On The Margins

Thursday November 15, 2007

[ Work ] 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.) 

2007-11-15 23:53:48.0 -- Comments [4] ; Permalink ; Trackback.
bangalore sun-microsystems

Tuesday January 09, 2007

[ Work ] 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 ...

2007-01-09 18:31:26.0 -- Comments [2] ; Permalink ; Trackback.
files linux os pc solaris systems windows

Friday December 01, 2006

[ Work ] 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.
 

2006-12-01 02:44:33.0 -- ; Permalink ; Trackback.
cities human norway scale trondheim work

Friday November 10, 2006

[ Work ] 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 ....

2006-11-10 13:28:22.0 -- Comments [1] ; Permalink ; Trackback.
client corporate ergonomics mobile sun technology work worker

Friday June 23, 2006

[ Work ] Managing the Java DB 10.2 Project

Managing and leading the Java DB 10.2 project, which involves the coordination of Sun's contributions and involvement in the Apache / Derby 10.2 effort, has taught me some valuable lessons regarding medium size projects composed from large global distribution of resources and effort crossing national as well as firm boundaries.

To begin with, when some 25 highly talented, technical and professional people from Sun work full or part-time in three major time zones (San Francisco Bay Area, Bangalore, Trondheim), the project manager as well as many of the leads and participants have to split and extend some amount of their working hours among all these zones.  (For example, this week while working from my brother's apartment in Wiesbaden, Germany, my own work day has extended most fully, spanning both the Bay Area and Trondheim and a good chunk of the time in Bangalore. When we first released our Java DB package for Mustang, we coordinated qualification, evaluation and anticipation of JDBC 4 API changes across all three geos. The sense, desire and volunterism for collaboration was astounding.)

It is time for a quick game of soccer! (JavaOne 2006, May 19, San Francisco)

The globe remains round, as our engineering director recently noted from Norway, but we literally live and work on a global scale across multiple temporal and spacial zones. What makes the system work is the existence of a critical mass of us who cross these boundaries on a regular basis. For all the work we do to be accomplished, work in one piece of the organization tends to seed, motivate and drive work in other pieces of the organization, and beyond, within the open source sharing community in which we are involved. (It is as if we pull and we're pulled by the larger context within which we pull.) Time and space are not limited to CET, PST, IEC and different geos. Even the configuration of our offices and physical environment varries widely as we move from place to place.

There are always communication challenges and misunderstandings that demand various types of solutions. E-mail, instant messaging, regular conferences and 1-on-1s as well as small, task-oriented one-time meetings bring together our focused attention on the goals of the global project and the larger community. However, these are simply the basic elementaries with which the team begins in order to bridge the gaps, build concensus and create group purpose. Other means involve conferences and meetings (e.g. JavaOne and ApacheCon) which bring us together in closer proximity to intensify the dialogues and create a more effective conversation within the team. Interactions of the less formal and ordered type prove particularly useful. The more random type of engagement, when possible, has tremendous effect in increasing harmonization of the "vibrating team strings" in all dimensions: personalities, tasks, external contacts, mini team and short-lived task force engagements, etc.

I've seen how people invited to a large room, in one of our global meetings or conferences, meander about to engage in conversations with many of their colleagues generating subtle but lasting effects on the project.

When it comes to the wider team of people who collaborate, on a larger scale, with various functions such as management, engineering, marketing, sustaining and support, the project demands a more formal and  broad-based weekly meeting in which the team properly handles and swiftly generates and dispatches necessary action items on the path to successful coordination. The ultimate goal are the deliveries of the project and a quality release.

When this happens, the sense of elation and team accomplishment produces the greatest, most fulfilling rewards for everyone involved, and those who have sat by the wayside either miss the opportunity to be part of it or mistake what is going on for some sort of cult feeling, which it is not. Full engagement in the work of the team and the community becomes necessary in order to understand its work and to appreciate and partake of its accomplishments. It is wonderful and most rewarding to see this when it happens.

The ideal team works like a watch. Conflicts are resolved and lines of communications are clearly set. Perhaps, no place other than a battle field, the proper placement of these lines of communications within the organization are more important than we see in a global project of appreciable size involving more than a couple of dozen people and interacting with even a larger community that goes beyond the members drawn from one's own firm.

The challenges of the open source environment are only matched by the rewards offered by the whole community engaged in veting and examining one's own biases and intentions as well as reaching out to offer help and share knowledge.

Sharing is economically valuable because of the "excess" intellectual powerhouse that exists in all organizations.

In many industries, the high-tech firm has stopped being the absolute work boundary because no firm can fully and optimally deploy the complete impact of its employed intellectual powerhouse due to cognitive limitations of the form noted by transaction cost economists--limitations that have to do with coordination of complex projects and activities, as these activities grow in their intensity, depth of intellectual requirement and fullness of impact.

The sharing environment of the "open source" movement literally brings to life that "excess" capacity that is only "excess" because within the boundaries of the firm it lacks the complete opportunities offered in a larger-than-any-firm community context that can increase and multiply the value of that intellecutal capacity.

More, later ---

2006-06-23 01:57:36.0 -- ; Permalink ; Trackback.

Monday May 01, 2006

[ Work ] Why Schedules Fail

Thus, writes Scott Burken in his The Art of Project Management:

Project schedules are the easy scapegoats for everything that can possibly go wrong. If someone fudges an estimate, misses a requirement, or gets hit by a bus, it is the schedule (and the person responsible for it) that catches the blame. If the nation's power supply were to go out for 10 days, or the team's best programmers were to catch the plague, invariably someone would say, "See, I told you the schedule would slip" and wag her finger in the schedule master's face. It's completely unfair, but it happens all the time. As much as people loathe schedules, they still hold them up to an unachievable standard. Even the best schedulers in the world, with the smartest minds and best tools at their disposal, are still attempting to predict the future——something our species rarely does well.

But if a team starts a project fully aware of the likely reasons schedules fall apart and takes some action to iminize those risks, the schedule can become a more useful and accurate tool in the development process.

We're lucky to have a good writer explain all that in a truly well-designed book.

2006-05-01 13:22:49.0 -- Comments [2] ; Permalink ; Trackback.