Complete: Coming home from town after having a few drinks at the Brit-Pop party thingy they did at Gromo Café and an extra Vodka Ahoi at the Sechs Millionen Dollar Club, I saw that Philipp's car was back in the parking lot and he had returned from Portugal. So our flat is complete once again. This also meant, we could finalise the date for the party we are going to have. I already came up with an invitation and will try what kind of paper we're going to use tomorrow. It's going to be huge – it's supposed to, anyway.
The Poincaré Conjecture may have been proved. I heard this a few days ago and it's already made its way into mainstream media. That text isn't very clear though (which should be partly caused by the conjecture requiring a little bit of specialist language to state – but not too much) and it seems to heavily borrow imagery from the Clay Institute's description of the problem. From what I've heard, the proof is quite technical and hard to understand and actually a more general statement is proved. But even that seems to have made it into semi-mainstream media.
Import: Also in Die Zeit, articles mentioning how in the name of
homeland security and
patriotism, getting into the U.S. can now be tricky for researchers if some dumbass bureaucrat thinks you might be from a
rogue country and your research could be dangerous, say because it contains words like Laser or atom. Probably not exceedingly clever for a country whose research and business has heavily relied on importing foreigners and welcoming them in the past.
Brackets in Objective-C is the alleged topic of Erik Barzeski's post. Except they aren't. What he – and I couldn't agree more – likes, is the way how you send messages and pass multiple quasi-named parameters in Objective-C. There are two things I'd like to comment:
Firstly, this isn't really about the brackets. While the brackets may be a consequence of using this kind of notation, they are the least convenient part of it. Why? I frequently find myself sending a message to an object and getting an object in return. Now, if I want to send another message to the resulting object, instead of just adding it at the back à la
object.message1(a,b).message2(c), I'll have to go back to the beginning of the line to add a
[ before being able to do this. That's not very convenient.
I tend to bump into this inconvenience regularly when under-estimating the number of the brackets I need at the beginning of the line or when wanting to call 'one more method' in bad programming mode. But also in perfectly legitimate situations, say when having to change statement from and convenience method
[object object] to hands-on intialisation
[[[object alloc] init] autorelease], the inconvenience of having to add brackets at the beginning of the line occurs.
Secondly, I guess it might be nicer to have properly named parameters rather than the 'message names with colons' in Objective-C. I think that using
[object setSize height:10.0 width:10.0] which is equivalent to
[object setSize width:10.0 height:10.0] would be more logical as it distinguishes the message from the parameters. And this would be easier on my memory. I guess there are good reasons for not doing this, say, performance-wise. It's probably all about compromising...
Oh, I still like the bracket notation. It's nice to read and very explicit in structuring everything. And it makes it easy to select complete sub-statements using a double-click in ProjectBuilder.
Conspiracy: Also via NSLog();: Other people sharing my suspicion that there is some kind of Mac conspiracy going on in the world of content creation. Of course it could just be Mac users being too narrow minded to look elsewhere...