Seashore

Seashore is an open source image editing program for Macintosh. It’s small and sleek. From the look of it, you’d think it was a very limited, simplistic program, and it certainly doesn’t pretend to have anything like the capabilities of Photoshop or the GIMP. But it has a surprising number of powerful abilities — painting in textures, multiple layers with many different blend modes (multiply, diffusion, blah blah), full alpha channel support, SVG import, and other things you wouldn’t expect out of a rinkydink paint program. It gets all that good stuff by leveraging the GIMP and the Mac’s native goodness — the guts are all GIMP code, the interface is all Aqua.

Unlike Photoshop, it’s Free Software.

Unlike the GIMP, it’s a native Cocoa application — no need for X11 to be installed, and it looks and acts like a proper Mac app should.

Unlike both, it’s super small — the entire app, installed, with optional extra SVG import added in, is about 8 megs on disk. The basic installer download is only 3.3 megs.

Also unlike both, it’s very new and probably hides some bugs. I’m gonna play with it a bit and report back. It looks like a very promising piece of Mac software.

Hatin’ on Ubuntu

Via Alterslash/slashdot, apparently the elder lords of Debian (including the Ian in Deb-Ian, but not the Deb apparently) are annoyed by Ubuntu because Ubuntu’s software, based on Debian’s unstable version, isn’t very compatible with Debian’s almost-stable-but-not-quite-there-still version.

I use Ubuntu now on my linux box. I used to use Debian exclusively (after having gone through a “try every distribution under the sun” phase where I’d always end up back with Debian cause it was the best thing I could find). But I stopped recommending Debian to friends about a year or so ago when I tried to help two friends install Debian’s stable version on a desktop and server machine of theirs and realized how bad it had become, at least for people like me and my friends.

The then-current stable version of Debian, Woody — which by coincidence is also the now-current stable version of Debian — had crappy-ass hardware detection, worse than any other linux distribution, I kid you not. X11 configuration was a total nightmare for the friend who wanted a desktop OS, and we had a dead rotten time trying to find drivers for the network card for the friend who wanted a server os. After those two installs I just couldn’t honestly say “hey, you should try Linux” to a friend anymore. Maybe to an enemy.

When I stopped recommending Debian it wasn’t because I’d discovered Ubuntu and liked it, it was because I stopped liking Debian and no longer had any Linux distribution I could confidently recommend. As far as I could tell, there were distros that installed well and had recent software but were hell to maintain and upgrade (anything but Debian), and one distro which was hell to install and had nasty old software but was pleasant to maintain and upgrade (Debian). I couldn’t happily recommend either of those. My linux advocacy took a dive. I drowned my sorrows in the beautiful Aqua interface of OS X and tried not to think about the proprietary software.

A few months ago I decided to give Ubuntu a shot, having heard some of the good buzz about it. As far as I’m concerned it’s everything that Debian hasn’t been for years and should have been. Yes, it’s dependent on Debian, and it is able to be so good because of the work Debian volunteers put into things. But why isn’t *Debian* able to be so good because of the work Debian volunteers put into things? Well, it’s probably because they support about 50 architectures and won’t go forward with a release till they can get an archaic version of GNOME to compile on Debian-SomeStupidProcessorIveNeverHeardOfAndOnlySixPeopleUse-64 (DEC). And real Debian guru types don’t give a crap about people less geeky and technically competent than themselves, because that takes time away from flaming each other in political wars on the dev lists. Not that I’m bitter. But if you want to see everything that’s loathsome about Linux people, try asking a question in a debian IRC channel. I only did it once. It was enough.

OK, so whatever, that’s fine, Debian has its goals and producing an operating system that I can recommend to friends and hope to keep them as friends is not one of them. That’s all good. But if somebody’s going to come around and actually produce such an OS (and Ubuntu is such an OS, usability bugs notwithstanding), then do the Debian developers have to come out hatin’ on it?

Yeah, of course they do.

I love Debian. And that’s why I hate Debian.

OS X Emacs

I’ve been using Carbon Emacs for my coding. It’s the official Mac branch of Emacs.

Just noticed AquaMacs, which is similar but “enhanced” to make it more friendly to general Mac users.

I tried it and my head exploded, because they remapped old-fashioned keys I use constantly (control-v to mean ‘scroll down a screen’). It may be better for new users but it was not good for me. Oh, and the proportionally spaced font — very pretty, but useless to me for coding in Perl, which is what I do in Emacs all day. Disaster. And new windows popping up all over the place all the time!

On the flipside, I also discovered Emacs-On-Aqua, which is based on the old NextStep port of Emacs, so it’s Cocoa-based, not Carbon-based.

Eh, it’s OK. I think I’ll just go with a recent build of Carbon Emacs.

Supporting Your Web Page with Ad Revenue Less Tenable Than Ever

Remember when Google had no ads? Want to use Google without ever seeing an ad again? Just use Firefox and Greasemonkey and Customize Google.

In the war between web advertisers and web browsers who want to avoid ads, Greasemonkey is the freaking Manhattan Project. I installed it and saw what it could do and whispered, “I am become Death, the destroyer of ad revenue.”

So how can people ever afford to publish content on the web, if not via ad revenue?

Well… one way is by giving up control of the content. Bandwidth is expensive, if it’s bandwidth on a server you control. Bandwidth is cheap if you give up control and let people share in the task of distributing your stuff. Peer to peer technologies let you publish like crazy. Open source projects are never lacking for a dozen friendly mirror sites for their code. You got creative commons licensed content? No prob. Publish it on OurMedia or Archive.org. They’ll give you the storage. But you have to give up some control. Not all control. Just enough to make things easy on the sharers, to give them a stake, to give them some rights with regards to the stuff they’re helping you distribute.

That’s one way to do it anyway. There may be other ways. But the “Ad Revenue” trick is not seeming like a very viable option anymore, in light of the power of things like Greasemonkey.