38MB sounds only a bit larger than just ICU (31MB on my machine), so Qt isn't adding much there. ICU is used by most GUI frameworks (Microsoft has their own version, but OS X ships it as part of the standard install) and includes things like fast unicode collation (locale-aware sorting is hard!) and fast unicode regular expressions. Most apps that need to work in places that aren't just the English(ish)-speaking parts of North America need most of that functionality.

I don't need to install an alternative shell. I've got one that works just fine out of the box. It's called the Windows 7 UI.

FWIW, it's not the start menu I'm bothered about. Since Win7 I hardly use it anyway, I just have my regular applications set out in the task bar and use jump lists probably 90% of the time I load one. This gets me to anything from a spreadsheet I worked with recently to a shell on a remote server I use regularly with two clicks and is one of the cleanest UI set-ups I've ever seen in an OS GUI.

The thing that annoys me about the Win8+ GUIs is how dumbed down and in-your-face they are. Huge areas of bright colours (yay for eye strain), boxy styles where you never quite know what you can click (sorry, tap) until you try, clumsy icons that don't really tell you anything anyway, and everything all spaced out so fat-fingered people with tablets don't accidentally reformat their disk instead of sending an e-mail. For someone using a keyboard and mouse with good screen(s), all of this is moving backwards. If I wanted dumb UIs for simple stuff, I'd buy an iPad and use web apps instead of desktop applications.

I do realise that some of this related primarily to what was then called the Metro UI in Win8 and some changes have been made since then. But from what I can see so far with Win10, it looks like they're pushing the overall UI theme even more in that direction, even if the default method of interaction looks more like a traditional desktop again.

This is true, but it's also completely irrelevant. Safari uses WebKit, including WebCore and JavaScriptCore. All of the Safari features that are not part of WebCore and JavaScriptCore are entirely user-facing and irrelevant to web developers. If you look at what's actually included in the WebKit nightly builds, you'll see that it's a build of Safari.

There are pretty good reasons for believing that a key to the improved environment on Earth will be the migration of many processes off the planet. I'm not a particular fan of Space Solar Power, but it's definitely in the running. According to experts in the field, SSP could eliminate all of the power plants on Earth (both fire-based and nuclear) and provide easy cheap power everywhere for less. (IMHO it would be cheaper in the short run to just build big solar facilities in the Sahara, 30 feet off the ground. This would generate plenty of power and provide a new resource underneath - shade where things could be grown.)

As someone who is involved (peripherally) in the "New Space" movement, IMHO the first purpose of space development will be the availability of new resources and technologies. An economist a couple of years ago predicted that space development would have the potential to increase the standard of living of everyone on Earth by a factor of 10. That seems optimistic to me, but a reasonable goal. One popular example (see Planetary Resources, Inc.) regards the availability of Platinum, which is a very useful industrial metal, but is unfortunately $1300 per ounce. Platinum mining is expensive, dangerous, and disastrous both ecologically and socially. This greatly restricts is usefulness although it is used in those expensive catalytic converters in your car - that's why they're expensive. The best astronomical physicists believe that some of the Near Earth Asteroids contain single-digit percentages of Platinum. If this is true, then a 100 meter asteroid would contain a dozen times as much Platinum as has ever been mined. Retrieving this material to Earth could drop the price to between $10 and $100 per ounce, and this would still be economically viable for the company to process in space and ship it down to Earth.

There are many other examples. Technologically, the range of industrial processes that are presently either expensive or impossible on Earth due to gravity and air, that could be done in the high vacuum and microgravity of space is broad but it is likely that an order of magnitude more new processes that have not even been envisioned yet will be discovered or invented. Orbital production of high quality integrated circuits might well be one - one of the most expensive aspects of IC manufacturing is the requirement to build a huge facility and maintain a high level clean room environment. In space that could be done with not much more than a bit of Mylar.

I expect those who grow up in space, or in a colony, to be habitually _very careful_. This puts 'kidproofing' at a whole new level.

Just fyi, an MSR can be built small enough (both weight and dimensions) to be driven around in a pickup truck.

Well the nice thing is that there would be plenty of open space. I'm not sure why one-inch steel - steel doesn't seem to be an ideal material for this. I don't know what the effects of all that sulfide would have on carbon, but if it can be made resistant I would think seriously about starting small with a probe that can produce a carbon-based skin and build a bigger balloon for itself.

Just to be clear - size is not largely irrelevant. The whole key to buoyancy is that the volume of a sphere goes up as the cube of the diameter while the surface area goes up as the square - for a non-sphere it's based on the three linear dimensions of course. So a very small craft can barely carry the skin, while a large one can carry much more in addition to the skin. There are other factors, but that's the primary one.

For example, a one-foot box made with one inch steel would not float well.

I recall not that many years ago when the prospect of a teraflop processor was science fiction. That was about 1992. A year or two before that I worked on some photometrically-correct ray tracing code, porting it to the Cray X/MP. That code took a month to make one 1024x1024 frame on the top-end Apollo workstation, and a few minutes to run on the Cray. It could probably run at close to 30Hz on my phone today, and today's supercomputers are in the 30+ petaflop range, i.e. 3x10^16.

So we're getting close - theoretically, if all of the top 100 supercomputers got together, the group performance would be in the 10^19 range. :) Actually that's not a bad idea - the powers that be could work a deal for all of them to work together for one week per year on the same problem, and the research time could be allocated the way that telescope time is allocated according to accepted/agreed value of a particular project.

I had an idea a while back, that actually relates to TFA. Genetically engineered bacteria or simple organisms that could float and live in the Venusian atmosphere and gradually begin to 'fix' the sulfides and whatever - maybe pooping out metallic sulfur. For the first long while, they would be working at the top of the atmosphere. Their poop would drift down and re-vaporize (absorbing energy and lowering the temperature). When they died, they would drift down into deeper layers and get to the point where their bodies would be heated back up to the point where the materials would be turned back into gas. But as they became more populous, gradually they would reduce the amount of solar energy (especially if their bodies were reflective), and the temperature. Eventually the might be able to reduce the temperature to the point where their poop, or that of their successors, would fall to the surface, permanently eliminating the sulfides from the air.

I'm happy to see that you don't hate systemd. That was the last shoe to drop. I'll complete the switch to BSD now!

Dragonfly BSD works quite well on the desktop, as does Funtoo Linux, which is systemd-free. Gentoo also works and still uses OpenRC by default, although there is growing concern some of the devs are quietly preparing to push a systemd agenda (kdbus patches in the kernel, one of the devs commenting he hopes systemd would become the Gentoo default, and a habit of the moderators in the Gentoo forums to shut down any discussions critical of systemd).

Linus may not be showing good leadership in this instance, but not everyone has drunk the urine just yet, and there are others stepping up to the plate to maintain or create alternatives.

Not everyone has a brand-new computer; The manuscript of the book I'm about to publish is in Open Office Word, about 400 pages and full of large images, and autosave is a real pain because it takes minutes to save the file.

Like another commenter said, I wouldn't make it the most important thing, overall efficiency is. But software speed is important to anyone with an older computer, especially a Windows computer, because the computer slows as the registry grows, and the registry never gets smaller, only bigger.

Oh.

I thought you sad "Homophones".

Nevermind. Back to discussing Youth in Asia.

Unfortunately, I'm in the UK, where the selection is much more limited.

For example, Dell UK's web site lists exactly one laptop with a 17+" screen and SSD, and it is also a touchscreen and comes with Windows 8.1.

HP do at least promote the Windows 7 option (via Win8 downgrade rights) for the high-end ZBook laptops on their site. However, the pricing on those tends to make the closest equivalent Retina MBPs in specification look cheap.

Also, Microsoft UK don't seem to have any high-end devices at all within their Signature Edition range, so it's invasive crapware city all the way with a lot of the big name brands, even on their expensive, high-end models.