"Well, I'll give him another twenty minutes; but that's it!"

It isn't necessary to have a square number size, it just means that you are forced out of having a puzzle comprising row-, column- and square-based "house"s when you do. Various polyomino forms have been done, from plain rectangular to the the "Squiggly" variety at dailysudoku.co.uk

Being reasonably competent at the various 9x9 forms (I seek puzzles online because the only puzzle I've found in the national press that I can't solve is the supposed-world's hardest [solution]), I can't see that there's any great complication possible in one of 4x4 size, although I'd imagine it was precisely the point to have *some* complexity but not *lots*.

You don't say anything about the model or age of this laptop, but if it's anything like the Compaq I had then its install disk will want to make multiple partitions (some of which will be specially marked) on the disk and get upset if there aren't enough free slots or decides it doesn't dare risk trashing a existing system if it finds one (especially likely since you've currently got one that clearly isn't Windows). This was a while ago, but something roughly analogous is true of the more-recent EeePC 700 series where Linux versions have (AIUI) a system, user, and and additional DOS-readable partition for handling BIOS updates.

If you're lucky you'll find that the Linux install you completed recognised a valid partition table and backed up the relevant disk sector, complete with any magic bytes that may or may not have been originally there [Google can advise if these are necessary!] compared to what's in the standard Linux MBR. If not, you couldn't do worse than 'dd' a sector's worth of zero bytes over the existing partition table and see if your reinstall disk starts to behave as a result of blanking it completely, putting Linux back again if you really have to.

Better still:

• Sze's announcement (greater longevity version [possibly])
• Previous 1 quadrillionth (lone-)digit record
• New 5 trillion consecutive digits record (additional details

You don't think calling the implementation "DistBbp" suggests they did?

Lazy, astigmatic, squinted, and very happy to have seen Avatar's 3D effect work :)

Background: I get kind of travel sick with FPS games and can't do Magic Eye at all, but have successfully observed 3D film presentations before (a six-minute ride at the National Space Centre which I felt only kind-of-worked but looking back I was probably suffering the FPS effect too much to gauge it effectively).

I had at least been firmly forewarned that staring at the out-of-focus things to try to "fix" them would disconcert, so I knew what to expect to an extent. One of my 20/20 friends had nothing of this, however, and he got the headaches something rotten! Go figure.

Mod parent insightful. I actually worked somewhere this was reasonably sensible; okay, the granularity of what we actually had was "hours", "mornings", "days", "weeks", "fortnights", "months" - (pretty much doubling again, looking back at it written down), but there really were so many meetings and other overheads that it made sense (that's probably what killed us, but I digress).

...and then sometimes management did the same again, which was nice. One demo I delivered in the predicted "few days tops" (including several additional features I anticipated followup requests for) which "we told [the customer] we'd need three months". The graphics design guy got plenty of time to beautify it and everyone was *very* happy :)

Sure it has. There's even a "parody" of it.

If you believe that, I've got a spray that keeps elephants off your garden to sell you!

At first people ignored that, then they laughed at it, and then they fou^H^H^Hfried it

Mod parent funny
...and the article he posted.

Nested comments maybe? If you hadn't noticed the outer comment markers, you'd get something that "didn't" work with the inner markers in place and "did" work without.

(In PHP, attempting to nest comments leads to the second '/*' being considered as enveloped by the first '/*' and the first '*/', leading to the second '*/' being flagged as erroneous, and the interpreter bailing out)

// writing my first bit of non-trivial PHP _today_

(Bind-)mount over ~/.gnupg? I do a similar thing on my EeePC for ~/.mozilla, although in my case it's to keep the browser cache off the internal drive when I've got the resources to do so. Whether it's possible to keep the command out of any history files made by $SHELL in an invisible/innocuous manner is another matter, of course (and not necessary in my scenario).

...at which point we could back up whether Commandos do actually "go Commando" with real statistics!

// Suddenly not at all curious about whether it's true...