Sunday Apr 29, 2007

3 years and still blogging...

Other Sun bloggers have been posting since Friday about the third anniversary of Dilbert even had a very timely series of comics.

Today is actually the third anniversary of my first public blog post (though as you can see, I later migrated some earlier posts from my now abandoned internal blog to it), as I was traveling to the first X.Org Developer's Conference on the day of the initial launch, but signed up once I was there (back when signing up was simply e-mailing Will Snow to ask for an account), and started posting from there.

In my first post, I worried I wouldn't be able to find enough things to blog about - now I know I have too much to blog about, and time to write it all up is the limit - I was posting the changes in each Solaris Express release for a while, and more recently, the X ChangeLogs for each Solaris Nevada build, but both fell behind after a few months. I've got a list of links to other blogs and interesting sites that I want to post that keeps growing longer and longer as I fail to take enough time to post them here. I wasn't sure I would even get this post done during the b.s.c birthday celebration.

As some of the other bloggers mentioned, the effect has gone beyond just providing a way for any employee to post - the company's culture has changed. No longer do we complain about hearing about new announcements first in the press - blogs are now considered a crucial part of many announcement plans, especially around OpenSolaris, and bloggers are briefed in advance. I joked with Ben Rockwood at the SVOSUG last week about how much he's been yanking Sun's chain lately - how long ago would it be unimaginable that the CEO, VP's and Chief Architect get personally involved when a sysadmin at a startup complained about issues with Sun?

So much change in 3 short years - who knows where the next years will lead us?

Monday Feb 19, 2007

What programmers look like...

Apparently, programmers look like me & Gary, according the Flickr title Alec put on the snapshot of us in his photos from last week's Sun Security Ambassador beerbash. At least I didn't get caught in a funny hat like the distinguished gentleman to our right (who was actually only a couple feet to our right when the photo was taken):

There's nothing like realizing you're surrounded by one of the inventors of public-key crypto and the architects of Trusted Solaris, having your photo taken by the developer of Crack, to make a X guy feel like he stepped into the wrong party. Though I wasn't feeling out of place for long, since just before Jim came down the hall and summoned me with the magic words “Free Beer” I had been working on the putback for the Xorg 7.2 vs. Trusted Extensions bug I'd been working on most of the week, so was answering questions about that, and then I got to meet a whole bunch of people who responded with “Oh! I know your posts on the [Sun internal] security-interest list.” In the end, I was once again feeling lucky to be able to work with such an amazing collection of talented and friendly engineers.

[If you couldn't tell, I'm the one just starting to show grey in his beard - or the one who is in most need of a trip to the gym. Maybe I should try Gman's new workout plan.]

Tuesday Jan 02, 2007

In honor of years of service

While the rest of the world has been paying tribute to James Brown & Gerald Ford, I have a much more mundane duo to pay respects to after many years of work which ended in December.

The first is John, who was the manager of the X engineering group at Sun for the past 5 years. He started at Sun long before that - so long ago that his first job in Solaris Desktop engineering was porting applications from SunView to XView. He later was part of Sun's Motif team, staying until he was the last member in the US after CDE & Motif development moved to Sun's Dublin, Ireland office. I can still remember him participating in the conference calls in which the CDE/Motif co-owners were trying to convince The Open Group to open source Motif, and only succeeding in convincing them to go halfway, cementing Motif's fate as a legacy toolkit and leaving the path clear for GTK+ and Qt to replace it as the toolkits of choice for new applications.

After nearly 8 years in the X group, first as an engineer and then a manager, he's decided to return to the academic life, which will be a big change for him and us. It will be interesting to see how things change around here once our new manager starts next week.

The second is the machine that has served as for the last 8 years. When it first went online in 1999, the Ultra 30 was intended to be a temporary machine, to be replaced soon by a more suitable server, but we never quite got around to that. During the 8 years since, we think it served almost half a billion http requests (we don't have all the logs, since we lost some to full disks, but that's about what the average hits from the logs we have would work out to). We've now moved the site to a hosted services provider, so that people with more time than us to do system administration can do so - one sign of how little time we had for that can be seen from this last set of commands I ran before the final shutdown:

# uname -a
SunOS amidala 5.8 Generic_108528-08 sun4u sparc SUNW,Ultra-30
# uptime
 1:08pm  up 1725 day(s),  9:41,  1 user,  load average: 0.07, 0.03, 0.02

Almost 5 years without even installing a kernel patch, much less an OS upgrade - but even though it was open to the internet at large, the network services were limited, and while we did get caught by spammers a few times who found ways to relay through us before we got sendmail patches updated, as far as we can tell, it was never broken into.

Thursday Mar 30, 2006

7 years and counting...

I didn't even notice the date yesterday, but when I looked at the calendar today, I realized I'd just passed my seventh Sunday. (If your birthday is the anniversary of your birth, wouldn't a Sunday be the anniversary of your joining Sun?) March 29, 1999 was my first day in the desktop group at Sun, though back then I was an intern in the PowerClient Desktop Release Engineering team, working on tools for the release engineers, fixing bugs in build scripts and Imakefiles, and filling in for the X, CDE, and OpenWindows release engineers as needed. After my internship was up and my degree was in hand, I was asked to stay on as a full time employee . A year after I first started, almost to the day, I moved into the X Engineering group as a developer, and have been here ever since.

Sunday Jan 22, 2006

People coming and going at Sun

Going: Richard Giles

Sad to see Richard Giles leaving Sun. I got to meet Richard in person in 2000 when he came to the US Sun offices. At the time, he was one of the first people in the field to spend much time with the then-new Xinerama feature in Xsun, and he stopped by to talk to the engineering team. He also created an internal Xinerama FAQ that helped save us from having to re-answer commonly asked questions from early users and helped everyone keep track of the various patches needed to Xsun and to other Sun applications to get everything working well together with Xinerama on. His enthusiasm will be missed at Sun, but I'm sure it will serve him well on his new projects and wish him luck there.

I almost wrote that Richard was Sun's first podcaster, for his I/O Podcast, but I suppose first external podcaster would be a better description, because Scott McNealy has been doing the internal equivalent of a podcast since before anyone hooked them to RSS and called them podcasts. About once a month for years now, Scott has recorded The McNealy Report to update people inside Sun on what's going on inside the company. Styled as a radio show, it often includes interviews with partners, customers, Sun executives, or internal project leaders. It's been going so long that it was once available for subscription as cassette tapes for listening in the car - now it's mainly distributed inside Sun as online audio. (Personally, I prefer to read the transcripts, since I can read faster and my brain just seems to absorb printed text better than listening - which drives my wife crazy sometimes.)

Coming: Thomas Haynes

I was amused to see a new Sun employee posting how much he liked the strict processes employed in his group at Sun compared to his previous employer. Especially when compared to another Sun blogger posting "Why I don't believe in code review", it makes for interesting food for thought. Of course both of these make the same mistake, talking about the "Sun" process, when it's really just the process used in their groups, and different parts of Sun have different standards for code review. Not even all of Solaris operates under the same rules, and our group is even learning to deal with the very different processes of the open source X.Org source tree vs. our Solaris X source trees. Somehow I don't see us all coming to agreement on how much code review is needed anytime soon.

(The other thing that drew me to the post was the title "Ex-NetApp and Damn Proud of It." It struck me that while I used to know lots of NetApp employees, I actually know a lot more Ex-NetApp'ers now than the few people I know still at NetApp. It's not that NetApp is a bad place to work - Fortune clearly thinks it's a good place - it's just that most of the people I used to know at NetApp were in the customer support groups, and when NetApp decided to move it's main support center from California to North Carolina - they chose staying with their families, homes, and Bay Area lifestyle over following their job across the country. A bunch of them ended up at TCP acceleration startup RiverBed.)

Coming Soon: More X hackers?

The posting for the job opening in our group was supposed to be posted last week to the Sun jobs site, so I went looking to see if I could find it to post here. I found a couple that could be our groups, but wasn't sure, so asked before posting. (You can still contact me directly if you're interested and I can pass on your resume.) I was surprised to see how many of the other groups we work with also have job openings for engineers to work on X & related technologies. For instance, the Project Looking Glass team is looking for someone to, among other things "Add 3D extensions to X server." The Sun Ray group is looking for someone with Audio & Video knowledge, including if possible experience with the XVideo extension. Someone to work on graphics drivers is being looked for by the SPARC graphics group. And for those who prefer working higher up the stack, Glynn recently posted an opening in the GNOME group.

[Technorati Tags: .]

Friday May 13, 2005

New bloggers at Sun

I noted in a blog posting last summer:

Elsewhere I noted that the movement towards greater transparency and customer communications via efforts like are a bit strange at the moment since they are both pushing up from the bottom of the org chart, with many engineers and other "individual contributors" participating, and pushing down from the top, with people like Jonathan Schwartz and John Fowler participating, but hasn't met in the middle yet, with the layers of middle management still out of the picture - where many of the decisions people want to know about are made and best explained. (For instance, you can find my blog here, and that of the VP I work for, Glenn Weinberg, but you won't find the manager I report to, the senior manager he reports to, or the director he reports to (who in turn reports to Glenn).) Perhaps it will just take time and growing numbers above and below to squeeze them out of the conference rooms and out here with the rest of us...

Recently, the gap has closed, with the two directors our group works for starting blogs — Andy Roach of the Solaris x86 group, and Stephen Harpster of the OpenSolaris effort. (At Sun, we find org charts that look like the traditional tree form to be boring, so ours are much more complex graphs, which is why my boss reports to two directors - a "solid line" report and a "dotted line" report.)

They're not the only new blogs springing up on though, and there's a bunch of new ones from those of us working down in the trenches, digging through the code. Those I'm particularly watching:

Jay Cotton
If all you had to go on was e-mail, you might think he's one of the quieter members of Sun's X11 team, but no one who has met him in person would make that mistake. Jay has done a lot of work on Xsun's Xinerama and power management code, and most recently has been working on delivering the Xorg 6.8.2 server for the Linux release of Sun's Java Desktop System.
What's Brewing by Randy Fishel
Randy works around the corner from me, and we chat a lot in the hall. I'm never quite sure what he's working on because he's got his fingers in so many parts of Solaris - the webmin packages included in Solaris 10, the initial JDS release for Solaris x86 last year, network drivers, power management, and every time I look it seems like he's working on something new. And when he's not talking about what he's cooking up next for Solaris, he'll tell you about the beer he's brewing.
OS Technology by Tim Marsland
The architect of the Solaris AMD64 port, he's now writing a series of posts about how that was done.
[] [] []

Monday Dec 20, 2004

Sun tied with IBM for 11th oldest .com domain name has trawled through the whois records of domain name registrations and published a list of the oldest still active .com domain names - Sun & IBM tied for 11th place with both domains registered on March 19, 1986. (At the time, I was in 7th grade, learning to use my school's new Macintosh 512k and using the Apple ][+ at home, neither of which had any sort of connection to any other computers other than via floppy transfer.)

My first e-mail address was at a slightly older domain though -, registered April 24, 1985.


Engineer working on Oracle Solaris and with the X.Org open source community.


The views expressed on this blog are my own and do not necessarily reflect the views of Oracle, the X.Org Foundation, or anyone else.

See Also
Follow me on twitter


« February 2017