« November 2007 | Main | January 2008 »
December 20, 2007
Now Shipping! The Cutter/QSM Benchmark Almanac
Great news! The Cutter Consortium is now shipping the Cutter/QSM Benchmark Almanac: Application Development Series, 2007-2008 Edition, the result of an unprecedented joint collaboration featuring industry research by QSM on software development (derived on projects mined from the QSM worldwide database), and expert opinions by Cutter authors like Tom DeMarco, Tim Lister, Jim Highsmith, Jim Love, E.M. Bennetan and... yours truly. For me, it's personally gratifying to have been the editor of this report since I also wear two hats: being a managing partner of QSM Associates, as well as serving with Cutter as the director of the Benchmarking Practice.
Read the Overview here, which also contains ordering information.
Here's an excerpt of Larry Putnam's preface to the report. As many of you know, Larry is considered by many to be the father of software measurement. He's written (along with Ware Myers) five books on the subject, and countless papers going back over two decades. I'm deeply privileged to have worked with Larry over the years, and to have been part of this project. We hope you find this report extremely valuable to your organization.
===============================
From Larry Putnam Sr., President, Quantitative Software Management:, Inc.
When I first started in the software measurement and estimating business more than 30 years ago I was frequently asked intriguing questions like, “What are the most productive languages to write my code in?” “How much productivity gain do I get from using a high level language v. assembly language?” “How much economic benefit does a best in class software development company achieve compared to a worst in class company?” “Is there a trade-off between cost and schedule? How does that impact quality?”
There were many more similar questions. I didn’t know the answers. Why? There just wasn’t any reliable data. And to answer such questions one needs to have a good body of carefully measured and recorded software data about completed projects in a wide range of different application types and development environments.
So, not having much data was a great incentive to start collecting some with the ultimate goal of being to answer many of the questions that people in the industry raised. Along with my colleagues I started collecting it. The early collection efforts with a few scores of systems were enough to develop good estimating algorithms that have stood the test of time. A few hundred systems collected and analyzed provided a baseline for benchmarking an organization v. peers in its industry. A little more data and it became possible to establish industry trend patterns and productivity determinations that could be quantified in dollars saved, months of schedule reduction and improved reliability of the product delivered to the customer.
Trends started to become evident. We started to be able to answer a lot of the questions by referring to the real data history. Today we have a very extensive data base of completed projects from which to analyze and draw inferences and conclusions.
The almanac you are about to get into was our first attempt to put together a number of analytical studies and present our findings about what the data were trying to tell us about software development and how the results might be used to plan and manage better.
The first part of the book is devoted to answering some contemporary questions about software projects that can be dealt with quantitatively – with numbers from the data. I won’t try to describe them here. You will enjoy it more if you dig into it yourself.
The second part of the book is a collection of excellent papers by some outstanding authors that support and elaborate on the analytical work presented in the first part. The papers have been published by the Cutter Consortium. The topic areas are: Software Estimation; Managing with Metrics; Risk and Quality as it relates to testing; Applying Metrics to Outsourcing, Extreme Programming and Agile Environments. The authors who have done this work should be well known to you – people like Tom DeMarco, Tim Lister, Jim Highsmith, Jim Love, E.M. Bennetan and Michael Mah. They have an immense amount of good, practical experience gained over many years working on complex projects on the forefront of the state of the art.
There is a lot here: A great body of experience to draw from. You should enjoy it. More importantly, you should benefit a lot. I hope you do.
Lawrence H. Putnam
Posted by Mike at 10:50 AM | Comments (0)
December 12, 2007
I'm So Excited...
... and I just can't hide it.
I just spent a day with our internal QSM team in McLean VA where we brainstormed about our HOT new SLIM 7.0 Release, which we're gearing up for a market launch next quarter. Building upon Larry Putnam's pioneering research on software lifecycles, we've got a new architecture that is completely adaptable to measuring, estimating, and planning Agile development projects for the software industry.
It's no surprise many organizations today are moving toward agile methods. Two burning questions for many organizations are 1) how to prove that these methods really work and 2) how to reliably estimate and plan software projects in a dynamically changing world when requirements are in constant flux. Many of our SLIM clients are facing issues like these.
So that's why I have this Pointer Sisters song ringing in my head. For those of you who might be reading this blog and are NOT in the software or technology industry, you've got to be thinking that Michael Mah has got to get a life. (I admit that it sounds a little odd that this subject gets me so worked up...)
Imagine being really psyched about something that sounds so technologically fringe and esoteric... Well, the reason I've got the Pointer Sisters in my head is because this is NOT fringe. I'm convinced that Fortune 500 companies today are going to realize that this newfangled thing called Agile Methods is going mainstream for many companies on the leading edge of technology. This is true whether you make software for financial services, medical, engineering, avionics, or embedded systems. As more companies race to build software that makes the smarts of the global brain, they're going to realize that building complex information technology systems - faster and more reliable than they've built them before - is going to be key to growing new companies, transforming existing ones, and creating disruptive technologies that take on economic challenges in a rapidly changing world.
Now, the MANAGERS who oversee the creating of these new applications are going to need to better understand, measure, plan, and manage projects like these under ever tighter deadlines. And to do that better than pencil and paper or rudimentary bottoms-up spreadsheets, they're going to need easy-to-use-planning and estimating tools like QSM's SLIM Model.
We've taken everything that we know about making a computer help us do this (to take the load off our already overtaxed and multi-tasked minds), and to do it better than ever before. In the next two days our QSM team brainstorms more about enabling our clients to do this, and I'll report about it here in the next few weeks.
In the meantime, we're building easy-to-use templates for people who want to jump right into running model simulations of agile, RUP, package implementation (i.e. SAP, Oracle), or traditional waterfall projects. Some of these are already done and ready for online Webex demos. Are any of you interested in seeing these? If so, contact Sean Callaghan by email and he can arrange an appointment for you over the web.
Stay tuned for more...
Posted by Mike at 01:54 AM | Comments (0)
December 04, 2007
It's Snowing in the Northeast....
... so I'm heading to Orlando FL for the Software Quality Engineering's Agile Development Practices Conference. Fun fun... Tomorrow (Wed 12/5) QSM Associates/Cutter and Rally Software will be presenting the following:
Case Study – Benchmarking Agile Productivity: What an Independent Assessment Revealed at BMC Software.
By Michael Mah, Senior Consultant, Cutter Measurement & Benchmarking Practice & Managing Partner, QSM Associates and Zach Nies, VP of Product Strategy, Rally Software
1. How data was captured on Agile releases.
2. What the benchmark findings revealed versus a worldwide database of 7,000+ projects.
3. How this was automated using QSM SLIM models to make measurement easy with Rally as the source of record.
What's going to be exciting about this session are these three segments. But what will likely be the most interesting part will be a "Charlie Rose" style talk show interview between me (as Charlie :) and Mike Lunt of BMC Software (as himself). I bet there's going to be lots of lively interaction with the audience during the "show."
If you're going to be in Orlando at this conference, come and join us!
Posted by Mike at 04:01 PM | Comments (0)