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

Re: [CMMi Process Improvement] Quality standards for delivered services

Expand Messages
  • Patrick OToole
    Jerry, Once an organization has completed a SCAMPI A appraisal, the results can be posted on the SEI website for up to 3 years. During that time, the
    Message 1 of 22 , Sep 30, 2008
    View Source
    • 0 Attachment
      Jerry,

      Once an organization has completed a SCAMPI A appraisal, the results can be posted on the SEI website for up to 3 years. During that time, the organization can certainly "claim" to still be operating at ML3 regardless of changes to methodology - or management, ownership, geographical location, etc.

      Whether they ARE still ML3 or not could only be determined by another appraisal. In fact, it's pretty easy to understand why the SEI introduced a 3-year time limit on the validity of appraisal results - an organization is likely to undergo some rather significant changes over that period of time and the likelihood of maintaining the stated level of process maturity is somewhat diminished.

      So, bottom line, an organization that is rated ML3 using XP and RUP can introduce SCRUM and still claim ML3 - at least until their 3-year time period expires.

      Regards,

      Pat



      ----- Original Message -----
      From: Jerry Zhu
      To: cmmi_process_improvement@yahoogroups.com
      Sent: Monday, September 29, 2008 8:58 PM
      Subject: Re: [CMMi Process Improvement] Quality standards for delivered services


      Pat,

      Still some questions.

      Methodology I meant development/management methodologies. Testing, peer review etc are practices in my oppinion. Testing could be SP 3.3 of PI "Evaluate Assembled Product Component" for example. There might be different testing practices but they are not methodologies.

      If a company adopted multiple methodologies, XP, RUP, SCRUM, FDD for example. They are very different methodologies with different work products and process descriptions. There will still be three focus projects to cover all four metodologies?

      From another standpoint, if a company is rated at ML3 with two methodologies XP and RUP. After that, can the company use SRCUM
      and claim ML3?

      Thanks

      Jerry

      --- On Wed, 9/10/08, Patrick OToole <PACT.otoole@...> wrote:

      Jerry,

      In theory, you are correct - if two different methodologies are used within an organization, one would want to put 6+ instances in scope for the appraisal - although admittedly that is not the way its always done in practice.

      First of all, it depends on what the methodology happens to be. For example, if we are talking about testing methodologies, or peer review methodologies, we would tend to cover the multiple methodologies with non-focus projects rather than focus projects.

      [Caution, Yogi Berra statement ahead...] Non-focus projects would allow us to focus solely on the unique methodologies better than focus projects would. That is, the evaluation of non-focus projects would concentrate solely on the way automated testing is planned and executed (as opposed to the manual testing methods used on the focus projects, thereby avoiding the overhead of adding 3 more instantiations of PP, PMC, TS, etc.

      Second, in order to constrain time and money, such an appraisal may only include 2 instances of each methodology - 2 focus projects that used methodology #1, and 1 focus + 1 non-focus project that employed methodology #2. Of course it would be more robust to include 3+3, but the risk vs. cost tradeoff will usually tilt toward favoring cost.

      Regards,

      Pat

      ----- Original Message -----
      From: Jerry Zhu
      To: cmmi_process_ improvement@ yahoogroups. com
      Sent: Tuesday, September 09, 2008 8:30 PM
      Subject: Re: [CMMi Process Improvement] Quality standards for delivered services

      Hi Pat,

      You made it cristal clear. Thank you so much. Most of the CMMI practices will be implemented three times by the three focus projects and the left over CMMI practices shall be picked up by non-focus projects. A fourth focus project could significantly reduce the # of non-focus projects needed. The non-focus projects should also include "deviate" practices in order to be representative as a whole.

      Now continuing Bruce's comments, I think that the three plus focus projects should be corresponding to a single methodology. If two methodologies are used in the organization and both are appraised, each will require 3+ focus projects to be representative for the methodology. That will make it 6+ focus projects. That is, three focus projects only can not be possibly reprsentative for the two methodologies.

      Really appreciated.

      Jerry

      --- On Tue, 9/9/08, Patrick OToole <PACT.otoole@ att.net> wrote:

      Jerry,

      [Note: I have not been following this thread as closely as some, so my comments may be a BIT out of context. However...]

      The minimum of three implementation instances for each practice (note - not necessarily from 3 focus projects) ensures that the appraisal team gains a more robust understanding of how that practice is performed across the organization, and provides greater confidence that the practice is implemented in all applicable projects. This allows the appraisal team to make more informed decisions with respect to Organization Unit practice characterizations.

      Remember that the 3+ focus projects are merely "representative" of the full organizational population, so inferences drawn by examining the activities in the focus project SHOULD be extendable to all (otherwise, they weren't very "representative" projects, were they?)

      Granted, there may be differences in the activities performed by other projects. If these are deemed consequential, then those activities should be put in scope for appraisal examination by including the projects that perform them as "non-focus" projects - in which case those projects would have a subset of their activities evaluated during the appraisal.

      Remember, too that three focus projects is the MINIMUM, the appraisal may require many more projects to get a good representative sample. I previously mentioned the continuous representation appraisal that I have coming up in November where we had to put 20 projects in scope for TS and VER - it's the only way we could get a robust view of how the 200+ projects do their work (across multiple platforms, application domains, modes (development vs. maintenance) , etc.).

      If a PA, say DAR, is not needed by any of the projects, then 3+ non-focus projects would be submitted to allow the appraisal team to evaluate those practices appropriately. Remember, too, that some process areas, like DAR, may also be performed at the organizational level, so those instances may be included as "non-focus projects" as well.

      I haven't been following the thread enough to comment on your PAL assertion, so I'll leave that to others....

      Regards,

      Pat

      ----- Original Message -----
      From: Jerry Zhu
      To: cmmi_process_ improvement@ yahoogroups. com
      Sent: Monday, September 08, 2008 8:05 PM
      Subject: RE: [CMMi Process Improvement] Quality standards for delivered services

      Hi Bruce,

      Very helpful. I was wondering why minimum three focus projects. It is because all CMMI PAs' goals and practices are covered by the three projects? In other words, the sum of all three projects' needs will need to cover all CMMI PAs? Which means all goals and their practices may be covered by three focus projects at least once? What about a PA is not needed by any of the three focus projects? The needs of projects maybe subjective.

      I think that good (normalized) PAL should have one to many mapping of goals/practices between PAL and project process, not the other way around.

      Thanks

      Jerry

      --- On Mon, 9/8/08, Bruce R. Duncil <brduncil@bellsouth . net> wrote:

      Hi Jerry,
      You stated below that your question to me (should EVERY method and
      process component in a PAL be used three times or at least once in
      doing projects to pass a appraisal?) didn't get answered. That is a
      different question! The answer to this question is 'no'.

      On 3 September, I provided the following in response to your earlier
      post: "CMMI appraisals (and ratings) should cover an organizational
      element that is fairly represented by all the processes it has in
      use. Excluding one or more methodologies that are used in the company
      from the appraisal should cause the organizational scope of the
      appraisal to be adjusted and well documented accordingly. "

      So, when an organization is appraised, its projects are appraised and
      their processes that they use are appraised. We don't look at the PAL
      to determine if every single thing in it is used.

      Hope this is helpful to you.
      Best Regards,
      Bruce
      www.alderonconsulti ng.com

      At 11:58 AM 9/5/2008, you wrote:

      >Jeff,
      >
      >Regarding mixed methods. I recommend to ready systems theory,
      >particularly Ackoff to really communicate and understand Bruce's
      >experience and his truthful explanation.
      >
      >Whatever are put in PAL and used in a project. The process/methods
      >belong to a single system. XP is a system different from RUP. Some
      >may say RUP and XP hybrid as merger of two systems. They are not, it
      >is a new system, called home made or priproiatory system as many do
      >to create their own system using altered parts of others.
      >
      >Assembling Buick's brake, BMW's transmission, Royce Roller's engine
      >together do not give you a best car, In fact it is not even a car
      >(parts do not match). YOu have altered the parts and fix them into
      >your system different from Buick, BMW, Royce Rller.
      >
      >So I agree with Bruce that in PAL, there is only one system used for
      >a project, it is either RUP, XP, or custom made. When custom made,
      >they often borrow and customize parts from other systems. My
      >question to Bruce and agreed with is that, it is rare to have two or
      >more INCOMMEASURABLE systems placed in the same PAL. It is possible
      >for a firm to have two or more sysstems. In economics terms placing
      >two in one PAL may cost more for unnecessary complicatedness of use
      >and maintenance. A truck is a system and convertable is another
      >system. Merging the two not only is expensive and useless. Use two
      >PALs for to separate systems maybe a good idea.
      >
      >My question to Bruce that I didn't get an answer is that should
      >EVERY method and process component in a PAL be used three times or
      >at least once in doing projects to pass a appraisal. Three focus
      >projects may cover about 50% of what is in PAL. This may relate to
      >the quality of PAL measured by normalization, precisness and conciseness.
      >
      >Lastly for the off topic issue. Is there light or heavy weight
      >methodology for constructing a bridge? or only one rightweight? A
      >thousand years ago there might have been two hundred ways of
      >building a bridge where scientific knowledge is primative. Software
      >industry we are now is as civil industry a thousand years ago. If
      >there is a righweight in softwre industry, most won't recognize it,
      >if even it were here now. So it is off the topic. back to busines as usual.
      >
      >best
      >
      >jerry
      >
      >--- On Fri, 9/5/08, Jeff Dalton - Broadsword
      ><<mailto:jeff% 40Broadswordsolu tions.com>jeff@ Broadswordsolu tions.com> wrote:
      >
      >Jerry,
      >
      >". In practice, does that ever happen?"
      >
      >Yes! Unlike Bruce's experience, I often see "mixing" of methods - and a
      >single, seamless PAL. I have one client , for example, that has their own
      >method, "xRUP," that utilizes concepts from RUP and XP together, and it
      >works for them. Being a purist often does not result in success.
      >
      >I don't understand your statement about "right-weight. " That seems like a
      >cop-out to me as it has zero definition or meaning (although it seems
      >impressive to me). What is that? The concepts of lightweight and
      >heavyweight (and everything in-between) are based upon the context of the
      >organization, their needs, constraints, and tailoring strategy. There is no
      >way to define a standard for "right."
      >
      >Warm Regards,
      >
      >Jeff
      >
      >Jeff Dalton, President and CEO
      >
      >Broadsword
      >
      >SCAMPI Lead Appraiser, CMMI Instructor, SCAMPI Team Leader Observer
      >
      >248.341.3367 (office)
      >
      >248.709.4775 (cell)
      >
      >248.341.3672 (fax)
      >
      >broadswordlogo. tif
      >
      >Visit our website at <http://www.broadswo>http://www.broadswo rdsolutions. com
      >
      >Visit Jeff's blog at: <http://www.asktheCM>http://www.asktheCM
      >MIAppraiser. com
      ><<http://asktheCMMIAp>http://asktheCMMIAp praiser.blogspot .com>
      >
      >From: cmmi_process_ improvement@ yahoogroups. com
      >[mailto:cmmi_ process_ improvement@ yahoogroups. com] On Behalf Of Jerry Zhu
      >Sent: Thursday, September 04, 2008 7:18 PM
      >To: cmmi_process_ improvement@ yahoogroups. com
      >Subject: Re: [CMMi Process Improvement] Quality standards for delivered
      >services
      >
      >Orhan,
      >
      >In non-theory, yes you can put anything in PAL. In practice, does that ever
      >happen? I think Bruce's rich exprience is common and representative. In
      >economic terms, placing two incommensurable methodologies together cost more
      >(both creating and using it) than creating two separate PAL. Well it depends
      >on how you organize it. It maybe between disjoint and cojoint sets to share
      >comment method elements.
      >
      >Both heavyweight and lightweight methodologies are unhealthy methodologies.
      >I'd use rightweight methodology (the methodology) . But it is off the topic
      >in this forum.
      >
      >Jerry
      >
      >--- On Thu, 9/4/08, Orhan Kalayci <orhan.kalayci@ gmail.com
      ><mailto:orhan. kalayci%40gmail. com> > wrote:
      >
      >From: Orhan Kalayci <orhan.kalayci@ gmail.com
      ><mailto:orhan. kalayci%40gmail. com> >
      >Subject: Re: [CMMi Process Improvement] Quality standards for delivered
      >services
      >To: "cmmi_process_ improvement@ yahoogroups. com
      ><mailto:cmmi_ process_improvem ent%40yahoogroup s.com> "
      ><cmmi_process_ improvement@ yahoogroups. com
      ><mailto:cmmi_ process_improvem ent%40yahoogroup s.com> >
      >Date: Thursday, September 4, 2008, 2:47 PM
      >
      >Hi jerry,
      >
      >It is the art of process architecture to develop a process asset
      >library that support both conventional approaches such as waterfal and
      >those of agile approaches at the same time. It is the art of
      >developing tailoring guidelines not allow creating unhealthy defined
      >process for a project.
      >
      >In short it it is possible to have waterfal and agile lifecycles in
      >the same PAL. It is not only possible but healhty as well.
      >
      >Peace,
      >Orhan
      >Toronto
      >
      >Sent from my iPhone
      >
      >On Sep 2, 2008, at 21:15, Jerry Zhu <jerryyz@yahoo. com> wrote:
      >
      > > Hi Bruce,
      > >
      > > I am not sure I was clear myself.
      > >
      > > I meant if there were two disjoint sets of activities and work
      > > products. Each alone can compose a single project. The two sets may
      > > overlap.
      > >
      > > For example, two methodologies waterfall and XP are both stored in
      > > the same process asset library. In a project, half waterfall and
      > > half XP actitivies and work products are used. Certainly the project
      > > failed outright but PPQA passed.
      > >
      > > My question really is whether a CMMI rating corresponds to a single
      > > methodology? If yes, then the company is rated on SCRUM or RUP or
      > > propriatory one. This seems to me that the company gets stuck with
      > > one methodology?
      > >
      > > Regards
      > >
      > > Jerry
      > >
      > > --- On Wed, 9/3/08, Bruce R. Duncil <brduncil@bellsouth . net> wrote:
      > >
      > > Hi Jerry,
      > > It requires multiple activities to create the many work products that
      > > compose a single project. Yes, there will typically be many
      > > procedures and product specifications/ descriptions in a robust,
      > > well-defined set of processes. QA selects procedures and
      > > specifications/ descriptions against which to evaluate compliance
      > > (PPQA SP 1.1) and conformance (PPQA SP 1.2).
      > >
      > > Best Regards,
      > > Bruce
      > >
      > > At 10:12 AM 9/2/2008, you wrote:
      > >
      > > >Hi Bruce,
      > > >
      > > >I have a question about "activities" and "products" mentioned in a)
      > > >and b) below. Is {activities, products} and project one-one
      > > >relationship? If not, there will be different sets of {activities,
      > > >products}to deliver the same project. In this case, there would be
      > > >multiple procedures and product specs and descriptions for the same
      > > >practice in a project.
      > > >
      > > >Thanks
      > > >
      > > >Jerry
      > > >
      > > >--- On Mon, 9/1/08, Bruce R. Duncil
      > > ><<mailto:brduncil% 40bellsouth. net>brduncil@ bellsouth. net> wrote:
      > > >
      > > >Hi,
      > > >PPQA evaluations are intended to ensure that a) activities are
      > > >performed in a manner that is compliant with policies, processes, and
      > > >procedures and b) products conform to their product specifications,
      > > >standards, and descriptions. You have to select which activities
      > > >(services) are to be evaluated. The application of CMMI GP 2.9 in
      > > >each CMMI process area is a good start toward deciding what processes
      > > >(services) to evaluate. I also suggest that you consider processes
      > > >(services) that are higher risk as needing more timely and complete
      > > >evaluation coverage.
      > > >
      > > >PPQA is not necessarily directly concerned with contract compliance
      > > >(other than perhaps Supplier Agreements/Contract s in SAM GP 2.9 -
      > > but
      > > >this GP is more concerned with internal compliance/conforma nce to
      > > >"SAM" practices in your own process).
      > > >
      > > >Many companies, however, do conduct management verifications to
      > > >assess contract compliance within major projects/programs. Although
      > > >this may be considered outside the scope of CMMI-Dev, I know it to be
      > > >a best business practice. Whether you consider these to be audits or
      > > >verifications is probably less important than having them done by the
      > > >right people for the right reasons using effective methods.
      > > >
      > > >Hope this is helpful to you.
      > > >Best Regards,
      > > >Bruce
      > > >www.alderonconsult i ng.com
      > > >
      > > >At 10:27 AM 8/28/2008, you wrote:
      > > >
      > > > >PPQA discusses objective evaluation of services against applicable
      > > > >process descriptions, standards, and procedures. The first question
      > > > >would be what services are contemplated in the context of the CMMI
      > > > >development constellation? System operations and maintenance? Help
      > > > >desk? The second question would be how do we distinguish compliance
      > > > >with a contract requirement (verification) from compliance with
      > > > >standards. If the contract called for ITIL compliance, would
      > > > >evaluation of the conformance be a PPQA evaluation or a
      > > verification
      > > > >evaluation? The draft CMMI for services model does not provide much
      > > > >additional insight.
      > > > >
      > > > >
      > > >
      > > >
      > >
      > >
      >
      >[Non-text portions of this message have been removed]
      >
      >[Non-text portions of this message have been removed]
      >
      >[Non-text portions of this message have been removed]
      >
      >

      [Non-text portions of this message have been removed]

      [Non-text portions of this message have been removed]





      [Non-text portions of this message have been removed]
    • Bruce R. Duncil
      Hi Jerry, You asked: if a company is rated at ML3 with two methodologies XP and RUP. After that, can the company use SRCUM and claim ML3? Yes, they can still
      Message 2 of 22 , Oct 2, 2008
      View Source
      • 0 Attachment
        Hi Jerry,
        You asked: "if a company is rated at ML3 with two methodologies XP
        and RUP. After that, can the company use SRCUM and claim ML3?"

        Yes, they can still claim to be ML 3 - on the methodologies that were
        included in the original rating. Any claim that their
        projects/organization was ML 3 compliant overall (or that included
        their SCRUM projects) without a follow-on appraisal that included
        that methodology, would most likely be at best misleading.

        This may also get back to your instantiation question. I typically
        see companies providing at least one project instantiation for each
        major methodology. I would not conduct an appraisal to rate an
        organization on only 2 of 3 methodologies IF the third (excluded)
        method was a significant part of their business, head count, or work volume.

        Bottom line: appraisal organizational scope must be representative of
        the organizational unit.

        Hope this is helpful to you.
        Best Regards,
        Bruce
        www.alderonconsulting.com


        At 02:38 PM 9/30/2008, you wrote:


        >Jerry,
        >
        >Once an organization has completed a SCAMPI A appraisal, the results
        >can be posted on the SEI website for up to 3 years. During that
        >time, the organization can certainly "claim" to still be operating
        >at ML3 regardless of changes to methodology - or management,
        >ownership, geographical location, etc.
        >
        >Whether they ARE still ML3 or not could only be determined by
        >another appraisal. In fact, it's pretty easy to understand why the
        >SEI introduced a 3-year time limit on the validity of appraisal
        >results - an organization is likely to undergo some rather
        >significant changes over that period of time and the likelihood of
        >maintaining the stated level of process maturity is somewhat diminished.
        >
        >So, bottom line, an organization that is rated ML3 using XP and RUP
        >can introduce SCRUM and still claim ML3 - at least until their
        >3-year time period expires.
        >
        >Regards,
        >
        >Pat
        >
        >----- Original Message -----
        >From: Jerry Zhu
        >To:
        ><mailto:cmmi_process_improvement%40yahoogroups.com>cmmi_process_improvement@yahoogroups.com
        >
        >Sent: Monday, September 29, 2008 8:58 PM
        >Subject: Re: [CMMi Process Improvement] Quality standards for
        >delivered services
        >
        >Pat,
        >
        >Still some questions.
        >
        >Methodology I meant development/management methodologies. Testing,
        >peer review etc are practices in my oppinion. Testing could be SP
        >3.3 of PI "Evaluate Assembled Product Component" for example. There
        >might be different testing practices but they are not methodologies.
        >
        >If a company adopted multiple methodologies, XP, RUP, SCRUM, FDD for
        >example. They are very different methodologies with different work
        >products and process descriptions. There will still be three focus
        >projects to cover all four metodologies?
        >
        > From another standpoint, if a company is rated at ML3 with two
        > methodologies XP and RUP. After that, can the company use SRCUM
        >and claim ML3?
        >
        >Thanks
        >
        >Jerry
        >
        >--- On Wed, 9/10/08, Patrick OToole
        ><<mailto:PACT.otoole%40att.net>PACT.otoole@...> wrote:
        >
        >Jerry,
        >
        >In theory, you are correct - if two different methodologies are used
        >within an organization, one would want to put 6+ instances in scope
        >for the appraisal - although admittedly that is not the way its
        >always done in practice.
        >
        >First of all, it depends on what the methodology happens to be. For
        >example, if we are talking about testing methodologies, or peer
        >review methodologies, we would tend to cover the multiple
        >methodologies with non-focus projects rather than focus projects.
        >
        >[Caution, Yogi Berra statement ahead...] Non-focus projects would
        >allow us to focus solely on the unique methodologies better than
        >focus projects would. That is, the evaluation of non-focus projects
        >would concentrate solely on the way automated testing is planned and
        >executed (as opposed to the manual testing methods used on the focus
        >projects, thereby avoiding the overhead of adding 3 more
        >instantiations of PP, PMC, TS, etc.
        >
        >Second, in order to constrain time and money, such an appraisal may
        >only include 2 instances of each methodology - 2 focus projects that
        >used methodology #1, and 1 focus + 1 non-focus project that employed
        >methodology #2. Of course it would be more robust to include 3+3,
        >but the risk vs. cost tradeoff will usually tilt toward favoring cost.
        >
        >Regards,
        >
        >Pat
        >
        >----- Original Message -----
        >From: Jerry Zhu
        >To: cmmi_process_ improvement@ yahoogroups. com
        >Sent: Tuesday, September 09, 2008 8:30 PM
        >Subject: Re: [CMMi Process Improvement] Quality standards for
        >delivered services
        >
        >Hi Pat,
        >
        >You made it cristal clear. Thank you so much. Most of the CMMI
        >practices will be implemented three times by the three focus
        >projects and the left over CMMI practices shall be picked up by
        >non-focus projects. A fourth focus project could significantly
        >reduce the # of non-focus projects needed. The non-focus projects
        >should also include "deviate" practices in order to be
        >representative as a whole.
        >
        >Now continuing Bruce's comments, I think that the three plus focus
        >projects should be corresponding to a single methodology. If two
        >methodologies are used in the organization and both are appraised,
        >each will require 3+ focus projects to be representative for the
        >methodology. That will make it 6+ focus projects. That is, three
        >focus projects only can not be possibly reprsentative for the two
        >methodologies.
        >
        >Really appreciated.
        >
        >Jerry
        >
        >--- On Tue, 9/9/08, Patrick OToole <PACT.otoole@ att.net> wrote:
        >
        >Jerry,
        >
        >[Note: I have not been following this thread as closely as some, so
        >my comments may be a BIT out of context. However...]
        >
        >The minimum of three implementation instances for each practice
        >(note - not necessarily from 3 focus projects) ensures that the
        >appraisal team gains a more robust understanding of how that
        >practice is performed across the organization, and provides greater
        >confidence that the practice is implemented in all applicable
        >projects. This allows the appraisal team to make more informed
        >decisions with respect to Organization Unit practice characterizations.
        >
        >Remember that the 3+ focus projects are merely "representative" of
        >the full organizational population, so inferences drawn by examining
        >the activities in the focus project SHOULD be extendable to all
        >(otherwise, they weren't very "representative" projects, were they?)
        >
        >Granted, there may be differences in the activities performed by
        >other projects. If these are deemed consequential, then those
        >activities should be put in scope for appraisal examination by
        >including the projects that perform them as "non-focus" projects -
        >in which case those projects would have a subset of their activities
        >evaluated during the appraisal.
        >
        >Remember, too that three focus projects is the MINIMUM, the
        >appraisal may require many more projects to get a good
        >representative sample. I previously mentioned the continuous
        >representation appraisal that I have coming up in November where we
        >had to put 20 projects in scope for TS and VER - it's the only way
        >we could get a robust view of how the 200+ projects do their work
        >(across multiple platforms, application domains, modes (development
        >vs. maintenance) , etc.).
        >
        >If a PA, say DAR, is not needed by any of the projects, then 3+
        >non-focus projects would be submitted to allow the appraisal team to
        >evaluate those practices appropriately. Remember, too, that some
        >process areas, like DAR, may also be performed at the organizational
        >level, so those instances may be included as "non-focus projects" as well.
        >
        >I haven't been following the thread enough to comment on your PAL
        >assertion, so I'll leave that to others....
        >
        >Regards,
        >
        >Pat
        >
        >----- Original Message -----
        >From: Jerry Zhu
        >To: cmmi_process_ improvement@ yahoogroups. com
        >Sent: Monday, September 08, 2008 8:05 PM
        >Subject: RE: [CMMi Process Improvement] Quality standards for
        >delivered services
        >
        >Hi Bruce,
        >
        >Very helpful. I was wondering why minimum three focus projects. It
        >is because all CMMI PAs' goals and practices are covered by the
        >three projects? In other words, the sum of all three projects' needs
        >will need to cover all CMMI PAs? Which means all goals and their
        >practices may be covered by three focus projects at least once? What
        >about a PA is not needed by any of the three focus projects? The
        >needs of projects maybe subjective.
        >
        >I think that good (normalized) PAL should have one to many mapping
        >of goals/practices between PAL and project process, not the other way around.
        >
        >Thanks
        >
        >Jerry
        >
        >--- On Mon, 9/8/08, Bruce R. Duncil <brduncil@bellsouth . net> wrote:
        >
        >Hi Jerry,
        >You stated below that your question to me (should EVERY method and
        >process component in a PAL be used three times or at least once in
        >doing projects to pass a appraisal?) didn't get answered. That is a
        >different question! The answer to this question is 'no'.
        >
        >On 3 September, I provided the following in response to your earlier
        >post: "CMMI appraisals (and ratings) should cover an organizational
        >element that is fairly represented by all the processes it has in
        >use. Excluding one or more methodologies that are used in the company
        >from the appraisal should cause the organizational scope of the
        >appraisal to be adjusted and well documented accordingly. "
        >
        >So, when an organization is appraised, its projects are appraised and
        >their processes that they use are appraised. We don't look at the PAL
        >to determine if every single thing in it is used.
        >
        >Hope this is helpful to you.
        >Best Regards,
        >Bruce
        >www.alderonconsulti ng.com
        >
        >At 11:58 AM 9/5/2008, you wrote:
        >
        > >Jeff,
        > >
        > >Regarding mixed methods. I recommend to ready systems theory,
        > >particularly Ackoff to really communicate and understand Bruce's
        > >experience and his truthful explanation.
        > >
        > >Whatever are put in PAL and used in a project. The process/methods
        > >belong to a single system. XP is a system different from RUP. Some
        > >may say RUP and XP hybrid as merger of two systems. They are not, it
        > >is a new system, called home made or priproiatory system as many do
        > >to create their own system using altered parts of others.
        > >
        > >Assembling Buick's brake, BMW's transmission, Royce Roller's engine
        > >together do not give you a best car, In fact it is not even a car
        > >(parts do not match). YOu have altered the parts and fix them into
        > >your system different from Buick, BMW, Royce Rller.
        > >
        > >So I agree with Bruce that in PAL, there is only one system used for
        > >a project, it is either RUP, XP, or custom made. When custom made,
        > >they often borrow and customize parts from other systems. My
        > >question to Bruce and agreed with is that, it is rare to have two or
        > >more INCOMMEASURABLE systems placed in the same PAL. It is possible
        > >for a firm to have two or more sysstems. In economics terms placing
        > >two in one PAL may cost more for unnecessary complicatedness of use
        > >and maintenance. A truck is a system and convertable is another
        > >system. Merging the two not only is expensive and useless. Use two
        > >PALs for to separate systems maybe a good idea.
        > >
        > >My question to Bruce that I didn't get an answer is that should
        > >EVERY method and process component in a PAL be used three times or
        > >at least once in doing projects to pass a appraisal. Three focus
        > >projects may cover about 50% of what is in PAL. This may relate to
        > >the quality of PAL measured by normalization, precisness and conciseness.
        > >
        > >Lastly for the off topic issue. Is there light or heavy weight
        > >methodology for constructing a bridge? or only one rightweight? A
        > >thousand years ago there might have been two hundred ways of
        > >building a bridge where scientific knowledge is primative. Software
        > >industry we are now is as civil industry a thousand years ago. If
        > >there is a righweight in softwre industry, most won't recognize it,
        > >if even it were here now. So it is off the topic. back to busines as usual.
        > >
        > >best
        > >
        > >jerry
        > >
        > >--- On Fri, 9/5/08, Jeff Dalton - Broadsword
        > ><<mailto:jeff% 40Broadswordsolu tions.com>jeff@ Broadswordsolu
        > tions.com> wrote:
        > >
        > >Jerry,
        > >
        > >". In practice, does that ever happen?"
        > >
        > >Yes! Unlike Bruce's experience, I often see "mixing" of methods - and a
        > >single, seamless PAL. I have one client , for example, that has their own
        > >method, "xRUP," that utilizes concepts from RUP and XP together, and it
        > >works for them. Being a purist often does not result in success.
        > >
        > >I don't understand your statement about "right-weight. " That seems like a
        > >cop-out to me as it has zero definition or meaning (although it seems
        > >impressive to me). What is that? The concepts of lightweight and
        > >heavyweight (and everything in-between) are based upon the context of the
        > >organization, their needs, constraints, and tailoring strategy. There is no
        > >way to define a standard for "right."
        > >
        > >Warm Regards,
        > >
        > >Jeff
        > >
        > >Jeff Dalton, President and CEO
        > >
        > >Broadsword
        > >
        > >SCAMPI Lead Appraiser, CMMI Instructor, SCAMPI Team Leader Observer
        > >
        > >248.341.3367 (office)
        > >
        > >248.709.4775 (cell)
        > >
        > >248.341.3672 (fax)
        > >
        > >broadswordlogo. tif
        > >
        > >Visit our website at
        > <<http://www.broadswo>http://www.broadswo>http://www.broadswo rdsolutions. com
        > >
        > >Visit Jeff's blog at:
        > <<http://www.asktheCM>http://www.asktheCM>http://www.asktheCM
        > >MIAppraiser. com
        > ><<<http://asktheCMMIAp>http://asktheCMMIAp>http://asktheCMMIAp
        > praiser.blogspot .com>
        > >
        > >From: cmmi_process_ improvement@ yahoogroups. com
        > >[mailto:cmmi_ process_ improvement@ yahoogroups. com] On Behalf Of Jerry Zhu
        > >Sent: Thursday, September 04, 2008 7:18 PM
        > >To: cmmi_process_ improvement@ yahoogroups. com
        > >Subject: Re: [CMMi Process Improvement] Quality standards for delivered
        > >services
        > >
        > >Orhan,
        > >
        > >In non-theory, yes you can put anything in PAL. In practice, does that ever
        > >happen? I think Bruce's rich exprience is common and representative. In
        > >economic terms, placing two incommensurable methodologies together cost more
        > >(both creating and using it) than creating two separate PAL. Well it depends
        > >on how you organize it. It maybe between disjoint and cojoint sets to share
        > >comment method elements.
        > >
        > >Both heavyweight and lightweight methodologies are unhealthy methodologies.
        > >I'd use rightweight methodology (the methodology) . But it is off the topic
        > >in this forum.
        > >
        > >Jerry
        > >
        > >--- On Thu, 9/4/08, Orhan Kalayci <orhan.kalayci@ gmail.com
        > ><mailto:orhan. kalayci%40gmail. com> > wrote:
        > >
        > >From: Orhan Kalayci <orhan.kalayci@ gmail.com
        > ><mailto:orhan. kalayci%40gmail. com> >
        > >Subject: Re: [CMMi Process Improvement] Quality standards for delivered
        > >services
        > >To: "cmmi_process_ improvement@ yahoogroups. com
        > ><mailto:cmmi_ process_improvem ent%40yahoogroup s.com> "
        > ><cmmi_process_ improvement@ yahoogroups. com
        > ><mailto:cmmi_ process_improvem ent%40yahoogroup s.com> >
        > >Date: Thursday, September 4, 2008, 2:47 PM
        > >
        > >Hi jerry,
        > >
        > >It is the art of process architecture to develop a process asset
        > >library that support both conventional approaches such as waterfal and
        > >those of agile approaches at the same time. It is the art of
        > >developing tailoring guidelines not allow creating unhealthy defined
        > >process for a project.
        > >
        > >In short it it is possible to have waterfal and agile lifecycles in
        > >the same PAL. It is not only possible but healhty as well.
        > >
        > >Peace,
        > >Orhan
        > >Toronto
        > >
        > >Sent from my iPhone
        > >
        > >On Sep 2, 2008, at 21:15, Jerry Zhu <jerryyz@yahoo. com> wrote:
        > >
        > > > Hi Bruce,
        > > >
        > > > I am not sure I was clear myself.
        > > >
        > > > I meant if there were two disjoint sets of activities and work
        > > > products. Each alone can compose a single project. The two sets may
        > > > overlap.
        > > >
        > > > For example, two methodologies waterfall and XP are both stored in
        > > > the same process asset library. In a project, half waterfall and
        > > > half XP actitivies and work products are used. Certainly the project
        > > > failed outright but PPQA passed.
        > > >
        > > > My question really is whether a CMMI rating corresponds to a single
        > > > methodology? If yes, then the company is rated on SCRUM or RUP or
        > > > propriatory one. This seems to me that the company gets stuck with
        > > > one methodology?
        > > >
        > > > Regards
        > > >
        > > > Jerry
        > > >
        > > > --- On Wed, 9/3/08, Bruce R. Duncil <brduncil@bellsouth . net> wrote:
        > > >
        > > > Hi Jerry,
        > > > It requires multiple activities to create the many work products that
        > > > compose a single project. Yes, there will typically be many
        > > > procedures and product specifications/ descriptions in a robust,
        > > > well-defined set of processes. QA selects procedures and
        > > > specifications/ descriptions against which to evaluate compliance
        > > > (PPQA SP 1.1) and conformance (PPQA SP 1.2).
        > > >
        > > > Best Regards,
        > > > Bruce
        > > >
        > > > At 10:12 AM 9/2/2008, you wrote:
        > > >
        > > > >Hi Bruce,
        > > > >
        > > > >I have a question about "activities" and "products" mentioned in a)
        > > > >and b) below. Is {activities, products} and project one-one
        > > > >relationship? If not, there will be different sets of {activities,
        > > > >products}to deliver the same project. In this case, there would be
        > > > >multiple procedures and product specs and descriptions for the same
        > > > >practice in a project.
        > > > >
        > > > >Thanks
        > > > >
        > > > >Jerry
        > > > >
        > > > >--- On Mon, 9/1/08, Bruce R. Duncil
        > > > ><<mailto:brduncil% 40bellsouth. net>brduncil@ bellsouth. net> wrote:
        > > > >
        > > > >Hi,
        > > > >PPQA evaluations are intended to ensure that a) activities are
        > > > >performed in a manner that is compliant with policies, processes, and
        > > > >procedures and b) products conform to their product specifications,
        > > > >standards, and descriptions. You have to select which activities
        > > > >(services) are to be evaluated. The application of CMMI GP 2.9 in
        > > > >each CMMI process area is a good start toward deciding what processes
        > > > >(services) to evaluate. I also suggest that you consider processes
        > > > >(services) that are higher risk as needing more timely and complete
        > > > >evaluation coverage.
        > > > >
        > > > >PPQA is not necessarily directly concerned with contract compliance
        > > > >(other than perhaps Supplier Agreements/Contract s in SAM GP 2.9 -
        > > > but
        > > > >this GP is more concerned with internal compliance/conforma nce to
        > > > >"SAM" practices in your own process).
        > > > >
        > > > >Many companies, however, do conduct management verifications to
        > > > >assess contract compliance within major projects/programs. Although
        > > > >this may be considered outside the scope of CMMI-Dev, I know it to be
        > > > >a best business practice. Whether you consider these to be audits or
        > > > >verifications is probably less important than having them done by the
        > > > >right people for the right reasons using effective methods.
        > > > >
        > > > >Hope this is helpful to you.
        > > > >Best Regards,
        > > > >Bruce
        > > > >www.alderonconsult i ng.com
        > > > >
        > > > >At 10:27 AM 8/28/2008, you wrote:
        > > > >
        > > > > >PPQA discusses objective evaluation of services against applicable
        > > > > >process descriptions, standards, and procedures. The first question
        > > > > >would be what services are contemplated in the context of the CMMI
        > > > > >development constellation? System operations and maintenance? Help
        > > > > >desk? The second question would be how do we distinguish compliance
        > > > > >with a contract requirement (verification) from compliance with
        > > > > >standards. If the contract called for ITIL compliance, would
        > > > > >evaluation of the conformance be a PPQA evaluation or a
        > > > verification
        > > > > >evaluation? The draft CMMI for services model does not provide much
        > > > > >additional insight.
        > > > > >
        > > > > >
        > > > >
        > > > >
        > > >
        > > >
        > >
        > >[Non-text portions of this message have been removed]
        > >
        > >[Non-text portions of this message have been removed]
        > >
        > >[Non-text portions of this message have been removed]
        > >
        > >
        >
        >[Non-text portions of this message have been removed]
        >
        >[Non-text portions of this message have been removed]
        >
        >[Non-text portions of this message have been removed]
        >
        >
      • Jeffrey Dalton
        All, There is a simple answer to this - and Pat came pretty close. Neither the SCAMPI Method nor the SEI speaks to verifying sustainability - it s not part of
        Message 3 of 22 , Oct 2, 2008
        View Source
        • 0 Attachment
          All,

          There is a simple answer to this - and Pat came pretty
          close.

          Neither the SCAMPI Method nor the SEI speaks to
          verifying sustainability - it's not part of the scope
          of the appraisal method. ISO has a concept of
          "surveillance audits" that don't exist in SCAMPI.

          An appraisal is a little bit like a new car . . .
          smells and drives like new but it only REALLY new
          until the moment you start the engine and drive down
          the block.

          Based on current rules, a company can assert they have
          reached ML3 for three years from the successful
          appraisal completion date REGARDLESS of what they do
          (SCRUM, XP, or chaos).

          Nothing about their appraisal is related to what
          happens after the final findings briefing. They can
          claim anything they want.

          The plain truth is that while a company "achieves" a
          level of maturity or capability, but that was never
          intended to address whether they are sustaining that
          performance.

          Regards,

          Jeff


          --- "Bruce R. Duncil" <brduncil@...> wrote:

          > Hi Jerry,
          > You asked: "if a company is rated at ML3 with two
          > methodologies XP
          > and RUP. After that, can the company use SRCUM and
          > claim ML3?"
          >
          > Yes, they can still claim to be ML 3 - on the
          > methodologies that were
          > included in the original rating. Any claim that
          > their
          > projects/organization was ML 3 compliant overall (or
          > that included
          > their SCRUM projects) without a follow-on appraisal
          > that included
          > that methodology, would most likely be at best
          > misleading.
          >
          > This may also get back to your instantiation
          > question. I typically
          > see companies providing at least one project
          > instantiation for each
          > major methodology. I would not conduct an appraisal
          > to rate an
          > organization on only 2 of 3 methodologies IF the
          > third (excluded)
          > method was a significant part of their business,
          > head count, or work volume.
          >
          > Bottom line: appraisal organizational scope must be
          > representative of
          > the organizational unit.
          >
          > Hope this is helpful to you.
          > Best Regards,
          > Bruce
          > www.alderonconsulting.com
          >
          >
          > At 02:38 PM 9/30/2008, you wrote:
          >
          >
          > >Jerry,
          > >
          > >Once an organization has completed a SCAMPI A
          > appraisal, the results
          > >can be posted on the SEI website for up to 3 years.
          > During that
          > >time, the organization can certainly "claim" to
          > still be operating
          > >at ML3 regardless of changes to methodology - or
          > management,
          > >ownership, geographical location, etc.
          > >
          > >Whether they ARE still ML3 or not could only be
          > determined by
          > >another appraisal. In fact, it's pretty easy to
          > understand why the
          > >SEI introduced a 3-year time limit on the validity
          > of appraisal
          > >results - an organization is likely to undergo some
          > rather
          > >significant changes over that period of time and
          > the likelihood of
          > >maintaining the stated level of process maturity is
          > somewhat diminished.
          > >
          > >So, bottom line, an organization that is rated ML3
          > using XP and RUP
          > >can introduce SCRUM and still claim ML3 - at least
          > until their
          > >3-year time period expires.
          > >
          > >Regards,
          > >
          > >Pat
          > >
          > >----- Original Message -----
          > >From: Jerry Zhu
          > >To:
          >
          ><mailto:cmmi_process_improvement%40yahoogroups.com>cmmi_process_improvement@yahoogroups.com
          >
          > >
          > >Sent: Monday, September 29, 2008 8:58 PM
          > >Subject: Re: [CMMi Process Improvement] Quality
          > standards for
          > >delivered services
          > >
          > >Pat,
          > >
          > >Still some questions.
          > >
          > >Methodology I meant development/management
          > methodologies. Testing,
          > >peer review etc are practices in my oppinion.
          > Testing could be SP
          > >3.3 of PI "Evaluate Assembled Product Component"
          > for example. There
          > >might be different testing practices but they are
          > not methodologies.
          > >
          > >If a company adopted multiple methodologies, XP,
          > RUP, SCRUM, FDD for
          > >example. They are very different methodologies with
          > different work
          > >products and process descriptions. There will still
          > be three focus
          > >projects to cover all four metodologies?
          > >
          > > From another standpoint, if a company is rated at
          > ML3 with two
          > > methodologies XP and RUP. After that, can the
          > company use SRCUM
          > >and claim ML3?
          > >
          > >Thanks
          > >
          > >Jerry
          > >
          > >--- On Wed, 9/10/08, Patrick OToole
          > ><<mailto:PACT.otoole%40att.net>PACT.otoole@...>
          > wrote:
          > >
          > >Jerry,
          > >
          > >In theory, you are correct - if two different
          > methodologies are used
          > >within an organization, one would want to put 6+
          > instances in scope
          > >for the appraisal - although admittedly that is not
          > the way its
          > >always done in practice.
          > >
          > >First of all, it depends on what the methodology
          > happens to be. For
          > >example, if we are talking about testing
          > methodologies, or peer
          > >review methodologies, we would tend to cover the
          > multiple
          > >methodologies with non-focus projects rather than
          > focus projects.
          > >
          > >[Caution, Yogi Berra statement ahead...] Non-focus
          > projects would
          > >allow us to focus solely on the unique
          > methodologies better than
          > >focus projects would. That is, the evaluation of
          > non-focus projects
          > >would concentrate solely on the way automated
          > testing is planned and
          > >executed (as opposed to the manual testing methods
          > used on the focus
          > >projects, thereby avoiding the overhead of adding 3
          > more
          > >instantiations of PP, PMC, TS, etc.
          > >
          > >Second, in order to constrain time and money, such
          > an appraisal may
          > >only include 2 instances of each methodology - 2
          > focus projects that
          > >used methodology #1, and 1 focus + 1 non-focus
          > project that employed
          > >methodology #2. Of course it would be more robust
          > to include 3+3,
          > >but the risk vs. cost tradeoff will usually tilt
          > toward favoring cost.
          > >
          > >Regards,
          > >
          > >Pat
          > >
          > >----- Original Message -----
          > >From: Jerry Zhu
          > >To: cmmi_process_ improvement@ yahoogroups. com
          > >Sent: Tuesday, September 09, 2008 8:30 PM
          > >Subject: Re: [CMMi Process Improvement] Quality
          > standards for
          > >delivered services
          > >
          > >Hi Pat,
          > >
          > >You made it cristal clear. Thank you so much. Most
          > of the CMMI
          > >practices will be implemented three times by the
          > three focus
          > >projects and the left over CMMI practices shall be
          > picked up by
          > >non-focus projects. A fourth focus project could
          > significantly
          > >reduce the # of non-focus projects needed. The
          > non-focus projects
          > >should also include "deviate" practices in order to
          > be
          > >representative as a whole.
          > >
          > >Now continuing Bruce's comments, I think that the
          > three plus focus
          > >projects should be corresponding to a single
          > methodology. If two
          > >methodologies are used in the organization and both
          > are appraised,
          > >each will require 3+ focus projects to be
          > representative
          === message truncated ===
        • Bruce R. Duncil
          There are at least three major concepts inherent in the CMMI/SCAMPI that address sustainability . Many may argue whether they are necessary and sufficient.
          Message 4 of 22 , Oct 3, 2008
          View Source
          • 0 Attachment
            There are at least three major concepts inherent in the CMMI/SCAMPI
            that address "sustainability". Many may argue whether they are
            necessary and sufficient.

            Institutionalization: This is defined as "the ingrained way of doing
            business that an organization follows routinely as part of its
            corporate culture". Our role as appraisers, in my view, is to verify
            institutionalization has occurred and that the culture can sustain
            the processes. This is embodied in whether or not the
            "institutionalization" practices (GPs) are in fact implemented and
            the goals (GGs) are achieved. As an appraiser, I do not accept the
            response I once received from a high maturity hopeful during an
            appraisal that said in effect: 'our company has implemented the GPs
            once in each project/organizational unit and the practice/process
            area is therefore institutionalized'.

            Capability: When applied to a process, this is defined as "a process
            that can satisfy its specified product quality, service quality, and
            process-performance objectives". While this is most often applied to
            processes in a high maturity context, it most certainly can and
            should apply to lower maturity (Level 2 and 3) processes in a non- or
            semi-quantitative way. The concepts of a "well-defined" and
            "effective" process have always been ingrained in the SEI models and
            appraisal methods and in process improvement, for that matter.

            Appraisal Record: Sponsors are charged with maintaining the appraisal
            record. Their affirmations in the Appraisal Disclosure Statement
            outline their responsibilities. That doesn't mean that things can't
            or don't change. It does strongly imply and encourage a
            responsibility for maintaining a "compliance" posture - if they
            continue to assert an SEI accepted rating.

            As appraisers, I believe that our responsibilities go well beyond
            just agreeing whether or not proffered evidence indicates practice
            implementation and that no weaknesses impact goals. I believe that we
            can, should and must evaluate whether the organization has actually
            put into place the capability of doing so on a repeatable,
            sustainable basis. After that, its up to the organization being rated.

            Nevertheless, it is always the responsibility of the customer to
            ensure they properly specify and actually receive what they pay for
            from any supplier.

            Best Regards,
            Bruce
            www.alderonconsulting.com



            At 02:24 PM 10/2/2008, you wrote:

            >All,
            >
            >There is a simple answer to this - and Pat came pretty
            >close.
            >
            >Neither the SCAMPI Method nor the SEI speaks to
            >verifying sustainability - it's not part of the scope
            >of the appraisal method. ISO has a concept of
            >"surveillance audits" that don't exist in SCAMPI.
            >
            >An appraisal is a little bit like a new car . . .
            >smells and drives like new but it only REALLY new
            >until the moment you start the engine and drive down
            >the block.
            >
            >Based on current rules, a company can assert they have
            >reached ML3 for three years from the successful
            >appraisal completion date REGARDLESS of what they do
            >(SCRUM, XP, or chaos).
            >
            >Nothing about their appraisal is related to what
            >happens after the final findings briefing. They can
            >claim anything they want.
            >
            >The plain truth is that while a company "achieves" a
            >level of maturity or capability, but that was never
            >intended to address whether they are sustaining that
            >performance.
            >
            >Regards,
            >
            >Jeff
            >
            >--- "Bruce R. Duncil"
            ><<mailto:brduncil%40bellsouth.net>brduncil@...> wrote:
            >
            > > Hi Jerry,
            > > You asked: "if a company is rated at ML3 with two
            > > methodologies XP
            > > and RUP. After that, can the company use SRCUM and
            > > claim ML3?"
            > >
            > > Yes, they can still claim to be ML 3 - on the
            > > methodologies that were
            > > included in the original rating. Any claim that
            > > their
            > > projects/organization was ML 3 compliant overall (or
            > > that included
            > > their SCRUM projects) without a follow-on appraisal
            > > that included
            > > that methodology, would most likely be at best
            > > misleading.
            > >
            > > This may also get back to your instantiation
            > > question. I typically
            > > see companies providing at least one project
            > > instantiation for each
            > > major methodology. I would not conduct an appraisal
            > > to rate an
            > > organization on only 2 of 3 methodologies IF the
            > > third (excluded)
            > > method was a significant part of their business,
            > > head count, or work volume.
            > >
            > > Bottom line: appraisal organizational scope must be
            > > representative of
            > > the organizational unit.
            > >
            > > Hope this is helpful to you.
            > > Best Regards,
            > > Bruce
            > > www.alderonconsulting.com
            > >
            > >
            > > At 02:38 PM 9/30/2008, you wrote:
            > >
            > >
            > > >Jerry,
            > > >
            > > >Once an organization has completed a SCAMPI A
            > > appraisal, the results
            > > >can be posted on the SEI website for up to 3 years.
            > > During that
            > > >time, the organization can certainly "claim" to
            > > still be operating
            > > >at ML3 regardless of changes to methodology - or
            > > management,
            > > >ownership, geographical location, etc.
            > > >
            > > >Whether they ARE still ML3 or not could only be
            > > determined by
            > > >another appraisal. In fact, it's pretty easy to
            > > understand why the
            > > >SEI introduced a 3-year time limit on the validity
            > > of appraisal
            > > >results - an organization is likely to undergo some
            > > rather
            > > >significant changes over that period of time and
            > > the likelihood of
            > > >maintaining the stated level of process maturity is
            > > somewhat diminished.
            > > >
            > > >So, bottom line, an organization that is rated ML3
            > > using XP and RUP
            > > >can introduce SCRUM and still claim ML3 - at least
            > > until their
            > > >3-year time period expires.
            > > >
            > > >Regards,
            > > >
            > > >Pat
            > > >
            > > >----- Original Message -----
            > > >From: Jerry Zhu
            > > >To:
            > >
            > ><mailto:cmmi_process_improvement%40yahoogroups.com><mailto:cmmi_pro
            > cess_improvement%40yahoogroups.com>cmmi_process_improvement@yahoogroups.com
            > >
            > > >
            > > >Sent: Monday, September 29, 2008 8:58 PM
            > > >Subject: Re: [CMMi Process Improvement] Quality
            > > standards for
            > > >delivered services
            > > >
            > > >Pat,
            > > >
            > > >Still some questions.
            > > >
            > > >Methodology I meant development/management
            > > methodologies. Testing,
            > > >peer review etc are practices in my oppinion.
            > > Testing could be SP
            > > >3.3 of PI "Evaluate Assembled Product Component"
            > > for example. There
            > > >might be different testing practices but they are
            > > not methodologies.
            > > >
            > > >If a company adopted multiple methodologies, XP,
            > > RUP, SCRUM, FDD for
            > > >example. They are very different methodologies with
            > > different work
            > > >products and process descriptions. There will still
            > > be three focus
            > > >projects to cover all four metodologies?
            > > >
            > > > From another standpoint, if a company is rated at
            > > ML3 with two
            > > > methodologies XP and RUP. After that, can the
            > > company use SRCUM
            > > >and claim ML3?
            > > >
            > > >Thanks
            > > >
            > > >Jerry
            > > >
            > > >--- On Wed, 9/10/08, Patrick OToole
            > > ><<mailto:PACT.otoole%40att.net><mailto:PACT.otoole%40att.net>PACT
            > .otoole@...>
            > > wrote:
            > > >
            > > >Jerry,
            > > >
            > > >In theory, you are correct - if two different
            > > methodologies are used
            > > >within an organization, one would want to put 6+
            > > instances in scope
            > > >for the appraisal - although admittedly that is not
            > > the way its
            > > >always done in practice.
            > > >
            > > >First of all, it depends on what the methodology
            > > happens to be. For
            > > >example, if we are talking about testing
            > > methodologies, or peer
            > > >review methodologies, we would tend to cover the
            > > multiple
            > > >methodologies with non-focus projects rather than
            > > focus projects.
            > > >
            > > >[Caution, Yogi Berra statement ahead...] Non-focus
            > > projects would
            > > >allow us to focus solely on the unique
            > > methodologies better than
            > > >focus projects would. That is, the evaluation of
            > > non-focus projects
            > > >would concentrate solely on the way automated
            > > testing is planned and
            > > >executed (as opposed to the manual testing methods
            > > used on the focus
            > > >projects, thereby avoiding the overhead of adding 3
            > > more
            > > >instantiations of PP, PMC, TS, etc.
            > > >
            > > >Second, in order to constrain time and money, such
            > > an appraisal may
            > > >only include 2 instances of each methodology - 2
            > > focus projects that
            > > >used methodology #1, and 1 focus + 1 non-focus
            > > project that employed
            > > >methodology #2. Of course it would be more robust
            > > to include 3+3,
            > > >but the risk vs. cost tradeoff will usually tilt
            > > toward favoring cost.
            > > >
            > > >Regards,
            > > >
            > > >Pat
            > > >
            > > >----- Original Message -----
            > > >From: Jerry Zhu
            > > >To: cmmi_process_ improvement@ yahoogroups. com
            > > >Sent: Tuesday, September 09, 2008 8:30 PM
            > > >Subject: Re: [CMMi Process Improvement] Quality
            > > standards for
            > > >delivered services
            > > >
            > > >Hi Pat,
            > > >
            > > >You made it cristal clear. Thank you so much. Most
            > > of the CMMI
            > > >practices will be implemented three times by the
            > > three focus
            > > >projects and the left over CMMI practices shall be
            > > picked up by
            > > >non-focus projects. A fourth focus project could
            > > significantly
            > > >reduce the # of non-focus projects needed. The
            > > non-focus projects
            > > >should also include "deviate" practices in order to
            > > be
            > > >representative as a whole.
            > > >
            > > >Now continuing Bruce's comments, I think that the
            > > three plus focus
            > > >projects should be corresponding to a single
            > > methodology. If two
            > > >methodologies are used in the organization and both
            > > are appraised,
            > > >each will require 3+ focus projects to be
            > > representative
            >=== message truncated ===
            >
          • Bruce R. Duncil
            Hi Jeff, With respect to sustainability and surveillance audits I would invite your attention to OPF SP 1.2. While a company MIGHT have only done this only
            Message 5 of 22 , Oct 3, 2008
            View Source
            • 0 Attachment
              Hi Jeff,
              With respect to sustainability and "surveillance audits" I would
              invite your attention to OPF SP 1.2.

              While a company MIGHT have only done this only once prior to their
              Level 3 appraisal, it is not written as a one-time activity.

              Best Regards,
              Bruce



              At 02:24 PM 10/2/2008, you wrote:

              >All,
              >
              >There is a simple answer to this - and Pat came pretty
              >close.
              >
              >Neither the SCAMPI Method nor the SEI speaks to
              >verifying sustainability - it's not part of the scope
              >of the appraisal method. ISO has a concept of
              >"surveillance audits" that don't exist in SCAMPI.
              >
              >An appraisal is a little bit like a new car . . .
              >smells and drives like new but it only REALLY new
              >until the moment you start the engine and drive down
              >the block.
              >
              >Based on current rules, a company can assert they have
              >reached ML3 for three years from the successful
              >appraisal completion date REGARDLESS of what they do
              >(SCRUM, XP, or chaos).
              >
              >Nothing about their appraisal is related to what
              >happens after the final findings briefing. They can
              >claim anything they want.
              >
              >The plain truth is that while a company "achieves" a
              >level of maturity or capability, but that was never
              >intended to address whether they are sustaining that
              >performance.
              >
              >Regards,
              >
              >Jeff
              >
              >--- "Bruce R. Duncil"
              ><<mailto:brduncil%40bellsouth.net>brduncil@...> wrote:
              >
              > > Hi Jerry,
              > > You asked: "if a company is rated at ML3 with two
              > > methodologies XP
              > > and RUP. After that, can the company use SRCUM and
              > > claim ML3?"
              > >
              > > Yes, they can still claim to be ML 3 - on the
              > > methodologies that were
              > > included in the original rating. Any claim that
              > > their
              > > projects/organization was ML 3 compliant overall (or
              > > that included
              > > their SCRUM projects) without a follow-on appraisal
              > > that included
              > > that methodology, would most likely be at best
              > > misleading.
              > >
              > > This may also get back to your instantiation
              > > question. I typically
              > > see companies providing at least one project
              > > instantiation for each
              > > major methodology. I would not conduct an appraisal
              > > to rate an
              > > organization on only 2 of 3 methodologies IF the
              > > third (excluded)
              > > method was a significant part of their business,
              > > head count, or work volume.
              > >
              > > Bottom line: appraisal organizational scope must be
              > > representative of
              > > the organizational unit.
              > >
              > > Hope this is helpful to you.
              > > Best Regards,
              > > Bruce
              > > www.alderonconsulting.com
              > >
              > >
              > > At 02:38 PM 9/30/2008, you wrote:
              > >
              > >
              > > >Jerry,
              > > >
              > > >Once an organization has completed a SCAMPI A
              > > appraisal, the results
              > > >can be posted on the SEI website for up to 3 years.
              > > During that
              > > >time, the organization can certainly "claim" to
              > > still be operating
              > > >at ML3 regardless of changes to methodology - or
              > > management,
              > > >ownership, geographical location, etc.
              > > >
              > > >Whether they ARE still ML3 or not could only be
              > > determined by
              > > >another appraisal. In fact, it's pretty easy to
              > > understand why the
              > > >SEI introduced a 3-year time limit on the validity
              > > of appraisal
              > > >results - an organization is likely to undergo some
              > > rather
              > > >significant changes over that period of time and
              > > the likelihood of
              > > >maintaining the stated level of process maturity is
              > > somewhat diminished.
              > > >
              > > >So, bottom line, an organization that is rated ML3
              > > using XP and RUP
              > > >can introduce SCRUM and still claim ML3 - at least
              > > until their
              > > >3-year time period expires.
              > > >
              > > >Regards,
              > > >
              > > >Pat
              > > >
              > > >----- Original Message -----
              > > >From: Jerry Zhu
              > > >To:
              > >
              > ><mailto:cmmi_process_improvement%40yahoogroups.com><mailto:cmmi_pro
              > cess_improvement%40yahoogroups.com>cmmi_process_improvement@yahoogroups.com
              > >
              > > >
              > > >Sent: Monday, September 29, 2008 8:58 PM
              > > >Subject: Re: [CMMi Process Improvement] Quality
              > > standards for
              > > >delivered services
              > > >
              > > >Pat,
              > > >
              > > >Still some questions.
              > > >
              > > >Methodology I meant development/management
              > > methodologies. Testing,
              > > >peer review etc are practices in my oppinion.
              > > Testing could be SP
              > > >3.3 of PI "Evaluate Assembled Product Component"
              > > for example. There
              > > >might be different testing practices but they are
              > > not methodologies.
              > > >
              > > >If a company adopted multiple methodologies, XP,
              > > RUP, SCRUM, FDD for
              > > >example. They are very different methodologies with
              > > different work
              > > >products and process descriptions. There will still
              > > be three focus
              > > >projects to cover all four metodologies?
              > > >
              > > > From another standpoint, if a company is rated at
              > > ML3 with two
              > > > methodologies XP and RUP. After that, can the
              > > company use SRCUM
              > > >and claim ML3?
              > > >
              > > >Thanks
              > > >
              > > >Jerry
              > > >
              > > >--- On Wed, 9/10/08, Patrick OToole
              > > ><<mailto:PACT.otoole%40att.net><mailto:PACT.otoole%40att.net>PACT
              > .otoole@...>
              > > wrote:
              > > >
              > > >Jerry,
              > > >
              > > >In theory, you are correct - if two different
              > > methodologies are used
              > > >within an organization, one would want to put 6+
              > > instances in scope
              > > >for the appraisal - although admittedly that is not
              > > the way its
              > > >always done in practice.
              > > >
              > > >First of all, it depends on what the methodology
              > > happens to be. For
              > > >example, if we are talking about testing
              > > methodologies, or peer
              > > >review methodologies, we would tend to cover the
              > > multiple
              > > >methodologies with non-focus projects rather than
              > > focus projects.
              > > >
              > > >[Caution, Yogi Berra statement ahead...] Non-focus
              > > projects would
              > > >allow us to focus solely on the unique
              > > methodologies better than
              > > >focus projects would. That is, the evaluation of
              > > non-focus projects
              > > >would concentrate solely on the way automated
              > > testing is planned and
              > > >executed (as opposed to the manual testing methods
              > > used on the focus
              > > >projects, thereby avoiding the overhead of adding 3
              > > more
              > > >instantiations of PP, PMC, TS, etc.
              > > >
              > > >Second, in order to constrain time and money, such
              > > an appraisal may
              > > >only include 2 instances of each methodology - 2
              > > focus projects that
              > > >used methodology #1, and 1 focus + 1 non-focus
              > > project that employed
              > > >methodology #2. Of course it would be more robust
              > > to include 3+3,
              > > >but the risk vs. cost tradeoff will usually tilt
              > > toward favoring cost.
              > > >
              > > >Regards,
              > > >
              > > >Pat
              > > >
              > > >----- Original Message -----
              > > >From: Jerry Zhu
              > > >To: cmmi_process_ improvement@ yahoogroups. com
              > > >Sent: Tuesday, September 09, 2008 8:30 PM
              > > >Subject: Re: [CMMi Process Improvement] Quality
              > > standards for
              > > >delivered services
              > > >
              > > >Hi Pat,
              > > >
              > > >You made it cristal clear. Thank you so much. Most
              > > of the CMMI
              > > >practices will be implemented three times by the
              > > three focus
              > > >projects and the left over CMMI practices shall be
              > > picked up by
              > > >non-focus projects. A fourth focus project could
              > > significantly
              > > >reduce the # of non-focus projects needed. The
              > > non-focus projects
              > > >should also include "deviate" practices in order to
              > > be
              > > >representative as a whole.
              > > >
              > > >Now continuing Bruce's comments, I think that the
              > > three plus focus
              > > >projects should be corresponding to a single
              > > methodology. If two
              > > >methodologies are used in the organization and both
              > > are appraised,
              > > >each will require 3+ focus projects to be
              > > representative
              >=== message truncated ===
              >
            • jeff@Broadswordsolutions.com
              Bruce, If one were to interpet the practice as another required SCAMPI appraisal, then I would agree, but as has been said many times on this board, there are
              Message 6 of 22 , Oct 3, 2008
              View Source
              • 0 Attachment
                Bruce,

                If one were to interpet the practice as another required SCAMPI appraisal, then I would agree, but as has been said many times on this board, there are many ways to do this outside of SCAMPI.

                If you re-read my post you'll note I was referring to the SCAMPI method, not the CMMI, and that practice does not exist in SCAMPI.

                Pehaps I'm wrong on this.
                Sent from my Verizon Wireless BlackBerry

                -----Original Message-----
                From: "Bruce R. Duncil" <brduncil@...>

                Date: Fri, 03 Oct 2008 09:16:32
                To: <cmmi_process_improvement@yahoogroups.com>
                Subject: Re: [CMMi Process Improvement] Quality standards for delivered
                services


                Hi Jeff,
                With respect to sustainability and "surveillance audits" I would
                invite your attention to OPF SP 1.2.

                While a company MIGHT have only done this only once prior to their
                Level 3 appraisal, it is not written as a one-time activity.

                Best Regards,
                Bruce



                At 02:24 PM 10/2/2008, you wrote:

                >All,
                >
                >There is a simple answer to this - and Pat came pretty
                >close.
                >
                >Neither the SCAMPI Method nor the SEI speaks to
                >verifying sustainability - it's not part of the scope
                >of the appraisal method. ISO has a concept of
                >"surveillance audits" that don't exist in SCAMPI.
                >
                >An appraisal is a little bit like a new car . . .
                >smells and drives like new but it only REALLY new
                >until the moment you start the engine and drive down
                >the block.
                >
                >Based on current rules, a company can assert they have
                >reached ML3 for three years from the successful
                >appraisal completion date REGARDLESS of what they do
                >(SCRUM, XP, or chaos).
                >
                >Nothing about their appraisal is related to what
                >happens after the final findings briefing. They can
                >claim anything they want.
                >
                >The plain truth is that while a company "achieves" a
                >level of maturity or capability, but that was never
                >intended to address whether they are sustaining that
                >performance.
                >
                >Regards,
                >
                >Jeff
                >
                >--- "Bruce R. Duncil"
                ><<mailto:brduncil%40bellsouth.net>brduncil@...> wrote:
                >
                > > Hi Jerry,
                > > You asked: "if a company is rated at ML3 with two
                > > methodologies XP
                > > and RUP. After that, can the company use SRCUM and
                > > claim ML3?"
                > >
                > > Yes, they can still claim to be ML 3 - on the
                > > methodologies that were
                > > included in the original rating. Any claim that
                > > their
                > > projects/organization was ML 3 compliant overall (or
                > > that included
                > > their SCRUM projects) without a follow-on appraisal
                > > that included
                > > that methodology, would most likely be at best
                > > misleading.
                > >
                > > This may also get back to your instantiation
                > > question. I typically
                > > see companies providing at least one project
                > > instantiation for each
                > > major methodology. I would not conduct an appraisal
                > > to rate an
                > > organization on only 2 of 3 methodologies IF the
                > > third (excluded)
                > > method was a significant part of their business,
                > > head count, or work volume.
                > >
                > > Bottom line: appraisal organizational scope must be
                > > representative of
                > > the organizational unit.
                > >
                > > Hope this is helpful to you.
                > > Best Regards,
                > > Bruce
                > > www.alderonconsulting.com
                > >
                > >
                > > At 02:38 PM 9/30/2008, you wrote:
                > >
                > >
                > > >Jerry,
                > > >
                > > >Once an organization has completed a SCAMPI A
                > > appraisal, the results
                > > >can be posted on the SEI website for up to 3 years.
                > > During that
                > > >time, the organization can certainly "claim" to
                > > still be operating
                > > >at ML3 regardless of changes to methodology - or
                > > management,
                > > >ownership, geographical location, etc.
                > > >
                > > >Whether they ARE still ML3 or not could only be
                > > determined by
                > > >another appraisal. In fact, it's pretty easy to
                > > understand why the
                > > >SEI introduced a 3-year time limit on the validity
                > > of appraisal
                > > >results - an organization is likely to undergo some
                > > rather
                > > >significant changes over that period of time and
                > > the likelihood of
                > > >maintaining the stated level of process maturity is
                > > somewhat diminished.
                > > >
                > > >So, bottom line, an organization that is rated ML3
                > > using XP and RUP
                > > >can introduce SCRUM and still claim ML3 - at least
                > > until their
                > > >3-year time period expires.
                > > >
                > > >Regards,
                > > >
                > > >Pat
                > > >
                > > >----- Original Message -----
                > > >From: Jerry Zhu
                > > >To:
                > >
                > ><mailto:cmmi_process_improvement%40yahoogroups.com><mailto:cmmi_pro
                > cess_improvement%40yahoogroups.com>cmmi_process_improvement@yahoogroups.com
                > >
                > > >
                > > >Sent: Monday, September 29, 2008 8:58 PM
                > > >Subject: Re: [CMMi Process Improvement] Quality
                > > standards for
                > > >delivered services
                > > >
                > > >Pat,
                > > >
                > > >Still some questions.
                > > >
                > > >Methodology I meant development/management
                > > methodologies. Testing,
                > > >peer review etc are practices in my oppinion.
                > > Testing could be SP
                > > >3.3 of PI "Evaluate Assembled Product Component"
                > > for example. There
                > > >might be different testing practices but they are
                > > not methodologies.
                > > >
                > > >If a company adopted multiple methodologies, XP,
                > > RUP, SCRUM, FDD for
                > > >example. They are very different methodologies with
                > > different work
                > > >products and process descriptions. There will still
                > > be three focus
                > > >projects to cover all four metodologies?
                > > >
                > > > From another standpoint, if a company is rated at
                > > ML3 with two
                > > > methodologies XP and RUP. After that, can the
                > > company use SRCUM
                > > >and claim ML3?
                > > >
                > > >Thanks
                > > >
                > > >Jerry
                > > >
                > > >--- On Wed, 9/10/08, Patrick OToole
                > > ><<mailto:PACT.otoole%40att.net><mailto:PACT.otoole%40att.net>PACT
                > .otoole@...>
                > > wrote:
                > > >
                > > >Jerry,
                > > >
                > > >In theory, you are correct - if two different
                > > methodologies are used
                > > >within an organization, one would want to put 6+
                > > instances in scope
                > > >for the appraisal - although admittedly that is not
                > > the way its
                > > >always done in practice.
                > > >
                > > >First of all, it depends on what the methodology
                > > happens to be. For
                > > >example, if we are talking about testing
                > > methodologies, or peer
                > > >review methodologies, we would tend to cover the
                > > multiple
                > > >methodologies with non-focus projects rather than
                > > focus projects.
                > > >
                > > >[Caution, Yogi Berra statement ahead...] Non-focus
                > > projects would
                > > >allow us to focus solely on the unique
                > > methodologies better than
                > > >focus projects would. That is, the evaluation of
                > > non-focus projects
                > > >would concentrate solely on the way automated
                > > testing is planned and
                > > >executed (as opposed to the manual testing methods
                > > used on the focus
                > > >projects, thereby avoiding the overhead of adding 3
                > > more
                > > >instantiations of PP, PMC, TS, etc.
                > > >
                > > >Second, in order to constrain time and money, such
                > > an appraisal may
                > > >only include 2 instances of each methodology - 2
                > > focus projects that
                > > >used methodology #1, and 1 focus + 1 non-focus
                > > project that employed
                > > >methodology #2. Of course it would be more robust
                > > to include 3+3,
                > > >but the risk vs. cost tradeoff will usually tilt
                > > toward favoring cost.
                > > >
                > > >Regards,
                > > >
                > > >Pat
                > > >
                > > >----- Original Message -----
                > > >From: Jerry Zhu
                > > >To: cmmi_process_ improvement@ yahoogroups. com
                > > >Sent: Tuesday, September 09, 2008 8:30 PM
                > > >Subject: Re: [CMMi Process Improvement] Quality
                > > standards for
                > > >delivered services
                > > >
                > > >Hi Pat,
                > > >
                > > >You made it cristal clear. Thank you so much. Most
                > > of the CMMI
                > > >practices will be implemented three times by the
                > > three focus
                > > >projects and the left over CMMI practices shall be
                > > picked up by
                > > >non-focus projects. A fourth focus project could
                > > significantly
                > > >reduce the # of non-focus projects needed. The
                > > non-focus projects
                > > >should also include "deviate" practices in order to
                > > be
                > > >representative as a whole.
                > > >
                > > >Now continuing Bruce's comments, I think that the
                > > three plus focus
                > > >projects should be corresponding to a single
                > > methodology. If two
                > > >methodologies are used in the organization and both
                > > are appraised,
                > > >each will require 3+ focus projects to be
                > > representative
                >=== message truncated ===
                >




                [Non-text portions of this message have been removed]
              Your message has been successfully submitted and would be delivered to recipients shortly.