So everyone in the compositing world (defined for me as fxguide.com and This Week in Media) is talking about how Apple has dropped the price of Shake to 500$ (from 2000$) and has stopped development on it. They plan to spend the next two years or so developing a next-generation compositor to replace it.
This is all fine and good for Macintosh users, but as a Linux user I see the writing on the wall: No more compositor from Apple. There is no way that Steve would develop a next generation compositor and maintain Linux compatibility. He’s going to make sure that their new software is going to fit snugly in with the rest of the Final Cut Studio, and that’s absolutely his right.
The support of Shake for Linux was only out of necessity and was half-assed to begin with (not even basic /dev/dsp sound output!) but with this announcement there’s no question about where the Linux graphics market is headed. It’s Autodesk on the ultra-high end or nothing.
Upgraded WordPress to 2.0. Thankfully since I’m not on planet gnome I won’t spam tons of people with revamped RSS GUIDs.
I popped my first tire today. I’ve had pretty good luck with bicycle tires over my lifetime, I don’t think I’d popped a single one until last Thursday. I was cruising down Summer St toward Davis Square when suddenly I heard a “whack whack whack” noise coming from the rear tire. I thought I had lost a spoke, but when I looked closer…
Not only did I perfectly hit a small rusty screw in the middle of a road, but I did it in such a way that the pressure from the tire formed an airtight seal around it.
Mike helped me patch the tire on Saturday, which was much more painless than I thought it would be. Thanks Mike!
Real life officially becomes Myst for a moment.
For a long time I’ve been trying to use Skype as a replacement for my cell phone, which gets crappy service in my apartment. SkypeOut sounds like a great idea, but in practice it was unusable. People on the other end of the line always complained that I was too quiet, and the voice would drop out so much it was unusable.
In contrast, Gizmo works perfectly. I don’t even remember where it comes from, but gizmo is a very nice GTK2 app that provides features very similar to skype. Call-out, call-in, all that good stuff. The real difference comes in actually using it. Gizmo worked perfectly the first time, and I was able to call out right away with near-perfect fidelity. I actually had to turn the mic down a little!
I’m probably not going to get a phone number for it, even though it has 617 numbers, because it’s really difficult to make sure the PC is always ready to take a phone call. But, it will be very useful for dialling out to order delivery, call work, call my parents, or what have you.
Why doesn’t this exist?
I have a mobile phone as my only phone, but it doesn’t get very good reception in my apartment. I am planning to get a bluetooth handsfree device so I can put the phone in the window where it gets good reception and carry the handsfree anywhere in the apartment and still get a signal.
That got me thinking. Why not create a device that has bluetooth and an rj11 connector on it? You set up the converter so whenever it sees the mobile phone in range it becomes a “handsfree device” connected to it. It then converts all of the handsfree functions — dialling, ringing, caller id, voice — to regular telephone spec. It could even fake a dial tone when it sees the mobile, and play silence when it’s not found. You then connect any cheap-ass phone to the rj11 and you instantly have a “home phone.”
I’ve done some searching on google, and the only devices I can find are converters to go from an existing landline to a bluetooth headset. What ever happened to marketing to us hip singles who don’t want a landline? This seems like such an obvious use-case.
Anyway, if there’s anyone out there who knows electrical engineering or has access to a bluetooth chip, let me know. We could create a new product and get bought by jabra or something!
So I had a good time at the Gnome Summit this year. I got up the courage to get on stage and show off PenguinTV, and despite wacky vga-output problems I was able to demonstrate the UI and functionality. It got a good reception, and a lot of people came up to me afterward with ideas for extended functionality, including support for stealing streaming real-audio sources, integration with Banshee, and integration with GStreamer. I also got my first patch by email, which is great. The idea that someone went through my messy code and learned enough to work with it (without complaint) seems a little unreal.
This positive feedback has motivated me to keep pushing penguintv toward 1.0, including setting up a CVS repository on sourceforge and getting any final bugs fixed. After 1.0 I’ll work on code cleanup and start looking at C# and Banshee. Fun times ahead!
Irony: The best way I’ve found to discover which filesharing networks I should use is to look at industry reports about what filesharing networks everyone is using. I don’t have any friends that are in to the whole downloading thing, so I don’t have anyone to ask directly. A recent slideshow mentioned edonkey2000 was incredibly popular. I had heard about it years ago but it never really registered as something worth checking out. I did, and it seems to work quite well!
By default wordpress only includes audio and video mimetypes when it reads your post to decide if there are any urls it should enclose. This means that if you want to enclose bittorrent in wordpress, it won’t do it without a little hacking.
in do_enclose, around line 775, there’s the line:
$allowed_types = array( ‘video’, ‘audio’ );
$allowed_types = array( ‘video’, ‘audio’ , ‘application’ );
Note that any link to some application/* will become enclosed, so this is an awful, awful hack.
ignore this, test
test was successful. When I put a url to a media file in wordpress, it puts enclosure information in the RSS feed. I’m not sure how it’s doing its type-detection, but it works well enough for now.