atomicbeanie

It’s about the data. Not the code.

Switching to Clojure did it for me.

I would say https://cognitect.com/blog/2014/7/22/transit is a better option, no?

Would be useful it seems to have a flight path marker on the display.

Super interesting. Thanks for the info!

I am guessing the "Tic Tac" ufos are drones with prop fans.

Not much, so much more is possible. But if Stack Overflow goes out of business, I quit. I do not miss trying to find obscure stupid things myself. The Internet is very good at making my time more valuable.

Have you looked at clojure.spec? (https://clojure.org/guides/spec#_generators)

And the truth is most systems of any kind seem to fail from simple things that everyone could have prevented but overlooked. Wasted engineering. Best to think like the Arpanet was designed. Have a way to heal once things are normal, not work around the bizarre failures.

I think that is inaccurate. One fleet was grounded quickly. The 30 million Kevlar tank linings were underestimated in cost I think dramatically. As for profit, I am not sure the aircraft was ever a profit center. Any reasonable observer saw Non-recoverable Engineering Expenses that the company any could never recover from as a result of having to mitigate a new safety requirement onto an old aircraft that was never designed with modern standards in mind.

I think it is a very good example of how old systems continue to operate while the public believes they are up to modern standards. We see it in civil engineering now too. A bridge collapses in Pennsylvania and the public is told that over 1/3 of US bridges are outside of their designed life span and more expensive to fix than replace with municipalities that cannot afford to do either. The Concord was old and had what we call Tech Debt. It could not be made modern. Risk is just negative opportunity cost. When it crashed that debt was realized and the company had to write it down. One thing for sure is correct in your statement. It is about money driving the decisions, not engineering.

There are many examples too. The concorde is one that has my current prize for being the most extreme example. One crash, retire them all. There was no reasonable way to operate that aircraft by modern safety standards. But it was certified in what, the 60's. One likely could have certified a barn door strapped to a Concorde engine in those days ;-)

I do not agree in general, but I admit to not knowing the details of that system. Those of us in the industry have been waiting for 737 rudders to fail for a long time. The data was there. This is the same as waiting for a pilot to rip the tail off an airplane as happened over New York. The certification basis of those aircraft was not in harmony with the information the operators used to fly them and gravely distorted the engineering that had gone into designing the aircraft.

In both cases, there was a certain inevitability given the processes at work, the conditions, the changes and the difference between engineering reality and what is presented in the FAA Type Certification process as the basis for the modification of the Type Certificate. As an engineer working on aircraft, one thing is for certain, engineering data is laundered to the bare minimum in the interests of certification. This is in contrast to the engineering NASA does. The cert process does not achieve transparency, as it is a legal not an engineering process.

One can blame manufacturers, but there really is no way for them to be transparent and ever ship anything (ref. Lear Fan). I am not, however, advocating retracting any responsibility from Boeing. They must bear the responsibility for their product. I am only saying that we have created an environment as a country that makes it unlikely we will see the behavior we desire from manufacturers. As a person who likes to land the same number of times I take off, my interest is in helping the government improve as well as the manufacturers.

So Boeing decided to conclude that the existing system was sufficient. And they proved it to the FAA. There is no precedent or standard applied to the parts of the airplane that are not changed, once they are deemed not to need changing. This is similar to the housing code. Replace a socket in a house with knob-and-spool wiring, and you do not need to fix a thing. Rewire a bedroom in a house with knob and spool wiring, and the code will not let you put new knob and spool wiring in, one must then bring the house's electrical system up to code.

The reason there still is a 737 is the tremendous economic benefit afforded to manufacturers who use safety analysis to show that the previous iteration of the aircraft's systems are sufficient given proposed changes on the aircraft. That's why it has ridiculous engine nacelles, ridiculous landing gear, in many 737's ridiculous avionics. A new airframe is required, but the cost of a new airframe's certification is much greater than the hacky re-certification of an old type certificate obtained when standards and processes were very different and much out of date.

Manufacturers take the blame for this. And they take the blame for things like no global transponder in the loss of the 777 over the Atlantic. Unfortunately the FAA processes, while enlightened in some ways, and firmly grounded in the science of safety, are effectively a strong deterrent for a manufacturer to avoid changing anything in a design.

The result is that many aircraft operate for a very long time with very outdated systems. Replacing designs is prohibitively expensive to prove to the FAA that there will be no corresponding degradation of the system's performance or new safety risk. Unfortunately such a process does not calculate the cost of not replacing the system. No cost is attributed with keeping something that is old and lacking in capability.

The result is that aircraft systems are woefully behind what technology can offer. And this is not just the hardware or the software, it includes the procedures and the overall set of capabilities. The result is that aircraft are being operated to the standards of the 50s, when in fact a much higher standard of crew and aircraft performance is possible. When I say performance I am also talking about safety performance, the ability to operate without harm causing failure.

-1 Moving the evolution of a groups code into the hands of the person who works with it the least is an anti-pattern. Driving change from within a team can be facilitated by a manager who reads the group's code, but any manager who thinks they are going to be the judge of what the team is achieving is deluded and overly enthusiastic about their authority.

Google Wave

Great news!

Very nice. I hope the outcome is as the artist intends! Thank you for pointing to this link.

Every human in Mumbai is probably hunting that cat now. I wish a conservationist had seen the photo first. The article is likely a death warrant for the cat.

Woot!

So very frustrating. We've done the same thing in this country when Europe had it together. Can't Africa get a break from our first world bureaucracy's?

My goodness. Future terrorism delivered as memes. Terrible genius?!

Nice article.

Great, teach our "authority" figures to lie, threaten a punishment "unthinkable" to the perp (victim), and have them plea bargain (admit to anything to avoid such a false accusation and punishment).

Should be titled, how principles are joining into the professional party state prosecutors have been enjoying for years. "Get more convictions, look like a pro!" All you have to do is act like a extortionist. Why let the criminals have all the fun?!

And we wonder why our children's live's are desperate now like the adults? And we loose some of our brightest stars... RIP Aaron Swartz.

I think the tools are challenging, but it is soooo refreshing to have tools that don't pass the complexity on to the user. The parser library that has done this for me is https://github.com/engelberg/instaparse. Using Instaparse is downright liberating. BNF in, parser out. Done.

Of course, if the police do stuff like this: http://www.theguardian.com/us-news/2016/feb/07/leaked-police... it is unlikely they are willing to take responsibility for enforcement. Rarely does enforcement happen without oversight. By hiding police records they protect themselves, the police department leaders. We should, however, be able to do better. Public records, paid for by the public, are easily considered public property. I think courts could handle this.

This is interesting. Normally police departments claim they have few ways to catch corrupt police. And I think there is more than a little truth to that. However, in this case the article shows that corrupt police are revealing themselves through the practice of destroying equipment. This could be an indicator that acts as a "canary" for police corruption! If police departments appropriately monitored such indicators, they may be on verge of discovering an easy method to spot criminal police officers before they do the damage to society.

It is comic that they would protect milk for children, when the DOE was experimenting on unwilling hospital patients: http://fas.org/sgp/othergov/doe/lanl/pubs/00326649.pdf

Not so much "Government for the People". Why such decision makers are above the law is baffling, really.

Try harder seems like it has been done in this area for some time. Can some creativity get into this process? It gets worse the more we invest. The cartels are ever more monstrous.

Since they're enjoying Star Wars while legislating, maybe Padme's quote is apropos: "So this is how liberty dies. With thunderous applause."

Of course if there is no privacy with the resulting data or the DNA input to the machine, adoption may become difficult. What will the government do with such information: http://www.innocenceproject.org/free-innocent/improve-the-la...