I have tremendous respect for Mr Musk and his team at SpaceX. To have designed and built the Falcon 9 and the Dragon, and to have them work perfectly every time, in the short time they had, is an amazing achievement.

On the other hand, this really isn't the first "privately built" spacecraft. Almost all of the "NASA" rockets and spacecraft were built by independent contractors. NASA did a lot of the design work on the Saturn rockets and the spacecraft, but the Redstone, Atlas, and Titan rockets were all designed by private contractors for the military. SpaceX has some advantage in that it's doing everything under one roof (literally).

It is impressive to see that hatch open -- showing the depths of the cooperation between NASA and SpaceX. NASA has to have been working on this almost as hard as SpaceX over the past year to develop the procedures for the rendezvous, capture, and berthing of the Dragon. The opening of that hatch might not be as historic as the Apollo-Soyuz docking of the '70s but it's right up there.

At that point, nobody had ever lost a remote control, so it would have been a fairly unappealing selling point for a lot of people.

Uhm. You are aware that using PHP for anything security related is like making a vault door out of lit sticks of dynamite, right?

There is nothing inherently dangerous about PHP. But, the phoChain login page is secured behind a normal HTTP / Apache login. So, we have it set up so you have to be logged in as a valid user before you can even see the phpChain login page. It's also on an internal server, so it can't be accessed from the Internet. (Or, if you can, we have far greater security concerns to take care of first!)

Lots of very large web sites use PHP in public facing applications every day. From what I understand, that includes Facebook. While it's easy to make vulnerabilities in PHP, there isn't any language where it is especially difficult to make them. In general, you introduce vulnerabilities in proportion to the amoung of pieces that are in motion. Something like phpChain really only does a few small things, so it is easy to see how every feature interacts with every other feature. Compared to something like Wordpress, something like phpChain is much easier to get right. It doesn't need to give content to untrusted users. It doesn't need to interoperate with other sites. It doesn't need feeds or pingbacks. It doesn't support public comments. It doesn't do embedded flash. It doesn't need a full SQL backend. It just stores some passwords. And, it does it well.

We just use a shared account for "engineering department (location XYZ)" passwords. You can also have an individual account if you want to have private passwords, and you could put the password for any shared phpChain accounts you need to access in your private phpChain account. We have it running on an internal server, rather then something exposed to the Internet, so the danger of a breach is minimal. (If anybody makes it that far, we are already hosed.) But, the passwords are all stored in an encrypted form, so we could safely send backups offsite if we needed to, etc. Also, you need to log into the page with a normal HTTP login (which is tied to the NIS account you use everywhere on our network) before you log in with the shared phpChain account, so we have an audit trail wrt which individual person accessed what, despite using shared accounts.

We use phpchain at work for this sort of thing. A few hundred accounts for various servers, devices, vendor support accounts, and logins for accounts at companies we work with. All stored securely. Google it if you arent familiar with it. It has been a huge win for us, and does everything asked for. We even wrote a simple search functionality for at that I think has been rolled into mainlIne at this point. Certainly better than a plain text file on a shared drive!

(tried posting this previously, but I wasn't logged in. Trying again now that I have gotten home. Hopefully it is more noticeable now.)

When movies were first made, they were single shots. A train approaching a station, something like that. Audiences oohed and ahhed.

But, the first time a cut was introduced, the audience was completely flummoxed. They had no idea what they were seeing. It's hard to believe that now, but we've probably seen 100,000 cuts by the time we are 5 now, and our brains are rewired to accept it.

Okay, sure, OpenCV is terrible. But, what do you reccomend as the alternative? I think it's similar to ffmpeg in a lot of ways. Sure, the API has grown like a cancer, but as inelegant as it may be, what else has the same functionality and reasonable licensing terms that you can use to pull off the shelf and start hacking? Complain about the speed if you want, but if I use OpenCV, it certainly runs better than if I had to try to reimplement all the code myself! IMHO, part of the problem is scope creep. OpenCV includes stuff like file IO and GUI development. If they just gave up on that, and kept easy interoperability with OIIO and Qt, and focused on only maintaining actual computer vision code, the scope might be more manageable. (And, focusing on documentation so I could have half a chance at actually using the mess woulkdn''t hurt either...)

I used to fly my lightplane back and forth from my home in the San Francisco Bay Area to my Los Angeles office on the fourth floor of a building in Hollywood.

There was an antenna across the street that looked exactly like the profile of an airplane heading toward us. Whenever I was walking down the hall and would glance out the window, I would see that and immediately, uncontrollably, startle. When you see a plane that close you literally have a second or two to make a decision, and it becomes a reflex to act immediately. Now, walking down the hall of a building no reaction is actually called for; but it didn't stop me from jumping!

I'm skeptical of the cost. $60B 2010 dollars is the estimated cost for high speed rail from SF and Sacramento to LA and San Diego. You're telling me I can get a maglev to fucking space for that much? Please do it if it's true, but I don't believe it.

Fair point, but when you look at how much cheaper-per-mile the LA-San Diego line will be compared to the LA expo line expansion, it's clear that these things scale very non-linearly, and cost more when you have to deal with existing infrastructure and population. Stick a megastructure in the middle of some accessible but vacant desert, and you will get a lot more bang for the buck than if you tried to stick it near extremely desirable California coastline.

The bathyscape Trieste used a similar electromagnet-holding-shot system that Cameron's sub uses. They had a bit of a surprise after many successful dives.

It seemed that the steel shot that they obtained in Europe had a substantial amount of impurities in it; and the system worked just fine. When they filled the hoppers with American shot, though, it was pure enough that the electromagnets didn't just hold the shot, it magnetized it! Even when the electromagnets were turned off, the shot stayed in the hopper.

If I recall correctly (and I read about this 40 years ago!) they were able to dump the entire shot canister to get back to the surface.