23 at 18:00 +1000, amos@... wrote:
                      > On 8/23/05, Omer Zak <omerz.at.actcom.co.il@...> wrote:
                      > > Now, that we have discussed criteria and clarified them a bit, can we
                      > > turn to the main subject - that of cultural influences on Israeli
                      > > project managers, which cause their projects to fail.
                      >
                      > Where do you expect to find such data? I'd look it up in publications of
                      > companies like Deloyt-Tusch(sp) and maybe the IDC, if it's available for
                      > the public at all...

                      Lacking access to such publications, and knowing that the publications
                      do not capture informal anecdotes, I threw the question to the
                      Hackers-IL mailing list to see whether some subscribers have anything to
                      contribute.

                      > > Are there such adverse cultural influences (the way there are in USA,
                      > > for example)? Are there any countries, in which project managers
                      > > consistently excel, thanks to the beneficial influence of their national
                      > > management culture?
                      >
                      > I don't know. I can only tell from my personal experience - the most well
                      > managed software development organization I've ever seen is completly
                      > run by Israelis (plus a russian-born R&D manager). Don't know how to
                      > extrapolate from this or my other experience on the rest of the industry,
                      > some are better and some are worse in any country.

                      What can you tell us about the average case - the average software
                      development organization in Israel (or at least among those which you
                      encountered)?

                      My own experience (which may be obsolete by now) is that I personally am
                      more aggressive than the average organization in developing or deploying
                      tools to assist them in the work.
                      (I remember an organization, for which I worked several years ago, in
                      which there was resistance to using 'make' and it was broken down only
                      after they added another employee in the role of 100% software
                      engineer.)
                      --- Omer
                      --
                      MS-Windows is the Pal-Kal of the PC world.
                      My own blog is at http://www.livejournal.com/users/tddpirate/

                      My opinions, as expressed in this E-mail message, are mine alone.
                      They do not represent the official policy of any organization with which
                      I may be affiliated in any way.
                      WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
                    • Ori Idan
                      Resistance to using make ? why because it was hard to use or because they did not realize the potential and the need for such a tool? -- Ori Idan
                      Message 10 of 13 , Aug 23, 2005
                        Resistance to using 'make'? why because it was hard to use or because
                        they did not realize the potential and the need for such a tool?

                        --

                        Ori Idan


                        Omer Zak wrote:

                        >On Tue, 2005-08-23 at 18:00 +1000, amos@... wrote:
                        >
                        >
                        >>On 8/23/05, Omer Zak <omerz.at.actcom.co.il@...> wrote:
                        >>
                        >>
                        >>>Now, that we have discussed criteria and clarified them a bit, can we
                        >>>turn to the main subject - that of cultural influences on Israeli
                        >>>project managers, which cause their projects to fail.
                        >>>
                        >>>
                        >>Where do you expect to find such data? I'd look it up in publications of
                        >>companies like Deloyt-Tusch(sp) and maybe the IDC, if it's available for
                        >>the public at all...
                        >>
                        >>
                        >
                        >Lacking access to such publications, and knowing that the publications
                        >do not capture informal anecdotes, I threw the question to the
                        >Hackers-IL mailing list to see whether some subscribers have anything to
                        >contribute.
                        >
                        >
                        >
                        >>>Are there such adverse cultural influences (the way there are in USA,
                        >>>for example)? Are there any countries, in which project managers
                        >>>consistently excel, thanks to the beneficial influence of their national
                        >>>management culture?
                        >>>
                        >>>
                        >>I don't know. I can only tell from my personal experience - the most well
                        >>managed software development organization I've ever seen is completly
                        >>run by Israelis (plus a russian-born R&D manager). Don't know how to
                        >>extrapolate from this or my other experience on the rest of the industry,
                        >>some are better and some are worse in any country.
                        >>
                        >>
                        >
                        >What can you tell us about the average case - the average software
                        >development organization in Israel (or at least among those which you
                        >encountered)?
                        >
                        >My own experience (which may be obsolete by now) is that I personally am
                        >more aggressive than the average organization in developing or deploying
                        >tools to assist them in the work.
                        >(I remember an organization, for which I worked several years ago, in
                        >which there was resistance to using 'make' and it was broken down only
                        >after they added another employee in the role of 100% software
                        >engineer.)
                        > --- Omer
                        >
                        >
                      • Omer Zak
                        ... They were used to typing in the commands to build and link the modules and were set in their ways. The software in question was simple, though (about 3-4
                        Message 11 of 13 , Aug 23, 2005
                          On Tue, 2005-08-23 at 13:11 +0300, Ori Idan wrote:
                          > Resistance to using 'make'? why because it was hard to use or because
                          > they did not realize the potential and the need for such a tool?

                          They were used to typing in the commands to build and link the modules
                          and were set in their ways.

                          The software in question was simple, though (about 3-4 modules per
                          build).
                          --- Omer
                          --
                          Delay is the deadliest form of denial. C. Northcote Parkinson
                          My own blog is at http://www.livejournal.com/users/tddpirate/

                          My opinions, as expressed in this E-mail message, are mine alone.
                          They do not represent the official policy of any organization with which
                          I may be affiliated in any way.
                          WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
                        • amos@amos.mailshell.com
                          ... At least the ones I took part in delivered their goals and as far as I can tell they have done so within the allocated resources (time, manpower, badget).
                          Message 12 of 13 , Aug 23, 2005
                            On 8/23/05, Omer Zak <omerz.at.actcom.co.il@...> wrote:
                            > What can you tell us about the average case - the average software
                            > development organization in Israel (or at least among those which you
                            > encountered)?

                            At least the ones I took part in delivered their goals and as far as I can
                            tell they have done so within the allocated resources (time, manpower,
                            badget). There were a couple of exceptions I am aware of:

                            1. The "Big Project" at Comverse to move the entire platform from an
                            ancient real-time 386 OS + PL/I + C etc to Windows NT + C++ failed,
                            possibly due to mis-management and unfamiliarity with the technology
                            and maybe also unfamiliarity with the development process they tried to
                            use.
                            (I haven't stayed there to watch the failure, I was just fed-up with
                            Windows NT and moved on to other projects which involved
                            UNIX programming).

                            2. During a one-year stint at Sun I was mostly responsible to help Amdocs
                            convert from HP to Solaris. Their way of working is to use lots of zombie-level
                            programmers who understand pretty much nothing at what they are doing
                            and just keep throwing more and more people at late projects. It works,
                            eventually, but at a much higher cost for the customer than he bargained for.
                            (who's apparently a captive by the time this happens).
                            You get to see wierd things in Amdocs (e.g. in parallel to the project I was
                            helping with, I slowly discovered three other projects to do the same (convert
                            their HP code to Solaris).
                            I don't know if this is considered "failure" since Amdocs generally succeed in
                            winning and retaining contracts but still their line of work is very distressing
                            just to be reminded about.

                            My refference: a few years of sub-contracting for various projects in Comverse,
                            and a few years of working for software startups. With side-kicks for the
                            army and others.

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