The jargon file

For someone who had first come across entries from The Jargon File via the fortune program on Unix, it was a pleasant surprise to find it on the web (once the WWW came into existance). This is what I consider original geekspeak, and as it states on the Jargon File pages, it has been published as the New Hacker’s Dictionary, interpreted with the “original” meaning of hacker

P S This is nothing specifically to do with telecom.

Back to basics: The mythical man-month

A classic written by Frederick P Brooks offers probably the most read and least followed principles for managing software engineering projects.  The book was published in 1975 and held enough relevancy to be republished with added material in 1995.  The author was a project manager in IBM running systems engineering projects for their operating system OS/360 and has put forth startling common-sense principles backed by mathematical theory.

The most famous quote that has been dubbed as Brooks’ law states:

“Adding manpower to a late software project makes it later”

Each chapter of the book deals with specific project management areas like identification of large and complex systems and their dependencies/relationships; scheduling and estimation & expecting the unexpected; sub-dividing tasks and different role requirements; following architecture concepts; safeguards against communication breakdown between sub-teams; importance of prototyping etc.

This book is a must read for anyone working in any pivotal role in software development, not just in the telecom area.  I read the book for the first time even before I had embarked on project management and found the concepts ringing true.

Wikipedia is your friend at The Mythical Man-Month
The Mythical Man Month by Fred Brooks, published by Addison Wesley; ISBN 0-201-00650-2 (1975 ed.), 0-201-83595-9 (1995 ed.)

Remembering first principles

As I am a computer science student who meandered into telecom as an application area (more due to circumstance than intent), I plan to refresh my basic principles once in a while. One of the seminal articles on sound principles for system programming has been published by ACM as far back as 1969. I think that systems engineers should re-read this article every few years to remind themselves of what should be vs what is. As I re-read it currently, the applicability to telecom systems – whether they be networks, services, OSS/BSS systems – clearly surfaces.

In brief, the article lists the fundamental principles of good systems programming (includes aspects like forgivingness and monitorability in addition to the well-known ones like modularity and availability) and defines all of them. It then goes on to discuss the problems that come in the way of adhering to the principles followed by cost metrics for systems. I think the cost metrics concept is something very important for software deployment, more so for telecom systems deployment. The article then discusses higher level implementation languages (remember that this was written in 1969 when handcoding in machine language was the order of the day. The parallel to modern day systems would be usage of meta-systems like application middleware or service delivery platforms in the telco world).

The article includes some good external quotes as well as creates its own quote-worthy statements like:
There is no substitute for intelligent forethought and coordinated planning using the principles of good engineering and development management. (T. J. Watson)

We don’t read. Very few system developers are familiar with work done outside of their own project.

PS Thanks to the NANOG poster who quoted from this and reminded me to re-read this gem.