Anyone who knows their own password, and is already logged into the gun, can fire the gun, if they just speak "sudo fire", and then say their password.

Plus they can keep killing people with just "sudo fire" with no password for a configurable amount of time since the last "sudo".

... plus "Wait, let me quit Skype and re-start; I will call you back".

For instance, it would be nice if the lastest version of Skype for Windows didn't sometimes freeze for 45 seconds at a time when merely sending an instant message, with no audio or video call in progress.

A translator for Skype? Bah, that just needs a fixed vocabularly of audio files consisting of "can you hear me now?" and "I cannot see you!" in 50 languages.

If you want subjects who don't mind not bathing for four weeks, just go to any CS lab.

You know, no matter how large a dinosaur you find, how can you prove that it's the largest?

Not without digging up every cubic meter of the Earth's crust to some reasonable depth.

Pass by reference is the norm? That is moronic and encourages bugs. We should avoid destructive manipulation such as variable assignment as much as possible in programming; reference parameters exist to make it possible to modify a caller's variable.

If you have reference parameters in the language, then any foo(var) call can potentially modify var. If it doesn't today, then someone can change it tomorrow to give himself access to var inside foo.

The C convention of taking an explicit address is safe against this.

Pass by value should be the norm. Pass by reference shouldn't even exist.

Lisp is purely pass by value; there is no pass by reference: just that some types have reference semantics (cons cells, arrays, etc).

Some old ideas indeed make much more sense. But all of those still-relevant old ideas came from Lisp, and none from Fortran.

So, Fortran issues extra instructions and pipeline stalls for accuracy, yet manages to be faster.

That is amazing!

Do these scientists develop friendly graphical user interfaces for their Fortran programs?

Do these programs have robust and secure handling of all input?

How about configuration: are there dialogs for setting up preferences, which are persisted somewhere?

Do they package up user-friendly installers?

How much of their stuff runs on new platforms like tablets and smartphones?

What non-Fortran-stuff do these programs integrate with? Anything over a network?

Where can I download a scientific Fortran program to evaluate its quality?

Has anyone written a viable program of the following in any dialect of Fortran, new or old?

- operating system kernel
- device driver
- web browser
- web server
- instant messenger
- audio/video telephony client
- etc

Wow, faster AND more accurate. They must use some mystical floating-point instructions that only Fortran compiler writers know about.

Is it possible that scientists work with throwaway programs that don't integrate with anything, are operated by command line interfaces, and assume that all their inputs are correct and trustworthy?

Fortran is a plastic shovel with a rolled up newspaper for a handle.

If it chooses what target to select and makes the call on whether to attack the target, it is autonomous.

If a human chooses the target and makes the strike call, the machine is not autonomous.

Complete no brainer.

Just anti-vms.

A sucker is born every minute, I guess.