|
|
We had an interesting little technical issue here today that was caused by a completely innocent change.
Sarah was rearranging things upstairs and removed an old PowerRF transmitter that was used by a wireless phone. She pulled the device and then plugged the phone straight into the wall removing a line filter from the equation. Shortly afterward, as we later pieced things together, we noticed significant network issues.
After a few hours and a trip out to dinner I finally called tech support to see what help I might be able to get.
After getting through the screening process, the tech informed me that my modem was not synced. Hmm, the light was on. Restart the router. It came up and synced, then lost sync several times in a row. Eventually it synced but using a tool on his side, he could see that there was a lot of noise on the line.
He asked if we had added anything to our phone network at home. My initial answer was no, but as we talked I remembered seeing the old device sitting on the bedside table upstairs. I mentioned that we had removed something, thanked him for his time and went upstairs to investigate. Sure enough, the device had a filter on it and it was completely removed. Restoring everything to it's state before the change restored network operation (after another router power cycle).
Tomorrow I think I'll go inventory everything making a connection and ensure that they've all got decent filters on them.
It's been quite a busy week.
In what little spare time I've had, I have been investigating alternatives to the Keyspan USB Serial Adapter so that I can do old fashioned serial communications on a new fangled Mac.
Mostly, it seems to boil down to a slew of generic devices based on the Prolific Technology PL-2303 USB to RS-232 Bridge Controller (I leave it to you to find your way through the maze). There is a lot of documentation on the tech site, but it's quite difficult to find anything directly. Driver writers want the file ds_pl2303_v14.pdf (or later) and maybe the linux driver ld_pl2303_v0213.zip.
The one I've seen several places is is the IOGEAR GUC232A. Some GPS units also use the 2303 part. Like the Deluo GPS if you choose the USB model (also see Unlikely's Deluo GPS article). Deluo also sells their Serial to USB converter separately at a low price.
The problem is that driver support is really spotty. There are lots of people looking for them. Hopefully, OS level device support will arrive. IOGEAR has a MacOS X driver (which comes complete with a warning about the quality of the 1.04 driver down inside the installer package) and I found a modified 1.06 driver in an unlikely place.
I'd originally wanted the Keyspan, but I can't find a Keyspan device anywhere locally (I was annoyed when I found that the Apple store didn't carry it anymore). The IOGEAR device is easily available, and if it doesn't work, I can easily get it back.
MacOS X Hints has a good article about patching the 1.06 driver which also contains information about updating the plist for other devices which has proved useful to some people.
So, why would you need one? Most people probably don't, but I'm slowly collecting a whole box of USB/FireWire dongles and I really do need this one.
Right now, I need to talk to the admin console on a Sparc and I'd like to see how it works with the Xserve. And of course, things just keep coming up which like serial connections and for the last three years I've been stuck.
LaCie's Bigger Disk is big alright, in both size (one terabyte unformatted) and form factor.
I was thinking about how far we've come in the last twenty years in terms of storage space and found this rather interesting historical perspective: Cost of Hard Drive Space.
Twenty years ago, a ten megabyte hark disk (which today would be large enough to almost hold a couple pictures from one of the newer digital cameras) cost roughly $1,500 (give or take). It wasn't until the following year that First Class Peripherals screwed everyone up by offering a 10 M drive (the Sider) for $700 (yeah, it was slow, but compared to the options of the day, it was a big win for people who needed *cough* mass storage — oh how things do change).
At a list price of $1,200, the biggest Bigger Disk is a 10,000:1 (plus a bit) improvement in cost per byte over that time which is quite amazing when you think about it.
We still have the same lurking problems (or I do anyway). How do you back it all up?
For a while in the early nineties, external storage had caught up with fixed media and it was relatively inexpensive to backup normal desktop system sized hard disks to magnetic optical and SyQuest cartridges. Tape backup finally seemed to get rolling with DLT in the mid nineties and while the cost was pretty steep for casual users, small offices could reliably back up somewhat large data sets (less than 40 G).
And since then? Not much as far as I can tell. VXA looked like it might useful, but the price point never moved and it's a single vendor technology. AIT drives which can handle about 70 G are available under $1,000, and there does seem to be downward price movement, but compared to hard disk prices, it's glacial.
Setting up home RAID arrays seems to be all the rage these days, because the cost of fixed media keeps plummeting. That works fine until you lose two disks in the array at the same time as happened recently to a friend. You still need a real backup somewhere and I just don't see how to do that effectively. You could just use mirroring I suppose.
Perhaps the answer is to split things up. A small data set that is extremely important would be backed up using more expensive techniques which ensure some sort of longevity. The larger set of temporary data (mostly media, like the 300 M movie animation of the MER landing) would be on big, semi disposable disks. If I lose it, oh well; I got it from the net, I can go find it again.
I still can't quite figure out the right approach, but I know from experience that it's going to get bigger.
Jason 
(Mr. Blogging.la) has a couple fabulous pictures up from a whale watching trip over the weekend. Even though I don't care much seagulls, I have to admit that he's captured a perfect shot of a pair. I'm looking forward to seeing the rest of the pictures when he publishes them.
It also remind me that I need to get off my butt and take the kids out on one of these trips while there is still some time.
When we arrived in Anaheim last week, the NAMM conference was just wrapping up and the hotel was crawling with music industry folks. Don't consider this in any way scientific, but there were Powerbooks everywhere (including the room next to mine, a suite where the adjoining door had been left open for some reason — I noticed the Mac and a lot of other gear before closing the door). TidBITS has more on Apple at NAMM.
That story led me to a TidBITS Talk thread running for a couple weeks now on ripping vinyl that's worth a read. I really wish you could just read the whole thing at once using something like mode=flat (if it's there, it eluded me). For instance, Matt Neuburg describes his rather elaborate process with quite a bit of detail (although he leaves out how to search for clicks and pops in Amadeus). Hmm...
Coincidentally, an update notice for version 3.7.1 of Amadeus II showed up in my mailbox this week. I was at 3.5.4, so umm, it was way past time to update. One of the replies in the TidBITS thread pointed to series on ATPM last year called Got Vinyl? See part 1, part 2 and part 3. Again, worth reading if you are interested in this sort of thing.
Finally, Tomoharu Nishino mentioned the M-Audio Transit USB and "Audiophile" gear (USB and FireWire). Eventually we'll need to get Jon some MIDI gear, but today is not that day (I had a little bit of sticker shock). I'll stick with the iMic for now.
I'm out of here and off to a company gathering near Disney. Since there will be all day meetings and the inevitable discussions after the meetings, I probably won't be posting much during the coming week (as if I've been doing much lately... oh well).
Adam got his voter registration card in the mail on Friday. I'm not sure if he'll be able to register in time for the March elections but he can certainly participate in the main election in November.
I'm feeling just a little bit older this weekend.
It has been 10 years since the Northridge earthquake struck Southern California at 4:30 AM. Unlike most people I was still awake at that time, although Sarah and the boys were sound asleep. I had been working on some code for Norton Utilities 3.0 that needed to be finished and then had begun tinkering on something else (I'm not sure what at this point).
When the earthquake first hit I was sitting at my desk. I had been through several earthquakes before, so I initially didn't think anything about it. However as the quake continued to roll and the world continued to shake more violently around me I realize that this was much more serious than I had initially anticipated. By the time I realized how serious things were I ran from my office winding up in the hallway leading to the bedrooms. Just as I cleared the door to my office there was a loud crash behind me as several bookshelves fell over and the door to my office was slammed shut from inside (which later required some major work to fix).
In the hallway I met my wife and sons as they came piling out of their rooms. Immediately afterward a large crash was heard outside, a bright light flashed and the house went dark. We ran towards the front of the house and bolted out the front door. Luckily, it wasn't too cold and of course we weren't alone.
Most of the neighborhood was out on the street in front of their homes. Everyone on the street was asking the same questions. Can you believe it? Are you OK? Is your house OK? Everyone was babbling at the same time about how powerful the shock had been and how long it had lasted and was amazed by the fact that all of our homes appeared to still be standing.
After getting everyone together and making sure we were all OK, I proceeded to check out the house. The gas line looked fine, and despite poor bracing the water heater was in good shape (it had moved some but not torn free). We went digging around to find a portable radio to listen to KNX and find out what was happening and I started across the street talk to our neighbor. Everything was good with their home but he was concerned about his elderly mother. She was pretty much bed ridden and was in no shape to leave the house after the earthquake. He and I stood in the middle of Avenida de Los Arboles talking for about 15 minutes. We never saw a car. Yes, it was nearly 5 in the morning, but it was weird. With no electricity and no cars, it was very quietly and lonely out there.
By the time I finished chatting with our neighbor, Sarah had brought the boys back inside, found some candles and was setting things up for us to listen to the radio. We spent the rest of the dark hours of the night sitting around three or four candles, listening to radio reports, running outside each time the earth started rumbling again.
Our electricity was out, the phone was not working, but we did have running water. So we filled the tub with as much water as it would hold. Once daylight arrived we had a chance then to check out the foundation and walls of our home. No major problems there, although we had a pretty big crack in front of the house (later characterized as 'normal'). We had a number of smashed glasses and crockery, but no major losses in the kitchen. Amazingly the China that I'd brought back from Hong Kong was still in one piece. Once we had phone service again, I started making some phone calls. We let the folks back east know that we were alive and uninjured. Later in the day I was finally able to get in touch with some friends from work and make sure that they were OK. Our friends Stuart and Connie, over in Simi Valley, had lost their gas main when the water heater was ripped out its mount but were otherwise undamaged.
During the day we picked up many reports from around the city of major damage. The 10, 14, an 101 in Newbury Park all had major road closing damage of some kind (although the 101 was quickly repaired and opened). We also learned of the terrible event in Northridge where an apartment building had collapsed and had buried 16 people inside. It was late in the day Monday when power was finally restored and we began to see what had happened around SoCal.
The Symantec offices of the Norton Group were (and still are) located in Santa Monica in what had become the MGM Plaza. MGM itself was located on the 4th floor of the same building, and a water main burst during the earthquake. Our offices on the second floor were covered in 4 inches of water while most of the monitors and many systems had fallen to the floor along with papers, CDs and various typical office materials. They sat in puddles for at least a couple days and most were later replaced. It was such a mess that it would be two weeks before we get back into our offices. Luckily for most of us, we were already accustomed to working remotely and used the email to arrange one-on-one telephone meetings during the next couple of weeks as we got back up to speed.
The most interesting aspect of the earthquake, was the aftershocks. Both from a physical and a psychological aspect. It was weeks before I slept through the night again. Every little shake woke me up. Was this another big one or just another aftershock? I'd wait until it passed, getting more nervous if went on for more than a few seconds. Afterward, I would be ill at ease and not really ready to go back to sleep. We talked to the kids a lot about it. From talking to others, they all had much the same response.
Another end result of the earthquake, was my taking the tour of the subterranean parking in Santa Monica. I would never again trust that parking lot. They were too many cracks in the foundation posts holding up the parking lot and the building itself. I still wonder from time to time about that place (and still have friends working there).
Ultimately, even though there wasn't much damage to myself or to people I knew, the Northridge earthquake profoundly changed me. I'm much more sensitive today when the world starts shaking.
John Berresford writing for the Harvard Program on Information Resources Policy has released a draft of How Government Can Bring New Communications to All Americans: Six Lessons from History Discovered by a Libertarian.
It makes for fascinating reading even if you reject the conclusions of the author. [via IP]
It looks as if Carlin's list of seven forbidden words is up for legislative expansion this year.
Bonus link (not safe in mixed company).
I could have sworn that we had some serious issues to deal with this year. Guess not.
I was in a conversation at work yesterday and mentioned seeing something recently about LZSS, which I thought was interesting. No there knew anything about it and I couldn't think of where I'd come across it.
I've been spending a lot of time wandering around Amit Singh's site because of something Bill Bumgartner had written (pre-explosion) which led me to What is Mac OS X? A quick search found that Amit had mentioned that the kernel was compressed with LZSS.
I don't know why that initially surprised me, but it did. I guess it's mostly because LZSS was largely considered a very good and fast text compressor when first described in Data Compression Via Textual Substitution. My own experience should have kept me more aware of the possibilities. When Kenjirou Okubo uploaded the original compression source to GEnie (where I found it), he included this note:
This article by H.Okumura explains various algorithms of Data Compression. The article, originally uploaded in his workshop, is posted here with his permission. Also includes three C programs illustrating lzari, lzss and lzhuf methods uploaded with permission of their authors. These are the compression schemes currently being investigated by Japanese hobbiest programmers.
Haruhiko Okumura had released a pretty good version of LZSS using Bells ideas about b-trees in the spring of 1988 (Michael Dipperstein has a very accessible description of LZSS). It had a small code and memory footprint and a very efficient decompression engine. It worked really well for textual data but was pretty iffy with binary. I tinkered a bit and moved on.
In late 1990, I was in desperate need of a way to compress a Davidson 'Blaster product to a single disk. The images were my single worst problem, so I focused there. I tried out a number of things, but the solution that worked the best was to use LZSS on PICT images.
Several things were in my favor. We were using 8 bit images. We used a lot of white around the edges because many images were used to create their own masks on the fly. PICT 2 encoding was mostly straight forward (scan line RLE with a few QuickDraw specific wrinkles as I recall).
My first pass on the raw image data was less than impressive. When I later tried compressing the already compressed PICT things changed a bit. I was getting nearly 50% compression for most of the images. I got better compression with other algorithms but was never able to achieve the same kind of performance.
After deciding that LZSS would work quite well, I shoved the original and compressed sizes at the front of the new compressed PICT resources as a simple verification method and a way of preallocating the result buffer during a decode. I also moved away from the stdlib calls to pure array access since I had to deal with handles (a pointer to a pointer, floating memory access mechanism used by the classic MacOS) anyway. Preallocation and direct data access helped improve performance even more. The application was IO bound by the floppy speed so LZSS decoding was almost unnoticeable there and only added about 15% overhead when running off a hard disk on an '030 class system.
Despite a bit of poking around, I can't find a reason for Apple (or anyone else) to pick LZSS as the mechanism for compressing the kernel. Perhaps it was just a timing thing (NeXT was doing a lot of kernel work around the time when people were tinkering with LZSS)?
From Squidly we get some insight into New Mexico's bear flight training program.
I know Gregg Easterbrook is going to be all over this come Tuesday with yet another kick early, go for it late rant.
I finally bought a new bowling ball a few weeks back and started using it a couple weeks before Christmas. First impressions on new balls are just flighty enough that I wanted to give it a little time. I think it's been long enough now and I really like the Columbia Wow!
When Brunswick phased out the original Red Alert, I bought one more, had it drilled about eight months later and that's been the ball I've used ever since. I'd never really made the jump from reactive to particle balls having tried twice; both were actually worse for me (we re-drilled the HPH for Adam and he likes it a lot, the other is useful on drier lanes but not on the oil we normally play on). As the amount of oil has increased (to deal with all the particle balls out there) and my ball has died, it's become progressively more exasperating. I've worked on reducing speed (which has always been an issue) but at some point a change was needed. With focus, I can get down to 16-17 mph and it can get up to nearly 20 if I'm wound up or paying too little attention. Slowing down also throws my timing right out the window at times. I was working too hard on something which is supposed to be fun and getting close to wanting to quit altogether.
As a 'tweener (neither a full roller nor a rev' master) with more speed than makes sense, I find this ball to be quite forgiving. It's a lot like discovering the Danger Zone all over again (the Red Alert was the last in the series of Danger Zone balls, all of them had a light bulb shaped core and for the time, a very aggressive cover stock — outer shell). Maybe it's new ball syndrome, but it's been a good three years since I felt like I could trust a ball to rollover on the backend on the swamp conditions we normally bowl on. The results have been better than expected and I'm actually looking forward to bowling on heavy oil again.
Paul Zimmerman (Dr. Z) answered a Packer fans complaint about his pick of Philly over Green Bay (24-20) thusly:
Steve of Appleton, Wis., offers many reasons why my pick of the Eagles over his beloved Packers is all wet, which is what a loyal Apple from Appleton should be doing. He wants my thoughts. Here they are. We're gonna get prohibition back, only this time it'll be different. The mobs'll be organized. From what movie? Key Largo. Thomas Gomez responding to Edward G. Robinson's, "Say something, say something." Wait a minute. You mean you want my SANE thoughts. You didn't say that. Packers' defense will give up too many yards and scores. Favre will throw two picks. Pack will get big rushing yardage, for a while, then will cut back on the running and play catch-up, which will be tough against that secondary, if everyone's healthy.
I enjoy Z's smart assed responses most of the time, but that one was pretty weak. While that's a reasonable opinion, a little backing data would have been nice. The other thing that bugged me was the prediction of two picks. Brett doesn't often throw two picks; it's more common to have a small number or bushels of them. Ya gotta pick one. 'Nough said.
On the other hand, Zimm does recount some of his all time favorite QB performances. Fran Tarkenton owns the number one position, which is interesting if you're a long time Packer fan. Tarkenton played for the Vikings for long enough to become memorable to me, not necessarily in a good way ('m not old enough to remember the first go around, but I was certainly paying attention during his second stay in Minnesota).
In between a blizzard of other things that are happening (life as it were, going back to work after several weeks and so on), I've been thinking about the game last week against Seattle and this weeks showdown in Philly.
For Green Bay it's going to come down to their first half offense and second half defense. Good weather would also help (freezing rain or snow will change the tone of the game in ways no one can predict).
If they can get out of the gate early and establish their offensive balance, they have a good chance. Despite what's been said and written this year, the Packers still run a balanced offense. The passing game was more active in recent years but it's still a big part of the plan (we're not talking about Carolina here). When the running games gets bogged down, it's up to the passing game to get things moving (and vice versa). The extended passing game has also contributed to defenses playing more honestly. When you're burned for a long pass play, you tend think twice before trying to cheat again. The steady, consistent play of the offensive line has made it easier for everyone.
It'll be interesting to see how this plays out against the blitzing attack of the Iggles.
If the defense can keep it together in the second half, which has been a problem all season long, they'll have a much better chance. Going up by two or three touchdowns isn't worth much if you're going to give it all back in the last thirty minutes. Once the Pack has a lead, they've got to maintain it. Several times this season, including last week against Seattle, things start going wrong late in the game. I don't have an answer for it (and thus far, neither does anyone on the Green Bay staff), but it certainly make for some nervous times near the end of games.
After watching the second half of the Philly game again, I believe that Green Bay can win. When the defense steps up and does their part late in the game, as they did in Philly, the offense has to step up and do their part by playing keep away for long enough to make sure there are no last chances. I also expect that this will be a hard fought game, in which case we'll be holding our collective breath as things come down to the wire.
Elgato Systems was the one place I was checking out as MacWorld approached. I'd seen the EyeTV 300 and 400 products, noticed that they were FireWire based and sniffed about it a bit. I held my breath, bit my tongue and didn't say anything.
Yesterday they announced the new stuff.
I honestly don't know what to make of the EyeHome product. If they had .avi support it might be more interesting. There are a lot of homebrew solutions in this area that just seem more useful without all the added expense (no cables locally). At this point, the product is tuned to what they generate rather than what a typical user might accumulate. For example, the movies coming out of my camera are .avi files. I want to play them on the TV right after getting them out of the memory card.
EyeTV 200 is what I'd been hoping for. We'll have to see how the retailers end up pricing it but it sounds like a terrific device.
I feel so stupid. As Kurt put it:
Never saw this before. I'd have to say operator error.
I've been trying to create some movies with iMovie that I can share with the family. Every single one wound up having two extra white frames at the end. Arggh! This is supposed to be easy.
Kurt came by to see.
I showed him my movie(s). Yep, white frames. He was sticking with his operator error theory. I open iMovie, do an export for web streaming. The bad frames are there. I explain all the expert settings contortions I went through. Kurt tells me that he exports to QuickTime for CD-ROM. We try it, it works. Arggh! It's a little bit smaller than the size of the movies I was trying to make originally.
Lessons learned? I haven't the foggiest idea.
What is this?
m002-01.bess.net - - [04/Jan/2004:20:01:16 -0800] "HEAD /blog/ HTTP/1.0" 200 0 "http://www.dickgephardt2004.com/" "StarProse Referrer Advertising System 2004"
Well... if you can sell it and someone will buy it, what the heck. Maybe Gephardt was asked to be the party sacrificial lamb to test out new ways of annoying people (I've heard from others that have received spam from the Gephardt campaign). The value here is hitting sites which make their raw referrer logs available to the public somehow (a lot of weblogs do this). With one quick hit you get mindshare. Since I don't, the usefulness is at least somewhat reduced.
Comment spam is all about doing the same sort of thing on a more permanent basis.
I wrote that last night and today I was sent some more information...
Doc wrote Say it ain't Joe on Saturday and then it hit MeFi. Shawn Allison has seen the StarProse journal nonsense too.
Susan Kitchens has been doing a bang up job covering last evenings successful Mars Spirit landing. I was up pretty late and she was still doing updates when I went to bed. She's also had all kinds of updates today. My thanks to Susan and countless others who are doing a great job of keeping us on top of the Mars news.
Has it really been six and a half years since we last did this? It feels longer somehow. Perhaps the failures color my perceptions. Rather than a gradual ramp up of Mars based research data, we've been waiting expectantly ever since Pathfinder went silent. I know I became spoiled by the string of successes, perhaps everyone did. Every failure was a dashing of hopes and the silly reasons for failure didn't help matters. Some of us wondered, is this ever going to work again?
Obviously, it did, and once again we're gathering data from our nearest neighbor.
My congratulations to everyone who made this possible.
One of the comment threads on a VoIP prediction column at Voxilla.com added a new item to the prediction list:
11. VoIP Spam As Asterisk hits it big and other systems start to accept VoIP calls from anybody (possibly because of ENUM-routed calls), the #$%^&*& spammers and telemarketers will start to use VoIP for their messages.
Ah yes, our friends the telemarketers.
Asterisk is an OSS PBX solution that runs on a number of platforms including MacOS X. Given call queuing and adoption of ENUM, it won't be long before the telemarketing community starts buying and selling IP phone lists and then things will get a bit crazy.
This ought to stand the whole Do Not Call controversy on its head.
Boing Boing mentioned From Vinyl to Digital, Hold the Crackle a couple days ago and I finally got around to reading it.
The NY Times article mentions the iMic from Griffin Technology but gets so many details wrong it's almost unbelievable (the company name, company URL, and so on). They certainly didn't discuss actually using the iMic. Yes, you can use the iMic alone, but it doesn't have enough amplification to deal with a phono connection nor does it have RIAA equalization. As I mentioned last April, you need a preamp, and I recommended the DeeJayPRE Phono preamp which still seems like a good choice.
Enjoy the article, but take it with a grain of salt.
My education in VoIP continues.
While trying to get some idea of what ENUM is all about from a back end perspective, I've been poking around on CircleID quite a bit. Parsing Hype From Hope: Will ENUM Spark Changes In Telecom? is a pretty darned good description of how ENUM works from the client side perspective. Convergence: ENUM is a Big Deal by John Patrick is the one you find most people linking to (and is also worth reading). Bret Fausett recently pointed out Letting DNS Loose by Paul Mockapetris (he played a large part in developing DNS). The article argues that DNS is available, scalable, extensible (via NAPTR records) and can be used now, so let's take advantage of it. I've been surprised that so little of the web services infrastructure has gone this route. Perhaps we'll see more of that sort of thing happening this year.
Jeff Pulver's blog (rss) is a very good place to check for VoIP related happenings. I should also point out the SIP Wiki since David Beckemeyer brought it up recently and the Wikipedia SIP page.
On the Vonage front, things are going well. I asked Sarah to kill off our second line yesterday, which says something. A Voice in the Calling Wilderness from the NY Times is a good introduction to Vonage from a couple weeks ago. The author had many of the same questions and concerns I had going in.
I've had a few interesting configuration issues (I somehow got calls to forward immediately by playing with the call forwarding and voice mail delays... a bit of fiddling restored the behavior I wanted). It does not seem possible right now to easily flip call forwarding on as I walk out the door, but I expect that'll improve. Our VTech dual line phone (20-2431) doesn't seem to pay much attention to Caller ID information from Vonage, but I'm not too concerned about that. When I asked a support rep about it he said that he'd heard of a problem. It's not that big a deal to me, YMMV.
I did encounter something odd. I was trying to call someone who had tried out Vonage and then decided to leave their service. From a local line or a cell phone, the call worked. Calling via the Vonage service resulted in a fast busy tone. A query to support straightened things out quickly, but I do have some concerns about their deprovisioning process.
Need a tutorial on using wget and/or curl?
SuperDeluxo4.2 wgets and curls is an interesting place to get some background information along with a bunch of working examples. The Elegant Universe is especially cool (I wasn't aware that PBS did a special on it). [via random($foo)]
I'd never used iMovie before last week, so please forgive me for being out of the loop.
Is it normal for iMovie to add a couple of completely white frames at the end of every exported movie? If so, what's the recommended way to remove them? I tried everything I could think of yesterday with no success.
I guess if all else fails I can purchase a QuickTime license and just snip them out but that seems to sort of defeat the purpose of iMovie.