Quarter Life Crisis

ADC Meeting¶

1296 words on Mac OS X

This Wednesday I went to Berlin to attend one of Apple’s preview meetings. Invitations for those seem to have been sent to all ADC members, even the cheapo online members that we are. This of course meant that there’d be no extremely useful information, but never having been to such an event before, I was curious to have a look.

Information-wise, the initial assumption turned out to be pretty much correct. The things we saw were slightly more than you can see on the web, but not a lot. The most notable parts were where we got to see some code to get a better feel of how easy stuff is going to be. Also, having only a single day – even a day packed with presentations – means you can only get a rough idea of what’s happening.

Before lunch, we were given a three hour overview of the new stuff. Nothing new there. However, it became clear that the guys who did the presentations (called Phil and Etienne from the UK and France, IIRC) were different from Steve Jobs or the Apple web site in two respects. Firstly, they wouldn’t present as flawlessly and most notably have partly ugly slides with too little contrast and, more gravely, partly tiny fonts. Secondly, they were much cooler and relaxed than any of the other PR work I’ve seen before. Heck, they’d actually admit that there are things in previous or current versions of OSX that suck(ed).

While that may not seem remarkable, and the shortcomings of OS X may be obvious to interested parties, I found this fact very relaxing: You don’t have to be upset all the time because the people talking to you just ignore the problems that exist (and may affect your work). When at other times you wonder whether Apple is in ignorance or denial about those problems, those guys left the impression that they are actually aware of the problems and doing their best to solve them. Just like it’s supposed to be.

So what did I learn in those talks? We’ll have more fancy file permissions. I guess admins may rejoice now. But I couldn’t care less. Hopefully this won’t affect the stuff I use. Then there’s the 64bit thing. Another thing I – and I assume most other people – couldn’t care less about. None of the apps I use or write even gets close to the 4GB memory limit. There’ll be a better kernel which apparently is important. There’ll be enhanced Rendezvous which sounded pretty cool with access across the world. It’s not clear to me how it works, but apparently we don’t have to worry about that anyway. With discovery of services being made easier that way, it was good to see that they immediately pointed out that you’ll have to make sure your apps have secure login facilities. And then there was Automator, which I would’ve liked to see more of and which everybody immediately started thinking will be a great way to spread malware easily (which I didn’t quite get, because AppleScript can do the same, but apparently the fact that this will give non-programmers access to such tools shouldn’t be ignored). Unfortunately they didn’t say anything about the progress we’re going to see in AppleScript, which naturally is a point I would’ve liked hearing about.

Another thing presented in those talks, but in quite some detail, was Core Data, which sounds very good to me. It looks like this will essentially let you design your objects graphically and remove the need to write all those accessor methods as well as code for reading and writing. To someone who secretly dislikes programming, particularly these kinds of trivial things, this is very promising. Less code means less mistakes for me to make.

After lunch, there were more detailed talks on Core Image, Dashboard and Spotlight. I was mainly looking forward to the last one as I already had a couple of Spotlight related ideas beforehand. We could see how a plugin for Spotlight has to be written and that it is reasonably simple. We could also see how indexing is practically instantaneous, which is nice. On the not so nice side, it was made clear that the whole indexing mechanism is tied to files (at this stage at least), meaning that one file gives one set of data. That was disappointing for me as my ideas for Spotlight were to make Bibdesk (i.e. BibTeX) files available record by record, and to give access to Unicode characters by their name via UnicodeChecker. Both these things look like they can’t be done (without spilling lots of otherwise useless files over your hard drive) initially. But Apple’s people seem to be aware of the potential and try figure out ways to handle this in a good way. I hope they succeed with this. Soon.

The session about Dashboard was probably the worst of the whole event. Not only was it made clear once more that Dashboard gimmicks will be by far too easy to make (the Perversiontracker people should make sure they have a specific Dashboard section ready soon!) and that they shouldn’t have a consistent look. This also implied that the presentation was inherently boring. Putting some pictures and JavaScript together isn’t rocket science. Quite the opposite, rather. Presentations about it aren’t inspiring. The only important thing that was said was that the gimmicks shouldn’t be just gimmicks. Nothing fancy or complex. They should be mere additions to applications or status displays. I guess it took people about 0,1 seconds to forget that essential bit of information. Another thing that I’m sceptical about is that the technology allows people to access system libraries in certain situations, e.g. the address book demo you can see on Apple’s pages: I hope this will be a reasonably secure thing.

And the session about Core Image was quite a surprise for me personally. I hadn’t paid a lot of attention to that topic earlier, because it’s quite clear that my old – pre Quartz Extreme, even – graphics chip won’t be able to use it anyway. But seeing the demos gave me two impressions. First, I’m not sure this will be too useful for the majority of people – sure, the guys sitting next to me who were into video performances started salivating, but how long will it take for mainstream graphics apps to use this rather than their own technologies?

Second, this is pretty darn cool. Using the apparently huge and mostly idling processing power of a graphics card to do computations is a good idea. Generating a framework that enables people to write filter plugins with minimal effort and without having to care about the graphics chips that are used sounds like it needs a lot of thought and work. And being able to use this resource not only for on-screen drawing but also for any other graphics operation is very powerful. I like it because it finally makes use of the modern graphics cards for things other than gaming. I’d even be tempted to write a little plugin, just to play around with it.

To summarise – Core Image looks like it may be the coolest technology, Core Data may be the most work-saving one and Spotlight seems to be conservatively (and thus hopefully workingly) planned but may need some more time to live up to its potential (add to that that I’m still not entirely convinced that you can actually manage your files with Spotlight exclusively).

So was it worth it? I’m not sure. Information-wise, there wasn’t too much. Just the code examples and demos went beyond what you can read on the internet. And admittedly those don’t help us a lot in these pre-release days.

December 16, 2004, 21:08

Tagged as Mac OS X.