David Hobby, the author of the Strobist blog, is taking a leave of absence from his “professional” job in order to spend more time with his family, and try to make the most of the opportunities presented to him by the amazing uptake of Strobist. I am glad to see that he’s worked out an arrangement that lets him see if Strobist has wings. I, and many others have learned a huge amount from David, and I am sure that this will be an exciting year for him (and us, by extension).
If you aren’t following the Lighting 102 series, you really owe it to yourself to do so. In the second unit, David filled in the one area that Zack Arias’s OneLight workshop didn’t cover that well, the impact of flash to subject distance. As always, no only did David show how it worked, but he explained it way better than just regurgitating the inverse square law: “Light has depth of field.” Great explanation.
I haven’t shot the assignments yet, having been on an “off the grid” vacation for a week (and now trying to dig out after that), but I am anxious to do so. We shot a variation of the first assignment on angle, at the OneLight, so I don’t feel quite so behind. Here’s one of the OneLight angle shots:
People working in open source have limited opportunities to meet each other in person. My own experience is that meeting someone in person, even if it is only once, can be a help in working with them in the virtual world. The Dopplr service is a social networking application oriented towards people who travel. I’m already using it to find out who is going to at OSCON and ApacheCon US later this year. Meeting folks from various open source communities in person has been an enriching experience for me, and I’m glad to have help at making that easier.
Let me know if you want a Dopplr invite.
Turns out that there is a little more technical information for us poor folks that aren’t at WWDC. So on the topics that I wanted more info on:
- Multicore – fixes to the scheduler, including processor affinity, multithreaded networking stack, more threading in Mail.app and Spotlight, NSOperation and NSOperationQueue, and OpenMPI
- 64 bit – 64 bit Java, MySQL and Apache.
- DTrace / XRay – Dtrace-ified Ruby 1.8.6, Python 2.5, Java, and Perl. XRay alone would justify the $129 upgrade.
There’s also a bit of news on ZFS [via Simon Phipps].
I guess I feel better now. Now if only someone would tell me that Spotlight is no longer crawling…
I have to say that I was pretty underwhelmed by the WWDC Stevenote today. Between the combination of last year’s keynote and the promise of super cool, top secret features, I was expecting a bit more than what actually took place. Let’s look at the ten features that were shown:
- New desktop – this is pretty, and I will appreciate a unified window look, but I’m doing just fine today. Unlike the rest of the world, apparently, I am using the plain old blue apple background, so a desktop that is friendly to my digital photographs just isn’t making me that excited, although in theory it should. Stacks the only thing that look like a new feature, and while they look cool, I’m not sure that I will use them that much.
- New finder – I’d like a new finder – so much so that I already use PathFinder on Tiger. The sidebar searches are nice — assuming that Spotlight actually has something like decent performance now. CoverFlow looks very pretty, but the only time that I can see using it is on collections of images or PDF’s. Looking at my Applications directory using CoverFlow isn’t really very exciting. The file sharing stuff would be good if I actually shared any file with people, but mostly we’ve been doing just fine using the existing, albeit clumsy networking.
- Quicklook – This is nice, and it’s nice that enabling quicklook support enables other things, like iChat Theater support. But I’d rather have working (i.e. performant) Spotlight.
- 64-bit – A genuine advance. Too bad it missed the window so that we could have 64 bit Photoshop
- Core Animation – I’m just not an eye candy guy. There wasn’t really any indication of how much effort is required to build something like the video wall that was demonstrated. Without a look at the code, it’s hard to know how impressed to be by this. It does look like Core Animation is an enabler for lots of what was shown.
- Boot Camp – I understand the rationale for Boot Camp, but continue to find the idea of rebooting into Windows a non starter. Give me VMWare or Parallels any day.
- Spaces – Virtual desktop management is so 1990’s.
- Dashboard – more specifically, a movie widget, and webclip. I can’t remember if we saw webclip last year, but it looked impressive. I am sure that people using Dashboard will find this a boon. Personally, I turned off all the Dashboard triggers because Dashboard is such a resource hog.
- iChat – It’s great that there is all the iChat eye candy. I might actually use the iChat theater features, but the big impediment to iChat is that I don’t run it. I run Adium because I need to talk to people not on AIM. Also, my experience has been that firewall tunneling for iChat doesn’t work very well, so it’s anybody’s guess as to whether video or audio chat will work on a given day. I’m basically using Skype for those features now. If Apple improved the firewall tunneling, then this might be interesting. I do give points for leveraging Quicklook to get things into iChat Theater
- Time Machine – well, okay, but we saw this last year.
Note that 64 bit support, Core Animation, Boot Camp, Dashboard, iChat, and Time Machine all appeared last year. That leaves the new desktop, finder and Quicklook as the Top Secret features that we’ve waited an additional year for. The things that I really want from Leopard are probably still buried in those NDA’ed sessions. I want my OS to stop leaking VM. I want XRay and ZFS. I want to know if the processor thread affinity problems have been fixed. I want garbage collected Objective-C – actually I want that one for the ISV’s. Perhaps most of all, I want Spotlight to actually be usable for me.
I”m not that excited about Safari for Windows. Since I do “Web 2.0 AJAX apps” for a living now, the last thing that I need is to add another browser/platform combination to the test matrix. It’s great that Safari is so fast on Windows, but that doesn’t really help me much. It kind of bothered me that when Steve talked about increasing Safari’s share, he overlaid the share that currently belongs to Firefox and “other” browsers. That did not make me warm and fuzzy. Nice that tabbed browsing is improving in Safari, but why couldn’t they do something slightly cooler like using stacks to organize the tabs?
As for the sweet SDKless iPhone development story? Well, it’s nice to know the whole Safari engine is in there. Basically what they said is: Hey you can write a dashboard widget that is deployed from the web. Ok. You can build some decent apps that way. I still think that a regular SDK is going to need to happen some day. In the meantime the problems that I have with iPhone remain, which means that I’m likely to end up with something like this. But that’s for another post, sometime in the future.