I think the verb “to democratize” is used here in the extended meaning “to make accessible to all”, like democracy makes governance an affair accessible to all by making it an affair of the people.

But I agree that this use does seem strange sometimes in some contexts.

A charming person, and a lovely, theatrical, speech.

If she had rehearsed it a couple more times, she could have achieved a perfect rhythm.

I was turned off a bit when she started using historical present to describe the incident at the Catholic school. She used it again in a couple more cases, but thankfully not for long. Most of the times I find historical present awkward. It is difficult to do well, and, to my eyes, it seldom succeeds in its purpose.

I imagine the “rather shockingly” clause was a misplaced attempt at irony. I cannot think of any other reason the author would put that there.

Thank you!

The Wiley page has some:

http://eu.wiley.com/WileyCDA/WileyTitle/productCd-1119962595...

I get confused with font stacks that have sequences like:

Calibri, Candara, Segoe, "Segoe UI", etc. etc.

Are there systems that do not have Calibri but do have Candara or Segoe UI? (Excluding, of course, cases where some of these fonts have been installed manually and individually.) I would suppose not. If so, then what’s the point of specifying Candara, Segoe, and Segoe UI after Calibri?

Am I missing something?

Segoe ships with some MS products (http://en.wikipedia.org/wiki/Segoe). I think the 'C' fonts are more tied to Windows releases (Vista and 7, probably 8, too)

So, I guess you can find Segoe UI but not Calibri on a XP system with a relatively (but maybe not too) modern Office. Maybe there are other requirements (newer IEs might install the 'C' fonts, too)

Looking at tasksel’s changelog on a Sid installation, I see the last change was v3.12 from 21 July 2012 and I cannot find anything about changing the default desktop environment for the desktop task. So, maybe this change is not meant for the frozen Wheezy, whose taskel is already behind Sid’s (v3.11 compared to Sid’s v3.12).

Maybe someone familiar with Debian’s decision-making and development processes could enligthen us.

Even better, put that in Preferences, Key Bindings, User.

Using the user file has two advantages:

First, you avoid mishaps when some future version makes changes to the default bindings file.

Second, you can quickly check what are your non-default settings.

Duly noted. Thanks!

In a few words:

250 of the 300 seats are allocated proportionally among the parties that got over 3% of the vote. The remaining 50 go to the first party.

So the "constituency" map has no actual electoral meaning?

Correct.

In fact, because the allocation of parliament seats is calculated on the basis of the total vote and because of the adjustments that this entails for the allocation of constituency seats, you can be the first party in a constituency and yet get fewer seats in that constituency than the second party.

Opera 12 is the first third-party web browser for Windows to offer an official 64-bit version. (Neither Firefox nor Chrome have official 64-bit versions for Windows. And I think Safari for Windows doesn’t have one either.)

Does this translate to any substantial improvements in speed? Does anyone here know of any benchmarks?

Very unlikely. 64-bit is mostly an advantage if you have to use a lot of memory or are doing processing on actual 64-bit integers.

For the kind of software like Opera (and Firefox and Chrome), the additional registers in 64-bit mode help a little, but that tends to be offset by most pointers now being 64-bits and the L1/L2 caches hence being less effective.

On my computer most memory hungry apps are browsers.

And by the way I use opera to open a lot of tabs and it handles those very well (compared with other browsers)

Right, but 64-bit computing isn't a magic bullet that speeds everything up. As gcp stated, there are both benefits and drawbacks.

Although it does break that annoying 4,294,967,296 open tabs limit!

This is nowhere near realistic. Opera's performance degrades linearly after the first 10000 tabs.

Do you have the 16bit build?

There are more registers in x64, so there may be some speed gain. Or a loss of speed due to larger pointer size and more frequent cache misses.

Registers, more instructions guaranteed (like SSE2) etc. It's up to the compiler to take advantage of that tough.

Did you just repeat exactly what I said?

There is an official 64-bit Firefox nightly built for Windows.

I tried it and to be honest I can't tell the difference.

Also bear in mind that it needs 64-bit Flash and Java plugins which are notoriously problematic.

Opera 12 uses a separate process for plugins like Flash and I believe it can actually use the 32-bit Flash within the external process while still displaying the content within the 64-bit browser.

Java and Flash are notoriously problematic (not to mention insecure) on 32-bits too.

64-bit is not really useful for browsers with multi-process architecture like Chrome. It is very unlikely that a single tab will need 64-bit address space.

Except once you have enought tabs Chrome starts putting multiple tabs in a single process. The cutoff seems to be somewhere in the low double digits. And at that point, you can in fact start running out of address space (not to be confused with running out of memory) without too much trouble, especially because system libraries on 64-bit OSes have a tendency to be mmap-happy, since they think they have lots of address space to work with.

64 bit address space significantly improves the effectiveness of ASLR, 64 bit tagged pointers can store way more data (7 byte strings, dunno if anyone actually does that, though).

I generally find Chrome likes to start falling over around the 3.5GB mark (as measured by chrome://memory) - larger images stop loading, pages stop rendering properly, and eventually it can end up crashing. Address space fragmentation I'm guessing.

This isn't helped much by Chrome's aggregate memory use being about 10x higher than other browsers. Takes a good 2-300 tabs in Opera to break 4GB, takes more like 20-30 in Chrome.