I have little to add on the substance other than to note that I find these posts (and the discussions here) incredibly informative. They shed light on an area that turns out to be far more interesting than I would have imagined.
However, the last example may be more realistic than you suppose. DEF CON last year contained a presentation[1] on security considerations in HFT networks, which I attended. It's thought-provoking and something I hope gets a little more attention.
What did you learn from this? I found the slides and read them, and it seems like he's just saying "the people who build these systems believe that firewalls and TLS add too much latency to be deployed".
(I have a lot of opinions on this subject but I am mercifully restricted from sharing most of them owing to professional obligation; we do a lot of work in this field).
Mostly, insight into the technical underpinnings. I had absolutely no concept of the existence of this world before that talk. In terms of take-home value, not much more than I got from your talk at BH last year, though largely because of the fire-hose effect there. :)
If you're looking for good talks on the security or technical implications of trading markets, look for microstructure details. Any talk that has a diagram of "orders" going to a "trading engine" is addressing itself at a higher level than you're interested in. Just the order entry side of a real firm OMS is too complicated to get one bubble in a diagram.
Again: strongly recommend _Trading & Exchanges_ by Harris. The TCP/IP Illustrated of markets. Supremely readable.
Funny, I just bought this on Kindle a week ago. While not exactly a page turner, it is indeed supremely readable. I'm about halfway through and remain fascinated.
However, the last example may be more realistic than you suppose. DEF CON last year contained a presentation[1] on security considerations in HFT networks, which I attended. It's thought-provoking and something I hope gets a little more attention.
[1]: https://www.youtube.com/watch?v=kjIdzBtTBnI