Posted on September 20, 2018
I became an engineer after graduation - although I also had a website on the side that was getting millions of pageviews a day. Then I became a startup founder. And then a CTO. And then another startup founder.
I was writing thousands of words, putting together pitches and decks, speaking all over the world, having partnership conversations and leading product development - but all the while, I was still described as an engineer. It was a label that stuck.
This is a disservice to the people who have spent their life in true engineering. It's also a misdescription: I'm not a top-level engineer and could never pretend to be, but I understand the technology and how it fits into the broader narrative, and the broader social context. I can lead products well because I can understand both the engineering and the business sides. I can use human-centered design and design thinking - both journalistic processes - to de-risk businesses quickly. I can wrap it all up in a narrative, and I can use that narrative to build a community of support that snowballs, Katamari-style. It's not something that fits into a neat pigeonhole, but I think it's more interesting.
I've become really appreciative of other people who don't fit into the pigeonholes that others try and fit them in, both in work and life. Observing from the outside, the people who are really making change are multidisciplinary, often guided by an overarching mission. They're not worrking on something because they want to become the best at a particular skill, but because they want to build something that achieves a certain effect. It's the difference between trying to ace an exam on a particular subject, and trying to create something that nobody knows exactly how to grade because it uses so many different skills. That can make it more difficult to find the right job - I've often had to make jobs for myself, and when I haven't (like now), I'm drawn to collaborate with similarly multidisciplinary outsiders. But for me, it also makes for much more fulfilling work.
Posted on September 19, 2018
We’ve been slowly entering an exciting new era of “computational” photography where software continues to overcome previous limitations of hardware. The iPhone XS is capable of running 1 trillion operations per image, and while extremely powerful, I’m also keenly aware of the fact that we know less and less about what’s going on to capture our image.
A key part of the creative process and achieving one’s artistic vision is troubleshooting. In order to troubleshoot, one must understand what is actually happening and what is causing the problem.
With a traditional SLR camera, if my image was too bright, too dark, too soft, etc., I knew exactly what to change/tweak to get closer to my vision. Today, with cameras heavily relying on software, sometimes things happen that I just don’t understand. Perhaps the tones in the sky don’t look quite right, or a vertical pano isn’t in focus like I wish. The difference is, I don’t know WHY it doesn’t look the way I want it to, which means I don’t know what to tweak to fix it.
Of course, the upsides of computational photography far outweigh the downsides, and almost always the software helps me capture exactly what I want, but I’m curious about how this conversation will develop over the next few years and how Apple will explore new ways to facilitate artistic expression.
Posted on September 19, 2018
For security policies to be followed, they must be known and enforced wherever possible and reasonable. If your users can’t follow your policies due to business process conflicts, or you can’t enforce the rules due to a lack of technology or another shortcoming you’re unwilling to mitigate, then you’re probably better off not having them at all.