Monday Dec 22, 2008

Conversation with Lenz Grimmer

You can read my conversation with Lenz Grimmer or look at other interviews conducted by the MySQL community team. 

Friday Dec 12, 2008

Janice Heiss' interview with me

Starting from a series of exchanges conducted mostly by e-mail and often across multiple time zones as I was traveling in Germany and Russia in the last couple of months, Janice Heiss cajoled me (and I should thank her for it) into this interview.

I hope you'll learn a few things reading it. I certainly learned quite a bit as I was exchanging these ideas with Janice and as I was trying to reply to some of her questions. For example, although I had always been curious about it, I hadn't earlier thought much about how I may respond to Bill Joy's famous essay until Janice actually asked me about it during the course of the interview. (Thank you very much Janice!)

I should probably add that Janice is a Sun staff writer as well as a blogger on Java.Net. She is also the person behind a wonderful series of other interviews with Sun's developers and software engineers—lots of amazing work and ideas are summarized in these interviews: "Meet the Engineer". Finally, I also recommend a reading of her tips for students coming from some of these top developers. There, you are bound to fin (as I did) many nuggests of wisdom.

Is Project Management Dead?

The PMBOK book comes to you courtesy of Project Management Institute.
It is considered a standard for project management.
Chapters 1 to 3 are "must" reads. The remaining chapters are further, very useful elaborations of the material in these earlier chapters.
When you read chapters 1 to 3, think of what it would mean to apply the concepts in some project you're facing: Perhaps, you're organizing a large conference, a wedding, or the construction of the next space shuttle.
See which concepts are applicable where.
I used the book, along with cases form the real world, to teach a semester-long graduate course in project management at NPU last summer.

Far from it.

Projects are about unique objectives attained within defined duration.

They are inherently different from operational work.

By the very nature of how we operate as human beings, any cooperative activity involving more than a two or three interactions per person contains within it the seeds of error, missteps and failures. (This may have to do with the common size of family units in some of our societies.)

The whole practice of project management involves instituting processes that meet in anticipation of these errors and failures, handle and check them when they occur and make the necessary adjustments in order to digest the uncertainties that future brings.

If future could be perfectly predicted, there would be no need for project management. If groups could cooperate with a guarantee that no failure or shortcomings would occur on the way to the objective, there would be no need for project management.

Sunday Jun 08, 2008

Subtle Significance of Job Satisfaction

I quote the following passage from the conclusion to Dennis W. Organ's paper ("The subtle significance of job satisfaction," Clinical Laboratory Management Review, (Jan/Feb 1990) 4, no.1, 94-98):

Management research and theory have taken a long time and a torturous path in catching up with the insights of Chester Barnard. More than half a century ago, Barnard noted the essential condition of the "willingness of persons to contribute efforts to the cooperative system." This quality of willingness "is something different from effectiveness, ability, or value of personal contributions...[it] means self abnegation." Willingness is characterized by "[an] indefinitely large range of variations in its intensity among individuals" and, within individuals, "it cannot be constant in degree." Finally, this "willingness to cooperate, positive or negative, is the expression of net satisfactions and dissatisfactions experienced or anticipated."

Barnard underscored the very nature of organizations as cooperative systems. Rules, structures, policies, job descriptions, sanctions, incentives—they all play necessary roles in collaborative endeavors, but as derivatives of, not substitutes for, the underlying disposition to cooperate. Such a disposition can be sustained only by a sense of organization as a microcosm of a just world. Occasional inequities can be tolerated if there is faith that the system works fairly over the long run, with self-correcting tendencies. When faith leads to a narrowly defined, quid pro quo contractual relationship, the disposition to cooperate ebbs. Surveys show that most of the nation's labor force begins work with a fairly high degree of job satisfaction and that most of the people, most of the time, will describe themselves as "all in all, satisfied." There is a generally prevalent inclination to give the employer the benefit of a doubt—"I'll assume you're treating me fairly until you persuade me otherwise." So the disposition is generally present to render a substantial contribution via OCB [Organizational Citizenship Behavior]. A good-faith effort by managers to provide a "square deal" will do much to ensure the quality of OCB.

Thursday Nov 15, 2007

Last Day in Bangalore

This is my last day in Bangalore and I'll be leaving the Sun office here in a few minutes to go to the hotel to collect my luggage and take a cab to the airport. It has been a great trip. We have had several very productive meetings, and I had the good fortunate of meeting some of the great folks at Sun, India Engineering Center, Bangalore. It made me particularly proud to meet folks from databases, documentation, SunMC and Sailfin ... What great teams!

This has been my second visit to Bangalore and I already know I'll be missing the city, friends and colleagues who are based here. I certainly look forward to seeing them again, either here or in other Sun sites or facilities. It has certainly been a great honor to have a chance to come here and to make and renew friendships. 

(If time allows, I will write some more later.) 

Tuesday Jan 09, 2007

Partitioning a Disk

Warning: This entry is the story of partitioning a disk.

I've recently moved offices within Sun and just got a new laptop. With a back-up work system, I figured it was a perfect time to go back to the Gateway desktop I've had in my office for some time and try to install Solaris on it.

As would be expected, we have weekly builds of Solaris here, and right across from my office, I can pick up the latest weekly build on a DVD. This seemed like a good place to start.

As a first step, I wondered if I should partition the hard disk on my Gateway machine which currently runs Windows. I didn't really need the Windows operating system any more. I don't use it for any application that would require it and all applications I run are either Java-based or available on Solaris, and I have used Open Office very successfully since 2003 to deal with MS Office based documents.

Nevertheless, I decided that the partitioning exercise was to be had not so much because I was interested in preserving my Windows files but because I wanted to see how easy it was to perform the task without paying for any software. James Liu had earlier mentioned QtParted tool available on Knoppix, which is a Linux OS possible to run from a CD. I had always wanted to use an open source partitioning facility, and this seemed like a good working choice. The alternative, of course, was just not to partition and install using the Solaris installation DVD.

When I was unable to produce my own working Knoppix CD, James kindly came to the rescue and gave me a working CD of Knoppix 5.1.1. James had burned this CD on Solaris. (The CD I had produced kept relegating me to a useless shell of Knoppix perhaps because I was producing it on a Windows XP system with a freeware CD image burner, probably not adequate for my purposes even at low burn speeds. There are commercial tools for burning CDs from CD images on Windows XP but I didn't want to use any of these.) 

The Knoppix OS on the CD works really well. I was now able to load the OS and then run QtParted to resize the existing partition and "create" new ones, and then run QtParted to "commit" these changes. I used suggestions from Richard Friedman which worked really well.

It turns out that the Ferrari laptop on which Richard installed Solaris Express has a similar size of disk to the Gateway machine in my office. The only difference is that QtParted performed the job of disk partitioning in less than 20 minutes on my Gateway machine which compares very well with the 2 hours in the Ferrari experience. As always, we shouldn't compare apples and oranges. The higher speed for partitioning has to do with the two CPUs and the large RAM available on the Gateway box in my office.

More later ...


Thursday Nov 30, 2006

A City on the Human Scale

I've been staying at Trondheim, Norway, for work-related meetings. Trondheim is not only an attractive university town with a rich history but also an urban area fit for human scale.

Friday Nov 10, 2006

Give Me Sun Ray

Good, timeless ideas keep reincarnating in better ways. 

We talk a lot about mobility and about devices. I have been mobile--moving around quite a lot recently among various Sun campuses and spaces in the San Francisco Bay Area, roaming through offices and conference rooms.

I now have a new office in Sun's Menlo Park campus and what I want more than the laptop that may be on its way (my laptop had a hardware failure some time ago), is a Sun Ray, even in my office. With a Sun Ray, my session is always there, and a card-key away, and because I do not have to carry anything but my cell phone and my corporate card-key, it makes me even more mobile--every pound counts. (The wear and tear on Sun Ray keyborads tell me I'm not alone.)

So, when do I use the laptop? When I go on trips where there is no Sunray, when I'm lying down on a bed or a sofa to work or when I'm trying to build,  test or demo a piece of software in the absense of a Sun Ray. Sun Ray is by far the best equipment for the corporate worker who is not doing any of these latter tasks in environments where Sun Rays are missing--and let's remember that few corporate workers are engaged in these sorts of tasks on a regular basis.

I can even leave this entry as it is, run to my next meeting and if my party is late, insert my card key in a Sun Ray and do a final edit at this very point, where I am. That typo is now gone .... next one for the next stop ....

Thursday Jun 22, 2006

Managing the Java DB 10.2 Project

Managing the Java DB project has thought me a few lessons about distributed collaboration.
[Read More]

Monday May 01, 2006

Why Schedules Fail

Schedules are about predicting the future in human affairs, and generally speaking, human beings have rarely shown any particular ability in predicting what is to come.[Read More]



« February 2017