Interesting assertion. Now, fire up Excel, and prove it.

I have no problems with pure R&D projects like this in peacetime - it's where any sensible civilization should be spending the bulk of its military money when there's no major conflict afoot. My problem is with immensely inefficient and corrupt programs designed from the get-go to produce a less-than-adequate result (F-35), and with enormous standing armies and fleets we neither need nor can afford. (12 carrier groups. Really? We need all that? The Chinese have, umm... one. The one they bought second-hand from the Russians, that's not even operational yet.

Until you get deployed to the SS Flounder...

Yeah, but then you're in the same attack group as the USS Otter, USS Blutowski and the USS D-Day, which is awesome.

It also shows a 60/70's naïvety toward how nasty our computing world has become toward attacking other users for personal and political gain.

Yeah, mitigating modern malware techniques, particularly trojans, is a non-trivial problem. Apple's solution, the walled garden, is probably the wrong one, but no-one has come up with another credible security model that works as transparently or effectively for the end user. This is really an area of OS research that needs a ton of attention and effort that it's not getting - anti-malware applications are not cutting it. The solution needs to be baked-in, not bolt-on, and pro-active rather than reactive.

This is exactly right. It's easier and more reliably profitable to work with domestic industries than to accept piecework from a country where there's a language and time zone barrier to collaboration. Companies in India were not capable of delivering a consistent result, because companies in the US were unable to offer consistent guidance and management. Pretty much as predicted by all the outsourcing skeptics. Once the domestic demand for IT services in India took off, outsourcing lost its luster for buyers and sellers both.

So, in essence, they don't want to pay IT staff what they're worth and can't find enough suckers willing to be underpaid, and believe the salesman when he says his company can do all of that messy IT work for you, dirt cheap. Heard that same song sung before - remember how everyone was going to lose their IT jobs to Indian outsourcing? How'd that turn out?

I'm thinking old fashioned extortion. Cough up some protection money, or bad things will happen to your network access... which is why the Egyptian authorities knew to look for them.

No, of course not. I draw the line at having an actual keyboard

A tablet has a keyboard, and also a mouse - the function of both is combined with the screen on tablets like the iPad and pocket computers like the iPod Touch or Nexus 4.

More to the point, "keyboard" is kind of arbitrary. "Can run software applications, and is not specialized to run one particular category of software" is probably a better definition - and both tablets and smartphones qualify. iOS and Android are both Unix variants, if you recall.

But no Holga? You are a heathen.

Ooh, look at you with your fancy high-tech camera. What are you, some kind of robot? Now, the Diana, that's a camera for a Real Man.

(In art school, I met someone who had a Holga, which he declared took the pretense out of fashion photography... he had it mounted to top-dollar Metz 45 potato-masher strobe. We all snickered at him uncharitably.)

Yeah, but to get an equivalent field of view at an equivalent speed, you're stuck with a wide angle design for a normal lens, and an ultrawide for wide angle. This introduces size and expense and has tradeoffs in quality and speed. (Thing start to work in favor of the smaller sensor once we start talking telephotos, tho...)

I hate them all.

- Full frame DSLR bodies are ludicrously expensive.
- All the glass worth owning is designed for full-frame - smaller DSLR formats get slow and/or soft zooms and a smattering of primes in useful focal lengths at useful speeds. Non-full frame DSLR owners are a despised afterthought, despite driving most of the profit.
- All of the mirrorless systems with great glass at good prices have crummy bodies.
- All of the mirrorless systems with great bodies have overpriced/slow/crummy/all-of-the-above glass.
- Modern primes are stupidly overpriced. Adjusted for inflation, N/C/S/P lenses with mediocre quality are pricier than equivalent Leica or Carl Zeiss lenses sold in the '90s.
- Ditto flash units.

I sold my Contax kit, I'm ready to take the plunge after sitting on the sidelines for a while - Fuji XE-1 with the kit zoom looks like the (reluctantly chosen) winner. It takes gorgeous photos, the zoom is sharp, contrasty and fast, the other lenses in the system are superb and (for primes) reasonably priced, the old-skewl controls make me feel at home - I just need to put up with craptacular EXF and The Worst Autofocus in Scotland.

What serious database (sql speaking or not) or operating system is written in a "non-legacy" language?

A few of them were written in ADA, FORTRTAN and COBOL. These are clearly non-legacy languages, and the engine that powers the bold new ideas of industry and computer science.

No seriously, it's all old shit, and anything remotely interesting is happening at a higher level these days. Sorry. Even app programming is being abstracted with stuff like Cordova. There are a lot of job openings, as operating systems and RDBMS's dating from the 70's need updating on occasion, so they've got that going on at least.

Python can be either interpreted or compiled.

No-one uses it compiled anymore for new projects. Hence legacy.

You can't transform from "an interpreted language" to "a compilation target"

Of course you can. You're arguing a point of semantics no-one in industry accepts as valid.

Google would probably be surprised to learn that Python can't handle that use case

No they wouldn't. They invented Go and Dart because Python can't handle that use case.

So, also, Python (JVM and CLR via Jython and IronPython), Ruby (JVM, CLR, and JavaScript via JRuby, IronRuby, Opal/HotRuby/Ruby2js), etc.?

That's cute - but no. Legacy languages with bolt-ons few people use. Time marches on, and in another ten years you'll be mocking all the Clojure and Node.js guys wondering where their glory has gone.

I remember when the O'Reilly Python manual came out and everyone was excited - this was back in Clinton's first term of office. Python is old. Interpreted languages in general are old, and increasingly less useful. Javascript is transforming from an interpreted language into a compilation target due to its integration with web technologies - Python is just python. Jython was a thing for a while, but it's not functional-language enough to handle new virtual and cloud infrastructures, where an app has to run on thousands of machines, scaling up to thousands more in response to load, all without operator intervention.

Languages that are not legacy compile to JVM, CLR and/or Javascript - Clojure, CoffeeScript, Opa.

Heck for pretty much any creation, you want a real system. A tablet is fine for watching a video, it would suck for editing one.

Ummm. No. This is knee-jerk conservatism. There are already video editing apps for the iPad, one of them by Avid, 3d modeling apps from players like Autodesk, and absolutely no end of audio apps, some of them being used to crank out pro-quality product.

Sure, a full-blown workstation will provide more screen real-estate, storage and processing power - but those little phones have more oomph in them than professional workstations used to model Jurassic Park or compose and master the Fight Club soundtrack. Software options are catching up rapidly as developers and users learn how to work with the new user interface constraints.

Projects like Blender need to start figuring out how their application is going to work without a mouse and keyboard, as that's where the users are going. Once upon a time, it was inconceivable you'd need to develop a GUI for your application, too - non-touch interfaces are starting to look a lot like Word Perfect for DOS.

Adjust their initial expectations. Explain very clearly on your website and download page that this product is free to download and free to try, but paid contracts are required for professional support. Do not refer to your product, anywhere on your site, as a being available for free, or "free software" - the term is confusing. Open Source is a better term.