Kickstarter for Henry Fool Part 3: Ned Rifle

Two years ago Hal Hartley successfully raised completion funds for his film Meanwhile through Kickstarter and now he’s trying to finance a whole film. It’s called Ned Rifle and it’s the third and final film in the Henry Fool trilogy. The script is awesome, and the idea of making three films over nearly 20 years with the same group of characters and actors is wonderful.

Part 2 of the Henry Fool seriesFay Grim, was the first feature film I worked on in the post production department. I was Hal’s assistant editor for Fay Grim and if we raise the necessary funds I will be the editor for Ned Rifle. Here’s a fun video in which I recount some details of the previous films’ plot along with the rest of the crew and the actors who will reprise their roles from the first two films.

Really Cool New Feature in My WordPress Video Plugin

Choose From Video

I’ve always been annoyed that my WordPress “Video Embed & Thumbnail Generator” plugin required FFMPEG to make thumbnails. Most people are on shared hosting and aren’t allowed to install software like that on their servers. And even if they are allowed, configuring and installing it is a pretty substantial hassle.

I started my most recent coding burst with the inspiration that I could show the video in a little player in the browser and use it to find the exact timecode a user wants to generate a thumbnail. I planned to send that number to FFMPEG in order to get the image, but when I saw the video in the browser it looked exactly like a thumbnail. I wondered why I couldn’t just grab the image that the browser had gone through all the trouble of decoding already. It turned out I could do exactly that, and it was surprisingly simple.

So as of version 4.2, you don’t need any special software on your server if you want to turn a frame of video into an image. There are some limitations though. Your server needs to have either ImageMagick or GD. Most servers have one of these enabled, so that shouldn’t be a problem. Your browser also has to provide native support for your video format. Plugins will not work. That means this doesn’t support FLVs, WMVs, AVIs, or MKVs. Browsers have built-in support for H.264 MP4s, WEBM, or OGV. There is a helpful chart on Wikipedia that details the array of browser support for these formats. The short version is if you use H.264 MP4 videos in Chrome then you should be fine.

I also moved my development onto Github, which I am loving. It’s much easier to keep track of everything, and it allows for savvy users to offer their own code to merge into the plugin. If you’re having any trouble with the new release, please post it in the issues section.

Where’s my $48 Kozmo.com?!

Kozmo.com is coming back and I couldn’t be happier because they’ve owed me $48 for 13 years.

In the summer of 2000 I lived in a $400/mo SRO in Chelsea. I assure you it was even worse than you’re imagining. I had an internship at CourtTV and was making minimum wage. I got my internet from NetZero, which at the time was one of those brilliant dotcom ideas where you got free dial-up internet in exchange for looking at ads. I had started using the Kozmo.com delivery service while at school that year and it was perfect for my solo lifestyle in New York. I spent a lot of my money renting DVDs online and getting them delivered to my door within the hour. Of course now I can hardly be bothered to go through the effort of putting a disc in my Blu-Ray player when there are so many instant streaming options available, but it was really cool at the time.

One night I rented Blue Velvet and watched it on my computer because nobody had stand-alone DVD players back then and I didn’t have a TV anyway. I enjoyed it very much. A day or two later (I don’t remember their rental policies, but it was within the acceptable time frame) I returned the DVD to the drop-off box in a donut shop on 23rd St, where I frequently returned DVDs from Kozmo.com. A few months later a $48 charge from Kozmo.com showed up on my debit card. I checked in with their customer service and they explained that I had never returned the Blue Velvet DVD and so was being charged $48 for the privilege of keeping it.

At the time they were having financial trouble, so I figured it was a scheme to stay solvent, but it was an unacceptably large percentage of my net worth so I was unenthusiastic about my part in this scheme. I explained that I did not keep the DVD and that $48 was a bit much for a DVD anyway. They apologized and promised to return the money. Months went by and I was frequently assured that I was going to get that money back. Then in April I learned that they were going bankrupt.

I found this in my archives:

From: Kyle Gilman <[email protected]>
To: [email protected]
Date: Tue, 17 Apr 2001 02:58:00 -0500
Subject: Still missing $48

Hi there Kozmo people. I guess you’re going out of business, but I thought that before you completely close up shop that maybe you could give me back the $48 you wrongfully charged me. I’ve been trying to get it back since January, and have been promised several times that it would be credited to my credit card and it’s never actually happened. I know you guys have money problems, but I never had anywhere close to $280 million to blow through.

Anyway, my username is ********, and I was incorrectly charged for “purchasing” an overdue DVD of Blue Velvet even though I returned it on time.

–Kyle

I never got a response and I never got the money, but I bet they have all kinds of cash on hand right now. How about a store credit or something?

Chromecast is Great But There’s a Big Feature Missing

I ordered a Chromecast as soon as I heard about it. It was $35 but came with 3 free months of Netflix streaming, which I would be paying for anyway, so it cost me about as much as lunch in Manhattan. It arrived on Friday and I’ve already given up streaming from my clunky Panasonic Blu-Ray player.

Streaming from apps that support Chromecast is fast, high-quality, and easy. Netflix, YouTube, and Google Play media show up on my TV with two taps. I ordered Europa Report from Google Play and had it on the TV in three taps on my Nexus 4. The movie was OK, but the purchasing and watching experience was as simple as I can currently imagine it. Casting a whole tab from Chrome is a different story. It’s really no good for video. It might be fine if you want something a little more static on screen but there was visible video stuttering when streaming from my MacBook Pro.

Here’s the big missing feature: on iOS if you have an Apple TV you can send any HTML5 video to it by pressing the AirPlay button in the video controls. Is there a good reason why every Chrome browser can’t send any HTML5 video to Chromecast today? I know that Chromecast works differently, and requires a receiver app, but it seems to me that Chrome could provide a generic HTML5 video receiver app without requiring every content provider to register as a developer and make their own, since every receiver app would be essentially the same. Maybe this is coming in Chrome 30 and I shouldn’t expect things to move so quickly, but I haven’t been able to find any discussion of it so far. I would love to incorporate it into my WordPress video player plugin as soon as possible.

Hey Look: I Have a New Website!

My website’s design was almost exactly the same for seven years. I built it out of the Blix WordPress theme back in 2006 and a few years ago in a fit of modernization I expanded the width of the main column. I always keep my resume updated, but mostly the site just sat there featuring months-old blog posts. But now I have a fancy, modern design thanks to the incredible WordPress theme designer Kriesi and his Enfold theme. Now everything works on the fancy smartphones nobody had in 2006. Rather than this musty old blog I’m now featuring my editor reel and many clips of projects I’ve worked on like this one from the Dead Possum episode of Maron featuring up-and-coming comedy star Josh Brener:

My favorite part is the tag system I added in. You can use it to look up all the projects I’ve done with a particular person. Want to see every Hal Hartley project I’ve been involved in? How about super-producer Jim Serpico of Denis Leary’s production company Apostle? Boom! Technology.

Check out all the cool new stuff. If you’re looking for an editor for your comedy series, might I suggest you peruse the Series section. Or if long-form is more your style, head over to the Features. I’ll do my best to keep the projects up to date, and maybe I’ll post in the blog a couple times a year.

At Least The New Mac Pro is New

After last year’s non update update, the Mac Pro is getting some real attention. This year Apple has announced a very powerful, very small, and very strange new computer.

2013_mac_pro

Sure, it’s an “innovative” design. I don’t really even care that it looks like a trash can or a coffee maker. My editing computer is hidden away in a closet anyway. What’s important about this new computer is that it redefines the idea of expandability in a powerhouse computer.

Remember this?

Remember this?

In 1999 the blue-and-white G3s introduced a tab that you could pull to swing the computer open and mess with its guts. Variations of this idea persisted in the G4, G5, and cheese grater Mac Pros. Just like on a generic PC, you could install hard drives, upgrade the RAM, and most importantly, install cards in its expansion slots. Since then there have been a number of other Mac models that offered limited expandability in exchange for “just working.” Those were and are some great computers. I’m writing this on a lovely 27″ iMac in an office full of iMacs and I like it very much. But sometimes you want your computer to do something that can’t be handled without direct access to the brain. Historically expansion cards have provided that power. The new Mac Pro has no slots for expansion cards and no space for SATA drives. What it does have is lots of Thunderbolt ports.

Thunderbolt devices can theoretically handle most of the things that expansion cards always did, but here’s the problem: they don’t do it yet. There are Thunderbolt to PCIe boxes that fill in the gaps in functionality, but they are very expensive. This new computer isn’t coming out for several months, so the world will be a different place by then, but Thunderbolt has been around for over two years and I rarely see Thunderbolt devices in the wild. The absence of Thunderbolt on the old Mac Pros has held the technology back from wide adoption in the professional realm, but the price premium is also a problem. G-RAIDs with Thunderbolt are almost twice as expensive as comparable Firewire/eSATA/USB drives. Glyph doesn’t seem to make any drives with Thunderbolt. The Thunderbolt version of the Blackmagic Intensity is $100 more than the PCIe card.

The only complaint I ever had about the Mac Pros when they were current was that their thick metal cases made them much too heavy. That was usually a problem once per computer: when I took it out of the box and installed it and never moved it again. I certainly opened them up to install cards and hard drives and RAM, but I didn’t have to move them to do it. I and most people I know who use Mac Pros would have been very happy with an upgraded processor, Thunderbolt, USB 3.0, and I guess no Firewire since that seems to be the way things are going. If we’re lucky, this will be like getting rid of floppy drives. It seemed crazy at the time, but then everybody had a CD burner in their Macs and CD-R prices plummeted. Same thing happened with DVDs when Macs all had Superdrives. Now we don’t have any optical storage, and good riddance. Thunderbolt is a very complicated technology and the high prices are not arbitrary. Will removing the option to use anything but Thunderbolt make Thunderbolt devices inexpensive enough to use for everything? I hope so.

I guess the big question is how much this machine will cost. An 8-core Xeon E5 is around $1500 depending on the speed. An AMD FirePro with 6GB of VRAM is $2400. There will be two FirePros in these things. Prices will go down by the time the computer is released, but his will not be a cheap computer.

Update to Video Embed & Thumbnail Generator Compatible With WordPress 3.5 and All Kinds of New Stuff

I’ve been working on this one for way too long. I was all set to release a big update to my Video Embed & Thumbnail Generator WordPress plugin when I found out in November that WordPress version 3.5 messed with the media windows enough that I’d have to do a lot of tweaking to even maintain the functionality I had before. I think I’ve finally sorted everything out, and I apologize to anyone who updated to 3.5 right away and hasn’t been able to make thumbnails for a month. I think the wait was worth it.

For me, the coolest new thing about this version of the plugin is the video gallery. It’s pretty and it was actually one of the easier things to do. I designed the basic functionality last summer in order to display rough cuts of the 50 short monologues I edited for CenterStage’s 50th anniversary. (For the finished product they designed their own site using Vimeo for playback.) Here’s a sample gallery of my short-lived animated series:

The most important new feature is probably the addition of Video.js as an alternative to the old Strobe Media Playback Flash player. Video.js solves the problem I used to have with native HTML5 players, which is that they were ugly and all look different. It’s also lighter and more flexible, and I highly recommend switching to it. I managed to add Google Analytics tracking too, if that sort of thing interests you.

I wasn’t keeping up with FFMPEG’s development and totally missed the fork to LIBAV. I know most of you haven’t updated FFMPEG in ages so it didn’t seem to cause much trouble, but I now support calling LIBAV directly.

MP4 files encoded by FFMPEG save the moov atom at the end of the file and that was causing problems with streaming playback. I knew the way to fix this was to implement a queue system for encoding videos and run qt-faststart or MP4Box once the file is encoded, but I also knew to do it right would stretch my meager programming skills to the limit. I spent a long time on it, but it seems pretty good now. One of the most difficult things was interpreting the output from FFMPEG because it’s not really designed to interface with other programs like I’m doing. And then I had to make everything look pretty and work with AJAX because what’s the point of making something ugly that forces you to refresh all the time?