|
|
|
David Rutherford, Treasurer BCS Leicester Branch writes:
It was a pleasure to attend a talk given by such an enthusiastic speaker. Having
some knowledge of software CMM (a precursor to CMMI) I was not looking forward
to this talk as much as others in the series. The speaker surprised me by putting the subject
in context and making it so accessible. His comments carry weight as he
is an experienced practitioner in software development in the same industry as
myself. A highly stimulating experience.
I've smartened up some of my ad-hoc, rough notes (included below) to
highlight examples, asides and main points that seemed relevant to myself. Some may not be contained in the presentation slides.
- The Thales experience is that software size multiplies by 5-10 times every 5 years, but the number of
staff involved stays the same. So productivity is key.
- Systems produced by Thales are software dominant and in 2001 reached 3500K
SLOC
(in comparison one estimate from the web gives Windows XP (in 2002) as being
40M SLOC)
- Agile Methods were a new term to me, regarding further investigation. The
umbrella term elicits examination of any method which can address
productivity. The speakers CMMI was fitted into this Agile Methods concept
which sounded just as important as CMMI itself..
- Agile Methods: Methodologies that work. Acceptable to engineers. Reduce risk. Deliver value as early as possible.
- Manifesto for Agile Software Development. (from http://agilemanifesto.org/)
We are uncovering better ways of developing software by doing it and helping others do it.
Through this work we have come to value:
- Individuals and interactions over processes and tools
- Working software over comprehensive documentation
- Customer collaboration over contract negotiation
- Responding to change over following a plan
- That is, while there is value in the items on the right, we value the items on the left more.
- CMMI does not replace other standards, but working to it means that it is
a relatively simple task to show that other standards are met.
- CMMI levels:
- 1 - Perform (not addressed by CMMI, hero use!)
- 2 - Managed (each project)
- 3 - Defined (each project follow organisational guidelines)
- 4 - Quantitatively Managed (Process driven by metrics)
- 5 - Optimising (whole organisation continuously involved in improvement).
- (As CMM) Key Process Areas broken down into Specific Goals and Generic
Goals. Specific Goals implemented by Specific Practices. Generic Goals
consist of Common Features each with Commitments to Perform implemented by
Generic Processes.
- It is expected that the implementation of CMMI be tailored for projects
and organisations.
- CMMI via use of SCAMPI ensures stated processes are being applied by
requiring compliance evidence (e.g. 125 types at Level 2) including direct
evidence, indirect evidence and affirmation (by interviews).
- SCAMPI can be applied to organisations, locations and projects at various
classes (A (1 week for 8-10 people ~£120K), B, C), with passing class A
meaning you are listed on the global website.
- Organisations who do not address CMMI may start finding it hard to hire
project managers.
- Knowing how CMM was structured helps understand CMMI.
A lively Questions and Answer session further brought home the subject by
discussing issues more specific to the people that attended, such as:
- Regularity of SCAMPI
- Applicability to business or organisation size (project size, SME's)
- How widespread is adoption
- Replacing use of other standards
- Implementation timescales
- Tailoring guidelines (remember to look up using only parts of CMMI)
- Commercial return
- CMMI forcing the use of any metrics collected
These points from my rough notes do not do justice to the presentation and
the speaker, but act as a personal review, indicating areas that could be looked
at in the future as part of both CPD and company related activities.
|
