Ultimately, you could have a unified realtime 3D model of the entire world, and you it would be completely robust and accurate, verified by millions or billions of video cameras. And you could still stream it to anyone, because the bandwidth optimizations. All you need is cameras everywhere. Privacy issues there, but surely in a few years we could have a non-realtime model of the entire world.
Multi-resolution images! Utterly brilliant. It seems so obvious after the fact, like all great ideas.
Aside from ignoring coarse-grained backgrounds which finer-grained images don't need to store, doesn't scalable vector graphics allow you to make those various resolutions? There must be a lot of room for compression, I wonder if you can expand on that (I'm only 20 minutes into the video however).
I'm a former neuroscience student and I think this is very interesting stuff and can imagine the idea being applied to many things.
Blaise, excellent work and great explanations. I want your monitor!
Great interview so far, Charles!
"I understand that was not the intention but as devs it's scary to watch automation carry into the things that used to once be the soul domain of devs and architects ..... the stuff that put food on our tables.... Lets just hope that the intended use is more often the case than the potential miss use......"
Automation is your friend!
It's way too late to turn back the hands of time; not to sound so melodramatic, but think about Ruby on Rails and Django, etc..., this is just the next step in web development and vital for .NET developers to compete.
I have already used BLINQ to generate a large website. The code is amazingly small and understandable to anyone who knows ASP.NET 2.0 and LINQ. I have never seen better than BLINQ personally, including Rails.
However it is still up to developers to work against that to meet all the business requirements - the pages are far from done is most cases. That's still a lot of work. Now you can focus on business programming, instead of technology programming, however.
I am just waiting for LINQ to be released to use it in earnest.
Very nice applications. I wish I had had ( oh, how I hate the had hads) XAML and WPF for a project I worked on under contract at Canada Post. A "simple" application to visualise address label layouts against a 2D (would of used 3D with WPF) representation of a mail sorting case(s).
The application had to automatically try to find the optimal layout for printing labels on mail sorting cases that made them easy to see, that covered the entire width of a slot in a case reserved for a particular address, and that had entire streets together in a row, when possible, or else, "line-broke" into another section of the case if it overflowed horizontally, and of course automatically re-sort and re-flow the labels as the user resized "address sections" to fit the case in an "optimal" way, including parameters such as the reach of the postal worker doing the sorting. And of course, allow the user to override certain labels as they know something the data doesn't, etc....
Printed addresses make up sections that vary in size but the application has to figure out how to make everything fit together right when printed out and applied to physical cases used to sort mail. We used simulated annealing to find the maximum size of street names without displacing other numbers and street names, and while staying within a specific range to match the physical slots within 0.5mm acceptable error. That was fun, and not very hard. But cool to say .
We did the graphics in GDI+. Managed DirectX was my first choice (in the non-beta category), but you needed another runtime, and we couldn't count on 3D acceleration anyways.
It took about 6 months, two developers (including me), one joining near the end to help out, and a great project manager.
Time is money for sorting mail. It's amazing that it's not 100% automated.