Cocoa apps must be able to load libraries developed by other authors if they are to prove optimally useful. Developers need to be aware of the various libraries that users might be adding to their setups and make authors who aren’t using the Cocoa APIs appropriately aware of their errors. It kind of needs to be a cooperative venture, rather than each developer viewing their application as some kind of island that can exist on its own.

Again, I know it’s tempting to view badly written code that sends users running to you with complaints as the fault of APIs like Input Manager (which is going away anyway but will likely be replaced with something else), APE, or SIMBL, but the problem is the author of the badly written code… not the API itself.

As a user, I find the ability to add functionality to all of my Cocoa apps a huge advantage over the prevalent “silo app” approach. Likewise, I’m a very big fan of APE and have never had any problem running them… it’s quite easy to disable a given APE in a particular app if it becomes a problem, and it’s easy to tell users how to do that. Unsanity has built a beautiful API for adding functionality to Mac OS X that Apple has left out, and there are scores of useful APEs out there from third parties… nearly all of them free (except those from Unsanity).

Just wanted to make that point. Oh, and the reason I happened upon your blog is that I’ve read that Apple is including a new API for method swizzling in Leopard, and I ended up here through a Google search.

Cheers,
Leland
Musings from Mars

And thanks for the Feeder comments too.

We’ve had the same experiences with Camino, specifically among bug reports targeted at 1.1 beta. One of our developers recently wrote a post that parallels yours, which you can read at http://www.escapedthoughts.com/weblog/camino/P070227caminohacks.writeback.

By the way, I’m a happy user of Feeder; It’s another example a well polished application that can be engineered by a single, independent developer working with awesome tools and frameworks.

Multitouch is too important to be confined to a phone. Think stylus-free, tabletlike MacBook that is indeed open to developers.

I don’t know of any planned events in the Spring. I see an Apple Event popping up sometime.

Word is that the iPhone will be closed for now. I don’t see Apple changing their mind in the near future, just because of the work they would have to do to open that platform, but I expect it will happen in due course.

I think the iPod taught Apple that the mass consumer consciousness was where the real money was. And that’s what Steve is betting the iPhone is going to be, and it does feel like the first truly 21st century device, and it’ll go like hotcakes when available.

Isn’t there a Spring developer’s meet? IF the iPhone is open for developers, Steve will want to focus on that for that conference. I think he’ll launch Leopard in advance of that.

I think Steve is going to have a Here’s How We See Things kind of presentation.

Apple have staked out a set of parameters that they think defines the digital media sector… and by and large they form a structure for business which works and they should be rightly proud of what they’ve achieved. They’ve let Flash take the scalability/interactive lead (by a long chalk) over Quicktime, but other than that black mark, they are the only credible market players out there.

But other than that, I’m hoping for Leopard’s release and seeing just what we have to look forward to….