June 14th, 2016

But I never got around to actually doing it

Around 2005 I was really into the whole field of non-photorealistic rendering (NPR). I’ve pored over dozens of research papers and spent months on implementing some of the basic building blocks and combining them together to recreate the results from a couple of those papers. It even got as far as submitting a talk proposal to SIGGRAPH of that year. I thought I had an interesting approach. All of the reviewers strongly disagreed.

Edge detection was one of the building blocks that I kept on thinking about long after that rejection. A bit later I started working on a completely different project – exploring genetic algorithms. The idea there is that instead of coming up with an algorithm that correctly solves a problem, you randomly mutate parts of the algorithm generation that you currently have and evaluate the performance of each mutation. The hope is that eventually these random modifications “find” a path towards the optimal solution that works for your input space. You might not understand exactly what’s going on, but as long as it gives you the right answers, that part might not be that important.

My homegrown implementation was to come up a set of computation primitives – basic arithmetic operations, a conditional and a loop – and let it work on “solving” rather simple equations. It was rather slow as I was basically doing my custom completely unoptimized VM on top of Java’s own VM. As I started spending less time working on it and more time just lazily thinking about it, I kept on bouncing two things in my head. One was to switch my genetic engine to work at the level of JVM bytecode operations. Instead of having a double-decker of VMs, the genetic modifications and recombinations would be done at the bytecode level, and then fed directly to the JVM. The second idea was to switch to doing something a bit more interesting – putting that genetic engine to work on image edge detection.

I’ve spent months thinking about various aspects of what could be done with such an engine, and how novel the entire thing would be when it’s all done. But I never got around to actually doing it.

Around 2007 as I was in the middle of working on a bunch of libraries for Swing (a few extra components, an animation module and a look-and-feel library), I fell in love with the idea outlined by some of the presentations around Windows Vista. I wrote about that in more detail a few years ago, but the core of it is rather simple – instead of drawing each UI widget as its own thing, you create a 3D mesh model of the entire UI, throw in a few lights and then hand it over to the rendering engine to draw the whole window.

If you have two buttons side by side, with just enough mesh detail and reflection on texture you can have buttons reflecting each other. You can mirror and distort the mouse cursor as it moves over the widget plane. As the button is clicked, you distort the mesh at the exact spot of the click and then bounce it back. Lollipop ripples, anyone?

I’ve spent months thinking about various aspects of what could be done with such an engine, and how novel the entire thing would be when it’s all done. But I never got around to actually doing it.

Around 2010 as I wound up all my Swing projects, I decided that it would be a good experience for me to dip my toes into the world of Javascript. So I took the Trident animation library that was written in Java (with hooks into Swing and SWT) and ported it to Javascript. It was actually my most-starred project on Github after it hit a couple of minor blogs.

I don’t know how things in the JS land are looking now, but back in 2010 I wanted a bit more from the language, especially around partitioning functionality into classes. Prototype-based inheritance was there, but it was quite inadequate for what I wanted. It probably was my fault, as I kept on going against the grain of the language. As the initial excitement started wearing down, I considered where I wanted to take those efforts. In my head I kept on going back to the demos I did for Trident JS, and particularly the Canvas object that was at the center of all of them.

Back in the Swing days my two main projects were the look-and-feel library (Substance) and a suite of UI components that had the Office ribbon and all the supporting infrastructure around them (Flamingo). So as I started spending less time working on the code and more time just lazily thinking about it, I thought about writing a UI toolkit that would combine everything that I’ve worked on in Swing and bring it to Javascript. It would have all the basic UI widgets – buttons, checkboxes, sliders, etc. It would all be skinnable, porting over the code that I already had in place in Substance. Everything would have animations from the port of Trident. I was already familiar with the complexity of custom event handling (keyboard / mouse) for the ribbon component in Flamingo. And it would all be implemented at the level of the global Canvas object that would host the entire UI.

I’ve spent months thinking about various aspects of what could be done with such an engine, and how novel the entire thing would be when it’s all done. But I never got around to actually doing it. Flipboard did that five years later. The web community wasn’t very pleased about it.

Some say that ideas are cheap. I wouldn’t go as far as that, even though I might have said it a few times in the past. I’d say that there are certainly brilliant ideas, and the people behind them deserve the full credit. But only when those ideas are put into reality. Only when you put in the time and the effort into making those ideas actually happen. Don’t tell me what you’re thinking about. Show me what you did with it. For now I’m zero for three on my grand ones.

February 17th, 2016


Gotta hand it to the FBI. Take their proposal of a completely custom system build that would circumvent various protections that are designed to keep people away from your information, and consider it on purely technical matters. It’s simple to explain to technical people, and it’s simple to explain to people that are not that well-versed in technology.

But consider this. There is no such thing as a safe-guarded backdoor. Do you really believe that once one government agency gets their hands on such a system build, it will only be used to help the “good guys”? If you are, I’d love to live in the fantasy world you’re inhabiting.

In the real world that the rest of us live in, this will be nothing short of a disaster. This custom build will get shared between various government agencies and branches, with an absolute guarantee of two things happening. First, it will get to an agency that is solely overseen by secret courts that rubberstamp pretty much every request. Second, it will get into the hands of general public, sooner rather than later – through social-engineered hacking or another Snowden-like act of political activism.

And then there’s another absolute guarantee. Let’s for a minute say that if you’re a law-abiding US citizen, then the US government is good guys. Then there are other governments who are our allies, which makes them good guys by proxy, and there are other governments which are our enemies which makes them bad guys.

What is going to stop other governments from demanding access to the same special system build? How many countries can a multi-national corporation withdraw their business from before it has no more places to do business in? How do you as a supporter of lawful information “extraction” decide on which laws you agree with and which step over “the line” that separates the good guys from the bad guys?

There’s not a single line in Tim Cook’s letter that is a gratuitous exaggeration of the dangers that lie ahead. I’ve spent the first twenty years of my life living in the communist USSR, where it was pretty safe to assume that the state had the capabilities and the means to do mass surveillance of anybody and everybody.

How does the self-aggrandizing beacon of democracy turn into the omnipresent surveillance state? Two ways. Gradually, then suddenly (in the mighty words of Ernest Hemingway). Just don’t tell your kids that you didn’t see it coming.

September 28th, 2015

Tech reporting

Two types of tech reporting / analysis out there.

The intellectually honest, hard working one where you look at publicly announced products (and maybe vaguely hinted future plans) and invest effort to make intelligent, thoughtful, researched predictions of how the different players in different interrelated markets will evolve in the next couple of years.

And then there’s “building your reputation” on top of some anonymous source inside the company that feeds you the actual plans and blueprints. The lazy kind where you’re just the conduit of plans and upcoming product announcements, always at the mercy of that source being continuously employed in that position of “close knowledge”. The kind that is disguised as “original content”, and is anything but.

Guess which one is more respectable even if you might make a mistake or two down the road. Play the long game.

June 17th, 2015

Table stakes

A long long time ago I wrote about companies being in control of their own long-term destiny. HP? Seriously? Silly me. And where’s Amazon? Anyways…

Not so long ago it used to be that a platform provider would give you a nice set of core widgets, along with access to the network stack, the local file system and a few lower-level graphics APIs. Good old times of shrink-wrapped software and platform updates that happened once every five years. In a good decade, that is.

Then the Internet happened, and after the dark old times of IE 6 a couple of big companies realized that not only they needed to have a fast and standards compliant browser available on their platforms, but that it’s such a basic block that it can’t be left to other parties. And so browsers developed by the platform providers became a table stake.

Somewhere in between, GPUs became more affordable to be included in the standard consumer-grade computers. Hello Moore’s law. They also happened to become quite powerful. Hello Moore’s law again. And OpenGL happened along the way, paving the road to gradually hardware accelerating all the things that find their way to the pixels. Along the way, local and streaming video became the expected part of a platform offering. Table stakes, if you will.

As a parallel track, platform providers started offering (built internally from scratch or acquired in a mad dash so as not to appear to fall behind) email services, ad platforms, office suite of apps (documents / presentations / spreadsheets) and search engines. Table stakes 2.0.

Then mobile happened. Not the kind focused on texting or plain-text email. Mobile as in having a frigging supercomputer in your pocket.

First came the app stores, along with the whole infrastructure around them such as authentication, serving downloads, replicating across multiple devices, user reviews, and search and recommendation engines. And oh, dipping the toes in the wonderfully macabre world of payments. Table stakes level++.

And you also had the other, more traditional media – books, movies, music and magazines. So the platform providers expanded the store capabilities to serve those, along with the matching applications to consume that content – reading books and magazines, watching movies and listening to music. Along the way came movie streaming, music streaming and music subscription services. You might also throw in the TV media players and TV streaming / casting sticks into the mix. Table stakes again.

The platforms themselves started branching into ever more form factors, from the traditional desktop screens and the just introduced mobile screens (a phone-sized screen and a tablet-sized screen) into a screen continuum. From phablets to TV screens, jumping into the wearable space (watches and glasses) and most recently into the automotive space. Along the way the platform providers started working on seamless integration flows between multiple devices owned by the same user. That included persisting the flow state and replicating it between devices, as well as unifying the user experience between the various “manifestations” of the underlying platform. And as the platform had successively more chunks moving into the cloud, the companies started investing ever more design resources into defining their own interface language and guidelines to help developers create apps that feel part of the platform. Table stakes raised another level.

And indeed the platform became this amorphous thing, with individual devices and screens at the edges and the data storage / sync layer in the middle. Platform providers offer their own storage and syncing solutions, with (mostly) generous free tier, spanning a variety of user-generated content, from documents to spreadsheets to, most recently, comprehensive photo services. And hey, let’s throw a mapping solution into the mix while we’re at it. Hello table stakes 2015.

There are quite a few pieces whose timelines I’ve generously rearranged in this article. There are a few pieces that were left out for brevity. To name just one, how about a bespoke system font that is used throughout the multiple manifestations of the specific platform. While we’re at it, how about the overall typesetting system in the world of high definition screens and the amazing “inventiveness” of various languages, living and dead. But I digress. Oh wait, how about a frigging voice-controlled assistant that predicts what you want even before you know that you want it?

Being a platform provider in 2015 is a big undertaking. And getting into the platform game might never have been as formidable. And when you’re in that game, you better up your game every year on the clock. At your own developer conference. But hey, no pain no gain. And the gains these days are enormous.