Wrong Number

Once upon a time, ’round about 1983, my mother was living in Houston, where I occasionally visited her while I was attending the University of Texas at Austin.

One evening during one of these visits, the phone rang in her apartment. She answered, the man’s voice on the other end asked for some name who didn’t live there, then apologized for dialing the wrong number. Mom told him to think nothing of it. Then, somehow, they ended up in a conversation, most of which I could hear (the voice and/or the phone was unusually loud). This complete stranger ended up pouring out his heart to my mother.

The crux of the matter was: he liked wearing women’s clothing, and he wanted to tell someone about it. He seemed lonely, looking for acceptance, which my mother provided. She was completely unfazed by his deep, dark secret, refusing to find it upsetting or disgusting, or to think that he was a bad person for wanting to wear lingerie and stockings. At some point she must have mentioned that her daughter was visiting, because he then wanted to talk to me, and ask all the same things over again: did I think he was weird or disgusting? No, not at all. He seemed relieved that we didn’t slam the phone down on him. He talked eagerly, asking what kinds of clothes we liked (neither of us had much interest in fashion; perhaps that part of the conversation was disappointing).

He wondered if he should be a woman instead of a man, and we didn’t find that strange, either. I had never, to my knowledge, met anyone who was contemplating a sex change. I felt shy to voice an opinion on so big a topic, but I tried to be supportive.

Eventually, he thanked us for listening to him, said goodbye, and hung up.

I’ve always remembered that call, because it struck me how isolated he seemed – so afraid that he was a horrible person, and that everyone around him would shun him if they knew who he really was, or wanted to be. He might well have been right about that, at that time and place. It seemed likely that the call had not been a wrong number at all, but a random dial in hopes of finding a sympathetic ear. Whoever he was, I hope he worked it out and finally got to be who he wanted, and that he didn’t have to do it alone.

What Linux Can Learn from Solaris Performance, and Vice-Versa

Brendan Gregg keynoted the Southern California Area Linux Expo this year, to a packed room, with this talk:

How does Linux system performance compare to other OSes, particularly the performance-focused Solaris family? What features inspired by them could be added to Linux?

Both are bristling with performance features and optimizations, and it’s difficult enough to fully understand the performance of the Linux kernel and its distributions, let alone other kernels and OSes for comparison. Brendan Gregg has unique insight into the performance features and analysis capabilities of both Linux and Solaris-based systems, which he covers in depth in his new book:Systems Performance: Enterprise and the Cloud. He also works at Joyent, a high performance cloud provider, where OS performance is core to the business, and frequently debugs head-to-head performance comparisons. It’s not just each OS’s baseline performance that matters, but also their analysis tools, and how quickly potential customer benchmarks can be debugged and tuned. This talk will include specific areas where SmartOS – an open source illumos kernel derivative of OpenSolaris – often beats Linux performance, and vice-versa. How does Linux compare today, and what can it do next? … and what could the Solaris family learn from Linux?

dtrace.conf 2012

In April, 2012, I organized and ran the second-ever DTrace conference (the first had been held in 2008). I found the venue and sponsors, did all the logistics, live streamed and filmed the entire day’s proceedings. It was run somewhat unconference style, with Bryan Cantrill emceeing, so the final list of talks you see below emerged over the course of the day.

Perhaps my greatest feat for this conference was persuading the Oracle DTrace for Linux team to attend and speak!

For a good overview and wrap-up, see Adam’s blog post on dtrace.conf.

8:30 AM registration, coffee by Stone Cobra, breakfast by DEY for illumos  
9:00 opening  
9:15 State of the Union – video Bryan Cantrill
  Setting the Agenda – video  
10:15 coffee break by Stone Cobra  
10:30 User-Level CTF – video Adam Leventhal
10:45 Dynamic Translators – video Dave Pacheco
11:15 Control flow & language enhancements – video Eric Schrock
12:30 lunch sponsored by Nexenta  
1:00 PM coffee by Stone Cobra  
1:15 Carousel ride!  
1:30 Clang Parser for DTrace – video John Thompson
2:00 Visualizations – video Brendan Gregg
2:30 Visualizations, Enabling toolchain for seamless USDT – video Theo Schlossnagle
  Visualizations – video Richard Elling
3:20 Coffee by Stone Cobra  
3:30 DTrace in node.js – video Mark Cavage
4:00 User-land probes for Erlang virtual machine – video Scott Lystig Fritchie
4:45 DTrace on Linux – video Kris Van Hees
5:30 ZFS DTrace provider Matt Ahrens
5:45 DTrace on FreeBSD – video Ryan Stone
  Bryan throwing big heavy books at people  
6:00 Barriers to Adoption – video Jarod Jenson
6:30 beer sponsored by Basho  
7:30 out of venue, go for dinner  

Many more videos about DTrace can be found in my YouTube DTrace playlist.

Jenolan Caves

After a somewhat abortive attempt to see some of New South Wales’ national parks, we decided to drive to Jenolan Caves, which Brendan had visited on a school trip as a child. I was dubious. Somewhere along the way I had spotted a poster advertising the caves, which showed formations lit in garish colored lights. It looked, well, …cheesy. But we decided to take a chance.

Brendan did not remember that the final piece of road approaching the caves was narrow, winding, and barely two lanes, carved into a mountainside. We also didn’t know that the time we were arriving – late afternoon – was when the big tour buses were leaving. So we had some excitement trying to get past a bus on the road. Italians have more practice at this.

You can only visit Jenolan Caves as part of a guided tour, one to two hours per cave. We arrived in time for the last tour of the day of Chifley Cave, led by a long-time employee, John. Neither my words nor my photos can do justice to the weird beauty of this ancient limestone (340 million years – the oldest known open cave complex on Earth). You’ll just have to go see it for yourself.

John did show us the room lit in colored lights, near the entrance of Chifley. He told us that they kept it to demonstrate how caves used to be lit for tourists (I remember this being the case in the US during my childhood as well), but now they use white LEDs so that you can see the beauty of the cave’s natural colors. In fact, all that we saw of Jenolan is very tastefully done, with well-designed lighting making the most of what nature has put there.

We were fascinated enough that we decided to stay overnight (there are several lodging options right there at the caves) so that we could go on another tour in the morning. We were tempted to also join that evening’s tour, but it had been a long day of driving and an hour of walking and climbing lots of stairs in the caves. We opted instead for a quiet walk around nearby Blue Lake, where we saw and heard wildlife, including the resident platypus (video here).

The next morning we toured Orient Cave, where I shot the video you see above which, again, can give you only a partial impression of the size and splendor. (The photo at top is also from Orient.) The tour guide you hear in this one is Ann; yes, she’s American.

The “very sparkly” bit you see in the beginning of the video is fresh limestone crystal that has formed in the 60 years since the new entrance tunnel to Orient cave was dug through (causing damage that you learn about on the tour). Old as they are, these are living and growing caves.

NB: This was a 90-minute tour with a lot of up and down stairs. Some parts were very narrow, and overall it was quite strenuous – my legs were sore the next day. The Jenolan Caves site lists a “fitness level” for each tour (Orient is listed as “average”); I suggest that you take that seriously!

 

That Marketing Thing

At the beginning of my Monktoberfest talk on Marketing Your Tech Talent, I said that: “I have occasionally had a marketing title; it has never been a good fit.” This was an overstatement. In the course of my career, I have had several good experiences as part of or working with marketing teams, for example at Adaptec, starting with: “…my new boss, Dave Ulmer… I told Dave what I’d been doing online (CompuServe, Usenet, answering emailweb pages) and he said: ‘Fine, do more of that.’ And that’s about all the direction he ever gave me. He not only understood the Usenet, he was frequently and visibly out there himself.” (more of this story here)

Even when not been formally part of a marketing organization, many of my tasks and activities could be classified as “marketing”, broadly defined as “communicating the value of a product or service to customers” (wikipedia). Such activities have included:

Continue reading That Marketing Thing

Deirdré Straughan on Italy, India, the Internet, the world, and now Australia