Facebook Twitter Linkedin

Monthly Posts: July

An Interesting Speech on Software Aquisitions

I've just read the speach of Paul D. Nielsen, Director and CEO of the Carnegie Mellon Software Engineering Institute before U.S. House of Representatives.
Very interesting for everyone in the software industry, including the 10 reasons for software aquisitions failures:

1. Technology key to program success is new to the organization
2. Software issues are considered too late in the system-development process
3. Inadequate planning and estimating
4. Size matters - large projects get into trouble more frequently than smaller ones, all projects grow larger over time
5. Software objectives are not fully understood or specified; they change frequently (and grow) during the project
6. Inadequate experiences and trained project management
7. Inadequate process emphasis and erosion of process discipline
8. Inadequate contract incentives to encourage use of proven software engineering practices
9. Acquirers and developers lack experience working as a team or the resources to do so
10. Insufficient senior staff and inexperienced software engineering cadre

In the last 10 years I have found these problems in speaches and presentations, every now and then.
Statistics showed a real improvement in companies that adopted solid models and methodologies. A lot more remain to be done.

I quote what I've enjoyed most in the speach:
The "unlimited" complexity of software is neither well understood nor well appreciated by many. Software is not rooted in the physical world like other engineering disciplines--civil, aeronautical, electrical, or mechanical engineering. Without physical constraints, the design space is so vast for large programs that you need strong architectural principles, disciplined processes, and talented people to be successful. The larger the program, the more important this is-and, as you are aware, many defense programs are very large.
Additionally, software is invisible; people don't buy code - they acquire systems that satisfy requirements. And software is intangible-you can't touch it or kick its tires. Nevertheless, in most defense systems, software is critical to the very success of the program. The systems just don't work without software.

I remember one "funny" experience we had recently in dealing with one of our Governement Agencies.
We in Romania have inherited a system in which the company has to paint/ stamp an inventory number on each object it owns. So far so good, it doesn't hurt.

But when that Government Agency requested that we have to paint those inventory numbers on our software, we gasped, and laughed, and choked and got nervous and exploded.
We've had this stupid request for years and years in several state owned institutions. Usually we dealed with it by presenting to the bureaucrats a CD or diskette or a licence sheet with an inventory number painted on it. Here Ladies and Gentlemen is your software !!!

This time we debated if we should refuse to obey. Enough is enough. This aberation, a favorite of old bureaucrats, has to stop. Dear all, software is invisible, intangible, is not an inventory object. Give us a break.

Finally, we gave up. Fighting bureaucracy was too much time consuming. We made however a technological progress. At the agency request, we put a label on a computer.

Ladies and Gentlemen, beware, BIS software is inside!

People CMM and Small Organizations

I fully agree with this quote. People CMM has the best practices needed by an organization to attract, develop, motivate, organize, and retain the workforce.
If you are a manager in charge of a whole organization or a team or a HRM Department, People CMM is the model to follow.
People CMM version 2 has just been released.
The People CMM model helps organizations to establish a culture of professional excellence which is the best remedy for brain drain.

Today, People CMM has been implemented in various types of industries (Banking / Financial Services, Governement, Insurance, Utilities, IT of course etc.)

The People CMM model must be interpreted flexibly especially when applying it to smaller organizations so that bureaucratic activities are minimized.
Pay attention to the fact that "size" for organizations in Romania (and other small countries) is not the same as in North America. What in Romania we say "large" in USA is "medium/ small". Only telecom organizations, the first 3 banks, several international companies might be considered "large".
From the point of view of People CMM Model, most of the organizations in Romania should be considered small/ medium size and as such, the model must be carefully interpreted.

We've had enough bureaucracy so far, no need to increase the forrests devastation.

I will add this: a collaborative platform, simple to use and affordable yet with a powerful functionality can make the difference between a successful implementation of the People CMM model and a new nightmare in a small/ medium size organization.

As I'm from BIS and I have been using successfully the collaborative software BIS Esfera Suite on a daily basis for more than 7 years so far, I can say Esfera is an interresting option. We are now in the course of migrating the product on Microsoft Sharepoint technology and People CMM proved to be extremely useful in improving the product requirements.

Emilia Dragne
BIS SPI Project Manager
Intertek ISO 9001:2008CMMI Level 2ISO  9001/2008 Dun & Bradstreet