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

Re: [CMMi Process Improvement] IT firm audit

Expand Messages
  • PACT.otoole@att.net
    Jerry, Well, for starters, an audit should be conducted “against something” – a standard, a model, a set of requirements, whatever. If not the CMMI,
    Message 1 of 50 , May 31, 2012
    • 0 Attachment
       
      Jerry,
       
      Well, for starters, an audit should be conducted “against something” – a standard, a model, a set of requirements, whatever.  If not the CMMI, then what?  Until you figure that out, you might want to avoid embarrassing yourself by suggesting that you perform an audit!
       
      I know that you aren’t necessarily a CMMI zealot, but check out generic practice 2.5 (GP2.5) – it kinda suggests that it’s best if the people performing an activity have the skills and knowledge to do so.  If you have to ask, “How do I do it?” I think you may want to rethink your capability of doing so.
       
      Just a thought...
       
      Pat
       
       
       
       
      From: Jerry Zhu
      Sent: Wednesday, May 30, 2012 8:12 PM
      Subject: [CMMi Process Improvement] IT firm audit
       
       

      Hello CMMI friends
       
      Will go to China next month to give a lecture for a software firm there. I was thinking to give an audit to assess their projects and processes. It is not CMMi related. Never did audit before. How to do it?
       
      Jerry Zhu
    • Joachim Bauchrowitz
      Orhan, When Lead Appraisers perform appraisals for customers, the first and most important aspect is RESPECT for all the people, who invested a lot of
      Message 50 of 50 , Jun 25, 2012
      • 0 Attachment
        Orhan,
        When Lead Appraisers perform appraisals for customers, the first and most important aspect is RESPECT for all the people, who invested a lot of ressources in improving processes, and I try to follow that idea under any circumstances.
        What characterizes a Lead Appraiser is basically, that (s)he points at "improvement opportunities" very clearly without fearing struggles, and I try to follow that, equally.

        When you re-read my statements, you will find, that I talked about your statements, and not about your person.  

        As you answered to my statements, you were extremely thin-skinned. You thought of slanderous statements against your person, but there aren't any. When you re-read my statements, you will find, that my "nasties" are positive, because they can lead to re-thinking of a cemented viewpoint. If not, I don't care.
        And do you want me to believe, that you don't ignore all the ideas and assists from a lot of professional volunteers, like Pat, or Ed, or David, ... ?
        And did you ever analyzed the fact, that the CMMI is a set of "Best Practices" from several hundred persons, so you ignore the experience of all the people by your changed modell without respection the trade mark of SEI?

        When I read your statements about failures of SEI, your hurting the feelings of every SEI employee and of every person who supports SEI statements in a very extreme way without any hestitation and doubt in very crude words.

        Conclusion:
        You don't have any right to claim for an apoligy: 1. Because there is no reason for an apology, and 2. Because you yourself don't fulfil the professional baseline for yourself, that you are demanding for others. That would be the pre-requisite to claim an apology

        And please stop to confuse everybody, who looks for insight on CMMI in this forum. You are hurting the bottomline from almost everybody of the community: CMMI is a good help for improving processes.

        Joachim



        Von:        Orhan KALAYCI <orhan.kalayci@...>
        An:        cmmi_process_improvement@yahoogroups.com
        Datum:        24.06.2012 05:15
        Betreff:        Re: Re: [CMMi Process Improvement] Clarifications - VER, VAL and MOA
        Gesendet von:        cmmi_process_improvement@yahoogroups.com




         

        Joachim,

        You are referring me as nasty and ignorant.  What makes you think that I am nasty and ignorant?  What did I say or do nasty?  How did you understand I am ignorant about the CMMI?  You don't know a thing about me but you feel free to label me as nasty and ignorant.  If you have any moral or civil integrity  (I believe you have) you will apology.  Why it is important for you to apology; probably, you noticed already that I don't care much about what people think about me as long as I know I am doing the right thing but I suggest you to apology for your own integrity and self-respect.   

        Regarding your two points below, see my response to Pat about how Maturity Levels apply to individuals, groups, and organizationals.  So, don't worry, Orhan's MLs will help organizations really improve their processes when the new home of CMMI changes the MLs according to Orhan's MLs,  you will see Orhan's ML1 applies to individuals, Orhan's ML2 applies to groups or projects, Orhan's ML3 applies to organization,...  For more details see my response to Pat's email in cmmi yahoo group.  

        I actually know why those certified CMMI experts don't like me, I simply know the CMMI better than them even though I am not a certified CMMI expert  :)  I am proud of being not certified by the SEI because the SEI was not able to understand/appreciate my level of CMMI know-how, that is one of the reasons why CMMI is moving out of the SEI into a new home.  In the past five years or so I was trying to educate the SEI and the rest of the community about the importance of not skipping levels, alternative practices, etc.  But they failed to understand my points and consequently failed to help organizations help improve their processes.  I am passionate about process improvement and I don't care if the SEI or the new home of the CMMI understand my points but I will continue to try to enlighten the community about the meaning (spirit) of maturity levels, importance of not skipping levels, alternative practices, etc.  BTW, without Orhan's MLs or with current definition of MLs, not skipping a level (i.e. not skipping ML2 and ML4) does not really make sense; to explain the meaning of skipping a level, one needs to have Orhan's definition of MLs that is, probably, why the SEI was not able to explain the meaning and importance of skipping a level.  

        I understand, it is not easy to swallow that someone who is not even a certified CMMI expert is talking about the CMMI with so many big words.  I know  you think you understand what I am talking about but I am pretty sure that you do not.  Anyhow, when you don't understand someone or you think he talks nonsense it does not make him nasty or ignorant.  if you don't understand someone talking on a subject which you are an expert on it; it means that he is either a mad person or genius.  I hope it is as simply as it is and clear to you.  

        If you will claim you understand what I am saying then please explain the meaning of skipping ML2 and going directly to ML3, give an example and explain why it is counter productive.  OR SIMPLY GIVE A REFERENCE FROM the SEI PUBLICATIONS IF YOU CAN FIND ONE :))  I will also very much appreciate if you give us a figure of % for how many organizations you know really experienced ML2 before going to ML3 i.e. not skipped ML2  -- I am not talking about skipping an appraisal for ML2, I am talking about experiencing the ML2 before going to ML3 and why it is important.  Unfortunately, in today's definition and understanding of MLs, the CMMI is not helping organization improve their processes, it is even counter-productive because it encourages organizations to skip "real" ML1 and ML2, i.e. Orhan's ML1 and ML2.  As I said, a further elaboration of Orhan's ML1 and ML2 see my coming emails in cmmi yahoo group.  

        Apparently, it is not enough to change only the SEI as home of CMMI, the SEI partners should also be re-considered to improve the overall service quality regarding the CMMI suites.  

        Peace,
        Orhan
        Toronto

        On Fri, Jun 15, 2012 at 4:40 AM, Joachim Bauchrowitz <Joachim.Bauchrowitz@...> wrote:
         

        Pat

        I strongly thank you for dealing with Orhan's ideas, because I often learned, that ignoring nasties leads to missing chances of improvement. Ignoring these inputs basically puts you on the same level of ignorance.


        Two things should be discussed additionally:


        1. CMMI is a model bound to support process improvement, and not process maturity measurement. I had some discussions with Watts, too, and he argued to leave the attention to management to get the cultural change, and not let the managers have the chance to point at "the engineerings" as a part of the problem by adding engineering processes to the first step of improvement.


        2. I had often ideas similar to N&I in development context, but found, that it's just too much for an organization to take. The base idea is right: you cannot calculate estimations without getting some structure in engineering processes, so you have to bring some basic professional level to these processes, but it doesn't have to be a level to pass an appraisal. In many organizations, the cultural shock of implementing and following processes of some maturity level leads to an implementation time of more than the average of 18 month for ML2 process areas in development organizations as reported by SEI. If you add engineering practices, you often get out of the maximum interval of patience of top level management for success. Top Managers simple plan to take the next step of career within the timeframe needed for that, therefore an implementation of N&I CL1 is simply not achievable in a market environment without Level-pressure from outside.


        The "best" thing in theory is not the best thing in practice.
        Being a process improvement guy means, that you have to have some patience and let experience grow. You cannot press experience into an organization, it has to grow, and that takes time.


        Just some ideas


        Joachim




        Von:        
        <PACT.otoole@...>
        An:        
        <cmmi_process_improvement@yahoogroups.com>
        Datum:        
        14.06.2012 20:07
        Betreff:        
        Re: [CMMi Process Improvement] Clarifications - VER, VAL and MOA
        Gesendet von:        
        cmmi_process_improvement@yahoogroups.com





         

         

        Orhan,
         

        Thanks – that helps.  In my text below, rather than using “Orhan’s ML1” – which could be perceived as pejorative, let me call it N&I ML1 (“New & Improved” ML1).

         

        So, in your view, N&I ML1 = CL1 in RD, TS, PI, VER, and VAL.

         

        Remember, that to be CL1, an organization has to perform all of the specific practices within that process area.  Failure to perform any single specific practice automatically results in a CL0 rating for that process area (which would then result in an overall N&I ML0 rating for the organization).

         

        Essentially you are suggesting that, to advance N&I ML0 to N&I ML1, organizations need to implement things like operational concepts and scenarios (RD SP3.1), alternative design analyses (TS SP1.1), peer reviews (VER SG2), and the establishment of integration strategies (PI) prior to implementing fundamental planning and tracking at ML2.

         

        You label N&I ML1 as “Performing” and describe it as “able to perform basic functions.”  If we were able to reinstitute the concept of base and advanced practices from CMMI-DEV v1.1, I might be more apt to agree – because then we could encourage an organization to “Gather Requirements” at CL1, and “Elicit Requirements” at CL2 or CL3.  But with the current one-size-fits-all approach for the specific practices, I perceive that many of the current practices go well beyond the term “basic functions.”

         

        Watts purposely positioned fundamental project management (PP, PMC, SAM) ahead of the Engineering process areas because he felt that most organizations needed to focus on establishing more reasonable plans – that is, allocating sufficient time and resources to do the engineering work – before they focused on improving those activities for which insufficient time and resources are allocated today.  When you’re trying to do 12 months of work in 8 months, it’s pretty hard to focus on implementing operational concepts, alternative design analyses, product integration strategies or even peer reviews.  (That is, when a man is drowning, it’s no time to teach him to swim.)

         

        I do understand and appreciate your basic point that the engineering work needs to get done and deserves some level of attention.  Watts would have argued that the RIGHT people to actually perform PP and PMC would be the engineers themselves – that THEY should be the ones planning their own engineering activities and tracking to make sure that they are going the way that they planned them.  Essentially, Watts would content that, at ML2, the engineers should be planning the work that would essentially accomplish CL1 in the various Engineering process areas.  After all, individual planning and accountability are the foundations of PSP and TSP.

         

        - - - - -

         

        Interestingly enough, the only difference between the current staged representation and the N&I maturity levels is that you have dropped the floor to ML0 and have inserted a new definition of ML1 – achieving CL1 in the Engineering process areas.  In fact, the N&I ML2 is exactly the same as the existing ML2 except organizations would also have to take RD, TS, PI, VER and VAL to CL2.

         

        As I’ve said before, the good news here is that you can adopt that EXACT approach today by using the continuous representation.  Convince your clients that you have built a better mouse trap and that they will achieve enhanced results by adopting your approach.  You could even get your approach MORE closely aligned with the established model by relabeling some of your terms as follows:

         

        N&I ML0 = “ML1” (that is, no change)

         

        N&I ML1 = “ML1+” (where the “+” indicates CL1 in RD, TS, PI, VER, and VAL)

         

        N&I ML2 = “ML2++” (where the “++” indicates CL2 in RD, TS, PI, VER, and VAL)

         

        N&I ML3-N&I ML5 – ML3-ML5 (no change)

         

         

        Rather than blaming the SEI for not listening, I’d suggest you get the data to PROVE you’ve built a better mousetrap.  Trust me, the SEI has 100+ Orhans screaming at them that the model should be changed to align with their view of the universe.  Empirical data goes a long way in making a more convincing argument.  The SEI has performance data related to the current structure of the model; demonstrate improved results with your preferred approach and you’ll get people’s undivided attention.  Unfortunately, until then, you’re just part of the screaming chorus!

         

        Regards,

         

        Pat

         

         

         

         

         

        From:
        Orhan KALAYCI
        Sent:
        Tuesday, June 12, 2012 9:11 AM

        To:
        cmmi_process_improvement@yahoogroups.com
        Subject:
        Re: [CMMi Process Improvement] Clarifications - VER, VAL and M&A

         

         

        Pat,
         

        Here is my model of maturity levels:
         

        ML0 - Incomplete

        ML1 - RD, TS, VER, VAL, PI (all at CL1)

        ML2 - REQM, PP, PMC, MA, PPQA, CM, SAM (all at CL2 + ML1 at CL2)

         

        The rest is the same as below list of yours Orhan's MLs.
         

        The term Specific PAs has been chosen for a reason.  The structure of Orhan's MLs are very similar to the CLs.

         

        Only ML1 is different for different constellations such as ML1 is Specific PAs at CL1 the rest is the same for all constellations (For SAM - in fact, I believe it applies to CMMI-Acq as well, when organization outsources management of its one or more of its suppliers to an supplier then it is the SAM for CMMI-Acq.  It is another flavour in OMMI - Orhan's Maturity Model Integrated)

         

        BTW, if we want to check if it makes sense in CMMI then we need to know business and technical requirements of CMMI; which I used to call them the spirit of CMMI.  In other word, we need to go back to the purpose and meaning of each maturity level and check if my suggested design fit the requirement/purpose or meaning.  Another result of poor performance of the SEI, they don't habit of implementing CMMI best practices in the development of CMMI; if they would have the habit of following their own best practices they should have a traceability matrix from requirements of CMMI to the design and other details of the CMMI.  

         

        The spirit/purpose/meaning of Maturity Models:

        ML5 - Quantitative continuous improvement

        ML4 - Stability (if a process is not stability, it is waste of resources to invest in quantitative improvement on that process)

        ML3 - Standard definitions (Stability requires quantitative understanding which requires standard definitions)

        ML2 - Internal best practices (standard definitions should be based on working practices that is why project-by-project implementation should be encouraged to identify what is working in the particular org.)

        ML1 - Performing (Org should be able to perform basic functions; e.g. Eng PAs for CMMI-Dev)

         

        Cheers,

        Orhan


        On Mon, Jun 11, 2012 at 1:00 AM, <
        PACT.otoole@...> wrote:
         

         

        Orhan,

         

        OK, so in your view, ML2 should achieve CL2 in the process areas that are unique to the given constellation (you really should avoid using terms like “specific PAs” as “specific” already has a pre-defined meaning in the model and it confuses newer members of our community – and some of us older members as well!)

         

        Since you didn’t say anything about the core process areas, I assume that they should stay where they are currently placed (alternatively, would you prefer that all core process areas should be deferred to Orhan’s ML3 or above?.  If the latter, I’m really not sure what you think should happen with SAM since it is common to the DEV and SVC constellations, but does not appear in the ACQ constellation).  

         

        For now, let me assume that the current ML2 process areas are also in Orhan’s ML2.  I assume, too, that you prefer to leave the ML4 and ML5 process areas exactly where they are, leaving

         

        Orhan’s ML2
        :

        REQM at CL2

        PP at CL2

        PMC at CL2

        SAM at CL2

        MA at CL2

        PPQA at CL2

        CM at CL2

        RD at CL2    

        TS at CL2

        PI at CLs

        VER at CL2

        VAL at CL2

         

        Orhan’s ML3
        :

        All process areas in Orhan’s ML2 at CL3 plus:

        OPF at CL3

        OPD at CL3

        OT at CL3

        IPM at CL3

        RSKM at CL3

        DAR at CL3

         

        Orhan’s ML4
        :
        All process areas in Orhan’s ML3 at CL3 plus:

        OPP at CL3

        QPM at CL3

         

        Orhan’s ML5
        :

        All process areas in Orhan’s ML4 at CL3 plus:

        OPM at CL3

        CAR at CL3.

         

        Did I get this right?  

         

        Pat

         

         

         

        From:
        Orhan KALAYCI
        Sent:
        Wednesday, June 06, 2012 11:00 PM

        To:
        cmmi_process_improvement@yahoogroups.com
        Subject:
        Re: [CMMi Process Improvement] Clarifications - VER, VAL and M&A

         

         

        Pat,

         
        Specific PAs are the PAs specific to a CMMI constellation such as Eng PAs for CMMI-Dev.

         

        They play an important role in my definition of MLs similar to role of specific goals in CLs.  

         

        My point is that the new home of CMMI should stop suffering of those organizations who have been invested in their MLs but did not realized the ROI they were supposed to realize because of the ill defined structure of staged presentation, i.e. specific PAs are delayed to ML3 which is too late an org to start working on them, there should be internal best practices should be developed in ML2 for specific PAs before starting to define org standards for them on ML3.  I hope the new home of CMMI will redefine the staged representation to bring the spirit (chaos -> performing -> **internal best practices** -> org standards -> stability -> continuous improvement)  to maturity levels.
         

        Cheers,

        Orhan

        Toronto


        On Tue, Jun 5, 2012 at 2:38 PM, <
        PACT.otoole@...> wrote:
         

         

        Orhan,

         

        First, where on earth (or elsewhere) did you get the crazy idea that “the SEI was not able to adapt to change and THAT is why they decided to establish a new home for the CMMI?”

         

        Second, who is this mysterious “they” you refer to in the preceding quote? Who, precisely, do you think made the decision to establish a new home for the CMMI?
         

        Third, why should the SEI listen you “your suggestions to lead blah blah blah” when they have already given you EVERYTHING YOU NEED?? All you have to do is convince your client to use the continuous representation rather than the stage representation and you can guide them in EXACTLY the direction you have indicated. Don’t make the rest of the industry change just because you are incapable of successfully altering your clients’ position regarding the staged vs. continuous representations.
         

        When I thought one of my clients would be better served to adopt the continuous representation rather than the staged representation (which, by the way, is the vast majority of cases), I present both options, the likely outcomes, and my professional recommendation. 90% of the time they agree and switch. It’s not all that hard if you simply use the tools that are already at your disposal and focus on helping one client at a time rather than trying to convert the whole industry to a new paradigm. Do you REALLY think you are going to “save the CMMI” by getting everyone to admit that approach taken by the SEI, and successfully adopted by many organizations over the past 25 years is completely wrong and that you are the only person in the universe that “really gets it?”

         

        One final question – what is the shortcoming in the current continuous representation that prevents you from implementing your approach to help your clients?  I’ll tell you what – you tell me precisely what your view of “Orhan’s Maturity Level 2” looks like and I’ll tell you how to get that PRECISE result using the continuous representation, OK?

         

        But you’ll have to be a bit more precise than what you stated in your previous post: “
        ML2 - All specific PAs are MANAGED,” because the term “specific PAs” is meaningless.  There are specific practices and specific goals, but there is no such thing as “specific PAs.”  So what did you really mean?  In your unique view of the world, specify precisely what you would like to see as “ML2.”
         

        There is no need to go into your discussion about the “spirit of the CMMI” – just tell me what the ML2 specification should be.

         

        Regards,

         

        Pat

         

         

         

        From:
        Orhan KALAYCI
        Sent:
        Tuesday, June 05, 2012 11:58 AM

        To:
        cmmi_process_improvement@yahoogroups.com
        Subject:
        Re: [CMMi Process Improvement] Clarifications - VER, VAL and M&A

         

         

        Hi Pat,

         
        The more I speak the more I misunderstood.  So I will keep it really short.

         

        You are right the SEI was not able to adapt to change that is why they decided to establish a new home for CMMI.  If the new home of the CMMI fails to adapt to changes then the whole purpose of moving the CMMI into new home will fail.  The new home of the CMMI should lead the change!  

         

        The new home of the CMMI better listens to my and others' suggestions to lead the SW development community which includes clients of sw, systems, or service developers, operators, and systems integrators.    
         

        The new home of the CMMI has a big power that they can use it to shape the community.  The power is that organizations need to maintain their official maturity levels. Using this power the new home of the CMMI can shape the community to increase the performance of organizations with MLs by just redefining the requirements of getting and maintaining an ML.  The new home of the CMMI should not be just new building and new offices; it requires new people who are not afraid of leading the market.  

         

        Cheers,

        Orhan


        On Fri, Jun 1, 2012 at 7:18 AM, <
        PACT.otoole@...> wrote:
         

         

        Orhan,

         

        You are correct that it is not easy to convince a client to convert from the staged representation to the continuous representation – it is hard (1) to convince them and (2) to work them through the change process.  So I’m a bit surprised that you think that SEI is simply “failing to do the right thing” by not adopting your rather radical changes to the staged representation – especially when a client could SO EASILY adopt your approach by merely switching to the continuous representation!  Go figure.

         

        I know that you prefer the staged representation, and that you think that your philosophy is better than everyone else’s.  But guess what – EVERYONE believes that they know better than anyone else as well!  It’s a funny thing about professionals – we don’t always see eye to eye (especially those of us that you say are blind!)

         

        I would simply suggest, as I have in the past, that you take a second look at the continuous representation.  As I suggested in my last email, you can quite easily formulate an alternative to the staged representation by cobbling together whichever process areas at whichever capability level you feel is optimal, and use that to assist your clients to achieve the best results.  I do this in my consulting work every day.  

         

        However, when I take this approach, rather than trying to come up with a one-size-fits-all solution as you propose, I am a bit more flexible in that I work with the client to understand their current state and then help them plot an optimal way forward to achieve their future-intended state – but that’s what MY experience tells me works best.  I have no problem working within the constraints of the current continuous representation to achieve that end.  Nor would I try to convince the SEI that they should change the staged representation to align with my preferred way of helping clients.

         

        Rather than pissing and moaning about the ineptitude of the SEI, try changing your OWN approach and mold the EXISTING continuous representation to best meet your client’s needs.

         

        Yes it is hard to get a client to convert to the continuous representation, but can you imagine how hard it would be to convert the entire industry to a radically different staged representation?  And yet you continuously wonder why the SEI isn’t aggressively adopting your ideas!  

         

        I know that you are well intentioned, I just don’t think you are going about this the right way.  You can’t convince many of the people in this forum that your ideas are worth the trauma they would cause in radically changing the CMMI, and you wonder why you can’t change the minds of the people responsible for the model itself!?!  Perhaps it’s time for YOU to try a different approach.. The CMMI’s continuous representation gives you almost everything you need to do what you want to do.  Get out of denial and quit trying to force the entire CMMI community to encounter trauma because you don’t want to use the continuous representation.

         

        Regards,

         

        Pat

         

        From:
        Orhan KALAYCI
        Sent:
        Monday, May 28, 2012 11:32 AM

        To:
        cmmi_process_improvement@yahoogroups.com
        Subject:
        Re: [CMMi Process Improvement] Clarifications - VER, VAL and M&A

         

         

        Pat,

         
        I am not interested in how EASY or DIFFICULT it is, I am interested in what is RIGHT! :)  (Also you will see it is not EASY to convince a sponsor targeting for ML2 to *invest* in VER and VAL since currently they are not explicitly required for ML2)
         

        Fact:  Number of organizations who prefer staged representation (ML) is HUGE.

         

        What you are saying is like saying;  it is EASY; all they need to do is to switch to continuous presentations.  I don't think you are so naive to assume it is easy for an organizations used to work in staged representation to switch to continuous representation and even it is not easy for so many organizations who is just about to start CMMI journey  because their business need does not allow them to do  so; they simply can not go with continuous representation because, in the eye of the sponsor, it is an unnecessary *investment*; the sponsor will say if it is not required to get the level then I don`t have money to waste for something unnecessary.  

         

        So, it is another result of poor performance of SEI and unfortunately it is so widely used for decades even a senior expert of CMMI like Pat yourself can not see the problem.  

         

        We don't need to look too far for an evidence, it is right here.  You answered a question about if they need to implement VER and VAL for ML2 and you said NO!  It is correct according the current understanding of SEI about CMMI MLs.  And you continued to say, if you really want to improve your processes go with VER and VAL even if they are not required for ML2 but if you only want to get the level you don't need them.  What does it mean, current definition of ML2 has serious weaknesses to help organizations improve their processes.
         

        Here is a question for you:  What percentage of organizations targeting for ML2 will try to implement VER and VAL (at CL2) after your answer?  This is one of the reasons why organizations with official levels can not perform well.  Because the current definition of MLs does not allow/help them to perform.  It does not allow them because it is kind of encouraging skipping levels; namely ML2 and ML4; these two MLs generally skipped because even the SEI and most of CMMI expert can not explain why they are important for the next level.  Apparently Pat, you don't care for MLs but as I said there are so many organizations live for and in MLs; that is why it is important the SEI (or the new home CMMI in CMU) needs to redefine MLs so that organizations targetting ML2 and ML4 can be guided correctly and can really improve their processes while they are *investing* their time and money to achieve a level.  I am kind of hearing what you say, "who will stop them implementing VER and VAL even though they are not required for ML2", let me answer for you :  it is THE SPONSOR who will stop them investing for VER and VAL (i.e. eng PAs) for ML2.  Most sponsors, especially in these recession times, try to minimize their investment to reach their targets; if there is something explicitly not required they (the sponsors) will not allow their staff to work it.

         

        Hope this help you and the new SEI understands the importance of revisiting the definition of MLs, (nothing personal it is just a fact - highlighting the ignorance for the need for the re-defined MLs)

         

        Cheers,

        Orhan

        Toronto

         

         



        On Thu, May 17, 2012 at 10:29 AM, <
        PACT.otoole@...> wrote:
         

         

        Orhan,

         

        Your approach is EASILY achieved if you migrate from the staged representation as it exists today to the continuous representation as it exists today.  The beauty of the continuous representation is you can “rack and stack” the process areas in any order you deem appropriate WITHOUT needing to have the SEI change anything.

         

        So if I simply modify your use of the term “maturity level” and change it to “Orhan’s Desired State” (ODS) we’d have:

         

        To make sense, ODSs should be defined as;

        ODS 0 - there is no specific PA in place; e.g. eng PAs for CMMI-Dev are not in place

        ODS 1 - all specific PAs are in place; e.g. eng PAs for CMMI-Dev are in place

        ODS 2 - All specific PAs are MANAGED.

        ODS 3 - All specific PAs are DEFINED.

         

        Voila!  I have solved your problem without involving the SEI!  Go forth and spread the word that this is how you perceive your clients would be best served to implement the model.

         

        Long live the continuous representation!

         

        Pat

         

         

         

        From:
        Orhan KALAYCI
        Sent:
        Tuesday, May 15, 2012 7:01 PM

        To:
        cmmi_process_improvement@yahoogroups.com
        Subject:
        Re: [CMMi Process Improvement] Clarifications - VER, VAL and M&A

         

         
        Hi Folks,

         

        Please allow me to explain what they don't teach in SEI about Maturity Levels.  

         

        To understand the spirit of maturity levels, we need to see the similarities between maturity (ML) and capability levels (CL).  

         

        You know specific goals of a PA play an important role in its CL.  In CL2, specific goals of PA are successfully MANAGED, in CL3 specific goals of PA are successfully DEFINED.  

         

        I strongly believe, to make sense, we need to establish a new structure for ML similar to CL.  First, I need to introduce you with the concept of specific PA and generic PA.  Generic PAs are the PAs applicable to all CMMI constellations; CMMI-Dev, CMMI-Srv, CMMI-Acq.  Specific PAs are the PAs applicable to only one of the CMMIs; e.g. Eng PAs for CMMI-Dev.  

         

        To make sense, MLs should be defined as;

        ML0 - there is no specific PA in place; e.g. eng PAs for CMMI-Dev are not in place

        ML1 - all specific PAs are in place; e.g. eng PAs for CMMI-Dev are in place

        ML2 - All specific PAs are MANAGED.

        ML3 - All specific PAs are DEFINED.

         

        So, in my definition of ML (which SEI is not there yet),  VER and VAL should be there in a project-by-project sense in ML2.  It is a necessary phase (ML2) in the journey of maturity which should not be skipped because thanks to these project-by-project implementation, internal best practices for eng PAs of CMMI-Dev or specific PAs of other CMMIs will be identified to establish the necessary foundation for ML3.  So org standards at ML3 will be able to base on internal best practices, tested and approved, in ML2 in a project-by-project manner.
         

        Hope, SEI will soon improve its performance and redefine the MLs to bring the spirit of maturity levels back so that they (SEI) will also be able to define the meaning of skipping a level and furthermore they can start encouraging organizations not to skip levels.  

         

        Cheers,

        Orhan

        Toronto

         

         

        On Mon, May 7, 2012 at 8:41 PM, <
        PACT.otoole@...> wrote:
         

         

        Hamsa,

         

        1.  If your primary objective is to achieve ML2, then ignore VER and VAL and leave them totally ad hoc as they will not be evaluated as part of the ML2 SCAMPI A staged-representation appraisal anyway, so who cares?

         

        On the other hand, if your primary objective is to enhance the performance of your projects and to achieve the business imperative (whether that is quick to market, low cost provider, highest quality producer, “greenest” organization in your market segment, whatever), then you may find that focusing attention on VER and/or VAL will help you immensely in that regard.  If that’s the case, then why WOULDN’T you focus some attention in such a high-payback area?

         

        Don’t “do the CMMI” so you can be appraised and get a maturity level; rather, EXPLOIT the CMMI to improve your projects’ performance and use appraisals to generate even more good ideas as to how work performance could be enhanced – oh yeah, and to VALIDATE your progress against model expectations.

         

         

        2.  Some of this depends on how you implemented Measurement and Analysis (MA) at ML2.  Some organizations take a project-by-project approach in which case each project defines its key metrics (size, effort, cost, defects, etc.).  Other organizations look to define metrics at the organizational level such that each project is using the same definition for <defect>, for example.  There are advantages and disadvantages to both approaches, so neither is inherently right or wrong.

         

        If you took the project-by-project approach, you’ll probably converge similar metrics terms and definitions as you evolve toward ML3.  This will happen relatively naturally as you start employing organizationally consistent processes, procedures, templates, forms, etc.  As work gets performed more consistently, it only makes sense to use more consistent metrics.

         

        You may also find that your mindset toward MA SG1 changes – especially MA SP1.2 – SP1.4.  At ML2, you spend a fair amount of time simply “backfilling metrics documentation” for existing metrics.  You’ve been measuring this stuff forever, but the model says you should specify the metrics, how they are captured and stored, and how they will be analyzed and reported – OK, so let’s just get that out of the way so we can move on to other things.

         

        However, at ML3, when new information needs arise, you will find yourself more naturally migrating toward these practices.  Big Boss wants us to be more productive, eh?  I wonder how we’d measure the productivity of our developers, testers, project managers, middle managers, etc.?  How are we going to combine these various measures to determine if the organization is meeting the challenge of increasing overall productivity?  How should we report this to show progress over time?

         

        At ML4, you’ll start getting very focused on things like potential sources of measurement error and the application of proper analytical techniques.  You’ll also figure out that some variation in the data is expected and that such variation really provides more noise than signal, so you won’t over-react like you did at ML2 and ML3.  Actually, you’ll wind up sleeping a bit better at night as well!

         

        Anyway, just a few random thoughts – hope they help!

         

        Pat

         

         

        From:
        energyminds
        Sent:
        Monday, May 07, 2012 11:40 AM

        To:
        cmmi_process_improvement@yahoogroups.com
        Subject:
        [CMMi Process Improvement] Clarifications - VER, VAL and M&A

         

         

        Hello,

        1)Suppose an Orgn. wants to get assessed at L2, how should VER and VAL be met or looked into? Can this be completely ignored and processes/practices carried out as Verification and Validation activities be left as they are (adhoc?). Can anyone please clarify.

        2) How is M&A implementation different in a L3 and L4?

        Hamsa

         
         

         

         

         



        _____________________________________________________________________________________________________________________

        ESG Elektroniksystem- und Logistik-GmbH
        Rechtsform/Legal Form: Gesellschaft mit beschränkter Haftung
        Sitz/Registered Office: München
        Handelsregister/Commercial Register: Registergericht München, HRB 8130
        Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: Mario Paoli
        Vorsitzender der Geschäftsführung/CEO: Dipl.-Math. Gerhard Schempp, Geschäftsführer/Managing Director: Dipl.-Kfm. Götz Graichen



        _____________________________________________________________________________________________________________________

        ESG Elektroniksystem- und Logistik-GmbH
        Rechtsform/Legal Form: Gesellschaft mit beschränkter Haftung
        Sitz/Registered Office: München
        Handelsregister/Commercial Register: Registergericht München, HRB 8130
        Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: Mario Paoli
        Vorsitzender der Geschäftsführung/CEO: Dipl.-Math. Gerhard Schempp, Geschäftsführer/Managing Director: Dipl.-Kfm. Götz Graichen

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