It takes a special kind of arrogance to assume these scientists (including the ones that peer reviewed the work for publication) aren't aware that "correlation doesn't prove causation". I mean seriously? You really think nobody thought of this incredibly obvious fact you're pointing out? You think the "after accounting for all variables" right there in the summary, means absolutely nothing at all?

As others (like Baloroth below) have pointed out, you clearly HAVEN'T "taken a good look" at the study, and the things they DID control for. Or in other words... you started with a conclusion of your own ("this study is BS"), and didn't even BOTHER looking for supporting facts.

to start with a conclusion and work his way back to find supporting facts

He started with a HYPOTHESIS, and established a method to test it. An intelligent critique of his specific methodology and control factors would require you to actually read the study. But knee-jerk reactions sure are more fun, aren't they?

When you've identified actual factors he hasn't controlled for that might explain the correlation, perhaps you can submit your own paper. I'll be right here holding my breath.

First of all, that's not even close to what you said. You said (emphasis yours) "MSDOS machines were pretty much only good for Doom". Now it's "ohh... what I REALLY meant was that other gaming platforms were more popular than PCs". Um, DUH. But that's not what you said.

Second, your new argument doesn't hold water either. Off the top of my head...

Civilization, Sim City, Command & Conquer, Warcraft I & II, Ultima, Commander Keen series, Duke Nukem series, King's Quest series, Space Quest series, Wolfenstein, Test Drive series, Master of Orion, X-Wing/Tie Fighter series, the early Wing Commander games, Maniac Mansion/Day of the Tentacle/Monkey Island/other Lucasarts titles, Prince of Persia...

Yeah, just a gigantic pile of "irrelevant edge-cases" there that nobody played on PC. Sorry, but you're on crack.

He's not talking about XBOX. He means that if you want to do a lot of gaming on a PC, you're running a Microsoft OS, and it's been like that for a very, very long time... and will be so for the foreseeable future due to sheer inertia & quantity of titles.

there was a time when an x86 box with MSDOS was pretty much only good for playing Doom 1

This is so wrong, I don't even know where to begin. Are you really implying that PC games were scarce before Doom?

There's an ocean of air above our heads, thousands of kilometers deep

Thousands of km???

Try "about a hundred"... and 75% of the atmosphere is within the first 11km. (Source).

This is why you rebuild your compromised environment. Which is exactly what the bulletin says they did.

Thinking back, one of the earliest things that hooked me was a Big Trak they brought into my grade 1 class. I was fascinated by the idea you could give this thing a sequence of instructions and it would carry them out.

Of course there was Logo on some school computers (Apple II's?) to play around with.

Then there was an illustrated children's book on BASIC programming from my local library... used robots running around putting things in boxes (ie, memory) to illustrate the concepts. It seems impossible to find similar books these days, which saddens me. I wrote my first programs on paper (we didn't own a computer) based on stuff I learned from that book. My first bona fide code was written on a C64, on display at a K-Mart. (10 PRINT "BRIAN" 20 GOTO 10)

Later, we borrowed a VIC-20 from a friend for a few months, and I began programming in earnest. Wrote a text-based casino game that quickly exhausted the 5K of user RAM.

Eventually I convinced my parents to buy a C64. Actually mucked around with assembler on that. Then (much later), our first PC, a 386... GW-Basic, QBASIC, Turbo Pascal (and more ASM), etc.

Another great reason to use Firefox during development is it's the only one (at least among FF, Safari, and Chrome) that can shrink down all the way to 320px (and narrower), which is important for Responsive Design/Mobile... everything else seems to stop at 400px.

It also has Responsive Design View, which is a godsend for precise info & control of viewport dimensions.

I used to love Firebug and hated Chrome Inspector, but am now the complete opposite. ;)

the arcane kludges needed to produce popular web-page layout effects easily achieved using evil tables

Rejoice brother, for the era of display:table-cell has begun (supported in IE8 and up, and IE7 is effectively dead, thank god). Nice clean semantic markup, but now with access to all the juicy table features like vertical & horizontal-block centering, shared column height, and automatic column sizing. (Good article: http://www.digital-web.com/articles/everything_you_know_about_CSS_Is_wrong/)

You know there's something wrong with a standard when Microsoft's broken box-model implementation makes more sense.

We now have the option to use this as well :) http://paulirish.com/2012/box-sizing-border-box-ftw/ Again, supported in IE8 and up.

Especially on responsive/fluid sites (where you often want to do things like, "I want this column to be 33.333% wide, oh but I also want 10px of inner padding"), this is a revelation. Sure we could've used a wrapper div, but sometimes this is not an option, and either way it makes the markup cleaner.

Other IE8+ stuff we can now use includes:
- :first-child (but hilariously, NOT :last-child)
- multiple class selectors (.foo.bar)
- direct-descendant selectors (.foo > .bar)
- :before and :after (IE8 still has some minor bugs, but easily avoidable)

Here's the BAD news:

IE10 has removed support for conditional comments, and .htc (something like PIE polyfill is now impossible). Go ahead and try getting something like border-image (or even a reasonable fallback) working in IE10, I dare you.

the lack of 'constants' to set standard colours and measurements

You're of course correct that native CSS still has no concept of anything like that. This is where LESS or SASS come in. And with the power of mixins (ie, to easily generate all the browser prefixes for more advanced stuff like gradients, shadows, etc), it's getting borderline irresponsible to not use a pre-processor. I instantly switched to SASS when I discovered this: http://jakearchibald.github.io/sass-ie/. This technique, I am convinced, is THE way to write maintainable, sophisticated responsive code (with a good path to easily strip out all media queries for IE8) right now.

Virtually every app (outside of games) is for viewing pre-generated content

I'm sorry but this is an increasingly unsupportable (yet oft repeated) claim. Maybe it was true for the first year or two after the iPad's release but it's certainly not any more. A quick scan on iTunes of "Top iPad Apps" (skipping games of course) shows virtually NONE of them are merely for consuming content.

Pages/Keynote/Numbers, various note taking and artistic apps, video/photography/editing, an INSANE number of professional grade music production apps (currently seeing a huge new surge thanks to Audiobus) by the likes of Korg/Moog/Steinberg, etc etc etc. There is an enormous diversity of creative apps for the iPad, and this will only continue to expand and mature.

Now... if you want to claim that the majority of iPad USAGE is for consuming content, fine... but that's a very different argument. And one that could be applied to conventional computers too... (are the majority of computer users really content creators? Or web surfers and YouTube/Netflix watchers?)

Well I'm just going to have to disagree.

Cutting up all your meat at once (and subsequently scarfing it all down at once, sans pauses) allows for greater batch processing efficiency. Switching back and forth between cutting and eating is analogous to excessive context switching.

This practice also allows one to drop the knife when the cutting phase is complete, and then switch the fork over to the primary hand for optimal input rate.

I am a skilled practitioner of this technique, and I can assure you that its overall throughput is such that the last piece is NEVER cold.

I suggest you read up on WHATWG (Web Hypertext Application Technology Working Group).

THAT is the real "browser vendors" group, formed as a direct result of W3C's horribly slow, ineffective, and impractical (remember strict XHTML?) output. It was WHATWG that pushed HTML5 forward (by actually going ahead and implementing it), which effectively forced W3C to abandon XHTML.

Parent's post is entirely correct... W3C made themselves irrelevant.

It's absolutely true that you need a tenacious thirst for learning new things to stay relevant in any technology field. But it's entirely possible to cultivate that without holding a degree.

Call it what you want, it's not claiming to have a degree without actually listing one (ie, "B Sc." or similar).

While I did not complete my degree, I do have a couple years of university-level Comp Sci education at a good school, why on earth shouldn't I use that fact to my advantage, in getting my foot in the door? I am entirely open about my education during the interview process, it's never caused any problems, including accusations of dishonesty, for me. It's just to get you past HR types sticking your resume in the "no university education whatsoever" pile. Why SHOULD I be in that pile?

The simple fact of the matter is, it really hasn't mattered at all past my first job anyway. My resume is a dense wall of experience, with a single line at the bottom regarding education.

And I don't consider your "Yale, Harvard, etc" example at all equivalent.. because it cannot truthfully name a specific academic program that was attended (ie, Computer Science), while I can.

Look, I am not trying to discount the value of a degree, far from it. But the OP asked for practical advice on alternatives to completing his CS degree, I'm just telling him what happened to work out very well for me, that's all.

I just wanted to add, I do consider my time at university very valuable, and likely would've considered switching to something like Software Engineering if it'd been offered at the time, and still think about completing a degree at some point.

That being said, I've always found this adage to be true: "Education is what companies settle for, when they can't find experience".

I went to the top CS program in my country (U of Waterloo), dropped out after 2 years in '99 to join some friends at a startup (originally doing Java dev but quickly switched to web work), and have never looked back. The startup didn't pan out, but it gave me the experience to get a "real" job. (I was also largely self-taught from hobbyist programming from before university). And you can still put "University of XXXXX, Computer Science" on your resume without lying, which will get you past certain gatekeepers. ;)

You know it's funny, back in the day browser limitations and incompatibilities made web development pretty challenging (especially during the transition from tables to early CSS). Then things have gradually gotten a lot easier as CSS support has slowly normalized to an extent (IE, especially 8, is still a thorn in the side of course). But now with mobile & touchscreen, and things like responsive design (also, post-Flash interactivity & animation) coming to the forefront, the pendulum swings back, and web dev is becoming "hard" again.

That's the curve that is ripe to get on right now. For front-end anyway. For backend you can pretty much either go LAMP, or .NET (Java's dead in this space). Experience with frameworks/popular CMS's will also be useful.