There's a reason why Pascal and Java made good teaching languages for so long.

I'm going to have to disagree with you, there, as far as Pascal is concerned. Pascal has always been a terrible teaching language: it's too complicated for beginners and not nearly powerful enough for experts.

It is a complex and fairly large chunk of code that "fixes" a nonexistent problem

I have to disagree with you, there. Unix-type systems have needed a new, dependency-based init system for at least 20 years, now. I'm amazed it took as long as it did to replace. I won't argue that systemd breaks the Unix philosophy of doing one thing well, and suffers from some overreach, but at least someone took some initiative.

Yeah, most people didn't have a mainframe in their house.

While it's true that microprocessors never really acquired mainframe I/O, they've had virtual machine support as far back as the 680x0 series, back in the 80s.

A few months ago, I was having lunch with two of my coworkers. Let's say their names are "Sean" and "Nate". Nate is a seriously hard core conservative; I'm guessing he's a Tea Partier, but he may be just your typical Republican.

At one point, Sean notices Nate is wearing a jacket in the style of a uniform from the original Star Trek - a blue one. Sean asks, "Why blue?". Nate replies, "Science officer."

By definition, a laptop has to compromise, for space and for weight.

That may be true, but some companies are better at compromise than others. I have an Asus GX-something-or-other. It's almost five years old, but it runs almost everything I've tried, and the fan hardly ever turns on, as long as I keep it clean. The laptop ran Minecraft perfectly well, with an HD texture pack, until the 1.8 update (but I think that's a Java garbage collection problem). It chokes on Kerbal Space Program, but then, it's five years old.

The irony is that it's only taken 40+ years to get to display resolutions for raster graphics to approximate vector graphics.

I'm not sure why that's ironic. When you're only displaying the outlines of objects, you don't need nearly as much memory (or memory bandwidth) as you do with a raster display. On top of that, vector displays only get that resolution in monochrome; you lose it when you try to do color. (A color display can't exceed the resolution of its shadow mask.) I can tell you from experience: Quake looks better on a 640x480x8 raster display than it could ever look on a vector display.

Please, not another useless Chrome clone.

I agree. Other than greed, I can't understand why they don't just make an agreement with Google or Mozilla - preferably both - to have one of their browsers automatically installed with Windows. Writing a browser from scratch is a huge project, and while I'm sure it's a tiny fraction of Microsoft's output, that's a fair amount of resources that could be directed elsewhere, while generating a fair amount of good will in the software community.

if you're not willing to build a bunch of nuclear power plants and shut down a bunch of coal plants, then yes you ARE arguing global warming to advance a political agenda

Some of the most prominent AGW scientists are strongly in favor of nuclear power: http://www.scientificamerican....

Others are a little more cautious, but still think nuclear is an important part of an overall strategy to reduce global warming: http://www.ucsusa.org/our-work...

To extend and extinguish, of course.

Yeah... My guess is that, after this announcement, developers are going to say to themselves, "Great, now we don't have to learn how to use new tools to create software for Linux", and do all their work on Windows. Fewer people will work on development tools for Linux, too, because they can use tools that already exist on Windows to create software for Linux.

Then, in five or ten years, when everyone's using Microsoft's tools, they'll claim no one's using them to port to Linux, anyway, and drop support. Developers will have no choice but to use Linux's (poor or non-existant) development tools, or drop support for Linux, altogether.

Brilliant!

As long as we're on the subject, I'd like to know about such software, too, but I'd like something that's OS independent, and stores images locally. My mom has an enormous collection of family photos, dating back to the early 20th century, that I'd like to catalog while she's still around. It would be nice if she could do the annotations on her Windows machine, while I organize everything on my Linux machine. Ideally, we could copy the images and associated data back and forth using a CDROM or USB key.

"Why do you need that $700 enterprise-grade AP? Just use the $69 linksys one like I do at home!"

I just got a very similar question, yesterday.

We're adding 350 workstations (and PoE phones) to our network - something for which we should be seriously looking at a Catalyst 6500 or a Nexus 7000, right? No, I cheaped out and got some Catalyst 2960s and a pile of SF500s. Total cost? $20000.

What's the first question they asked me? "Nowadays, I can get two terabytes of hard disk for $100. Why do we have to pay so much for our network ports?"

They use my bandwidth (without permission)

You know, I don't even have a problem with that. I kind of feel like looking at their ads is the price I pay for viewing their content, and (assuming the ads were less intrusive) it's better than having to explicitly pay for every site we access. I held off installing Adblock for a long time, but when their ads are so CPU intensive they hang my browser for minutes at a time, that's where I draw the line.

That's because computers at the time couldn't keep up with the drives (unless you really knew what you were doing). I read a bunch of old QIC tapes I have, a few years ago, on a relatively modern machine, and the drive only stopped at the ends of the tapes.

Like it or not, human field of view is much wider than it is taller.

Citation needed.

Interrupts worked fine. It was bus errors (i.e. for off-chip memory protection and/or mapping units) that were a problem. The 68010 fixed that particular issue if I recall.

You're correct, except for the fact that it wasn't a bug. The original 68000 simply wasn't designed for use with demand-paged virtual memory. To make that happen, you need to either save the processor state somewhere (which the 68010, 68020, etc. did) or have restartable instructions (the approach used by National Semiconductor, for their 32000 series). I vaguely remember reading that Motorola switched to restartable instructions in the 68040 or 68060, but I'm not sure.