Rss

Archives for : httplivestreaming

Streaming the Streaming Talk (CocoaConf Raleigh, Dec. 2012)

Since Jeff Kelley demanded it, here’s a dump of the setup I used to livestream my CocoaConf Raleigh session about HTTP Live Streaming (and the Core Audio session before it). Attendees of these talks are welcome to skip to the bottom of this blog for promised links to code and slides.

First off, links to the recorded livestreams on UStream:

  • Core Audio in iOS 6 – stream
  • Mobile Movies with HTTP Live Streaming – stream pt. 1, pt. 2

The second talk is split in two parts because Wirecast crashed partway through the session, breaking the stream and forcing me to start the broadcast anew. And that speaks volumes about what a seat-of-the-pants operation this was.

Continue Reading >>

The Sound of Plaid Friday

CocoaConf Plaid logoOh, this is nice. In the spirit of all these aggressive Black Friday offers, CocoaConf has announced a substantial deal for what they’re calling Plaid Friday. On November 23 only, you can score a registration to next week’s CocoaConf Raleigh for $50 off and get free admission to my Thursday All-Day Core Audio Tutorial (or the introductory iOS Tutorial taught by Walter Tyree).

In another fit of geeky Black Friday goodness, Telestream is offering 30% off all their products, including the screencast recorder ScreenFlow (which I use), and the livestream production suite Wirecast, which I’ll be getting for $!50 off, thankyouverymuch.

And that takes us back to CocoaConf, actually, because I’m going to be livestreaming my two Saturday morning talks from there on UStream. The events are now scheduled on the invalidstream: Core Audio in iOS 6 at 9:00AM EST (14:00 UTC), and Mobile Movies with HTTP Live Streaming at 10:45 (15:45 UTC). Apparently, I can’t deep link into the scheduled events on Ustream, so please visit the invalidstream channel and sign up for a reminder if you like. And yes, the live streaming talk will itself be streamed, and that stream will discuss how the stream is created and played (on OS X and iOS devices). So that’s ridiculously meta, at a minimum. Also, I’m bringing a camcorder, tripod, and 15′ FireWire cable, so I should be able to get a halfway decent longshot for the talks that can capture both me and the screen (and the audio we’ll create in the Core Audio talk).

CocoaConf Portland ’12 and the AudioQueueProcessingTap

CocoaConf Portland was this last weekend, and the conference continues to grow in scope, prominence, and depth with each installment. Visiting the US west coast for the first time, it picked up Brent Simmons (famous for NetNewsWire, Glassboard, and MarsEdit) and Daniel Pasco of Black Pixel as keynoters, plus James Dempsey playing some of his famous WWDC developer-oriented songs, such as the timeless luau of Cocoa memory management, “The Liki Song”.

For my stuff, I kicked off Thursday with a second run of the all-day Core Audio tutorial, which I’ll be doing again at CocoaConf Raleigh. It’s nice to teach these advanced classes, because I keep learning things about Xcode and Obj-C from the attendees as we work through the projects together.

Continue Reading >>

My CocoaConf Columbus file-dump

I spent too much time away from sessions (readying my sessions, revising the book for iOS 6) at CocoaConf Columbus to do my usual “What You Missed…” style blog entry. I guess you could say I missed it too. But at any rate, I still owe the attendees some files.

Everyone in the all-day Core Audio tutorial should already have the five sample apps that we built in class, and a fallback zip was given to the group first thing in the morning, but if you some how missed it, here you go:

The HTTP Live Streaming talk is little changed from its CocoaConf DC version, but for the addition of a few new slides that show off more HLS apps (like NBC’s Olympics apps) and some off-the-shelf encoding boxes that’ll set you back $25,000. Here are the updated slides; sample iPad code is the same as DC’s:

Twice now, I’ve tried to show actual live broadcasting, and been thwarted both times. There’s actually nothing in Apple’s toolchain that provides the MPEG-2 transport stream needed by the mediastreamsegmenter tool. There is a trick on Stack Overflow that gets a seemingly-suitable stream out of VLC, but while the .ts segment files and the .m3u8 files are created and look correct, QuickTime X is unable to open the resulting stream. It doesn’t help matters that VLC can only capture from the webcam and can’t do audio capture, nor can it use screen:// as an input in Lion or later, which breaks Erica Sadun’s HLS screencast trick of a few years ago.

I’m going to do the all-day Core Audio tutorial and the HLS talk again at CocoaConf Portland in October, possibly something new too. Hope to see some west coast media developers out there.

Midsummer streaming

So for some reason, I’ve got an itch to experiment a little more with streaming. My talk at CocoaConf DC went well enough, but the sight of seeing mediastreamsegmenter dutifully dropping off webcam captures as 10-second HTTP Live Streaming segment files into my web server Documents folder, or the fact that I could turn my Dropbox public folder into a streaming video server, has me thinking about getting deeper into this stuff.

That, or the fact that I’m looking for an excuse not to get depeer into AV Foundation.

Personal broadcasting seems a bigger deal with the younger people in my circles; my college peers have largely never heard of it, while my anime tweeps are all happily streaming away with their video reviews and retro-gaming sessions on UStream, Justin.tv, Twitch.tv, etc. I’ve dropped by JesuOtaku’s show a few times and she’s got over a hundred anime fans in her chat room (it helps that she’s well known and quite entertaining). In my HLS chat, I showed a screenshot of some dude playing Diablo III to a stream with over 1,000 viewers. Watching someone else playing games! And in the screenshot from the Twitch.TV app below, nearly 3,000 people are watching a speed run through The Legend of Zelda: The Wind Waker.

20120712-210337.jpg

So anyways, research project for the next month or so is going to be evaluating the various services in terms of Mac and iOS compatibility, technical feasibility of producing something more than just a headshot, how good a stream I can get out with 1.5 Mbps of upstream bandwidth, etc. Streaming Media’s Jan Ozer published a nice comparison of UStream, Justin, LiveStream, and Bambuser in the latest issue, so that’s one place to start. I think I also need to figure out some of the Mac software packages (CamTwist and EvoCam, I think) that convert a video input (perhaps mixed from screen cap and webcam) into the Flash stream needed by these services.

Another option would be to host my own HLS-only stream, which is highly appealing (no ads!), but there seem to be few free or cheap options for generating the MPEG-2 Transport Stream over UDP (or a pipe) that mediastreamsegmenter wants as input. I asked about this on the devforums and word from Apple is that none of Apple’s tools provide this, although there are nice third-party products.

Surprisingly, VLC is capable of producing the input transport stream, although the resulting HLS stream wasn’t playable when I demo’ed it in DC. Maybe I’ll get it working for Columbus.

Speaking of CocoaConf Columbus, Friday is the deadline for early bird registration. I’ll be doing HLS and an all-day intro to Core Audio (which is an add-on to the main conference), which is my chance to do a really substantial run at Core Audio teaching, even covering a few things that aren’t in the book, like network streaming.

What You Missed at CocoaConf DC 2012

…a big freakin storm, strong enough to spin the A/C fan backwards and blow wind and rain back into the room, but that’s neither here nor there.

So, I spoke at CocoaConf DC 2012, presenting a new talk, “Mobile Movies with HTTP Live Streaming”. It was a good excuse to get me off my ass, download the tools, and build a few of my own streams. They’re in the demo, so here ya go with some links:

HTTP Live Streaming demo app playing encrypted stream
I also did my iOS-5-oriented “Core Audio Cranks It Up” talk again, which focuses on the effect audio units and the AUSampler that we got in iOS 5. It ‘s still good and still gets high marks, but with iOS 6 on the way, it can’t help but seem like old news before long, so I’m going to retire it at this point. If you want to see it, I think Safari has the original version from Voices That Matter Boston, and InformIT will sell you the 90-minute video, for $15 or so.

Not getting high marks, surprisingly, was the Reverse Q&A session. Some of the feedback rightly dinged me for a somewhat unfocused panel — in Chicago, we had three panelists, but in DC, I did a scattershot invitation which everyone accepted, leaving me with a too-large panel of five. I think it works better with a small number of panelists who are willing to interrupt and steer the conversation. As it was, I let it linger too long on App Store grievances (even though that was where attendees wanted to go, rebuffing my first attempt to change topics). I still think the format is superior to the regular panel, at least with a manageable number of attendees (doubt I’d try it with more than 150 or so), but this experience teaches me that the format won’t take care of itself, and that I and the panelists have to own the conversation to a greater degree than might be readily apparent.

Anyways, next up is CocoaConf Columbus on Aug 9-11. This is where I’ll be doing an all-day intro to Core Audio, as part of my de facto 2012 Core Audio tour, and an experiment in doing an opening-day tutorial for advanced attendees. Although I should probably change that title — the tutorial is for advanced developers, but it will presumably be their first time getting really deeply into Core Audio.

Also a little tempted to do a new eyes-forward session if I’m retiring the Core Audio talk, and there are some very advanced video compression bits in [MOUNTAINDACTED] that might be worth shining a light on. Question is whether I’m fool enough to try to prepare the tutorial and a whole new talk over the course of the next few weeks. Watch the schedule, folks…

What Anime Central Taught Me About Mac/iOS Media

I spent the weekend in Chicago at Anime Central 2012, the first anime convention I’ve been to since Anime Weekend Atlanta 2007. Many of us geeks have our escapes: mine is the mythical realm of “Japan”, where children are taught the value of empathy, and people default to a position of kindness and respect in their dealings with others. ぼく の ゆめ きれい です。

invalidname's ACen Badge

Along with loading up on collectibles (Angel Beats! Anohana!) and finding new shows to watch (Toradora!), I had a few encounters that, surprisingly, led to some insights about where digital media is and is going on the platforms I work with.


That iPad UStream Guy

In the Sentai Filmworks panel, I noticed a guy in the front row shooting video with his iPad. Not a bad idea — the battery life is enormous, so you might only be limited by using up the flash storage. But what I discovered later was that he was the animecon-industry channel on UStream, and was streaming all these panels live on the internet. From the iPad.

Multiply this by a bunch more iPads and a bunch more interesting pursuits and I start to wonder why I’m not already combing UStream for cool stuff streaming now.

BTW, I asked three questions in the Sentai Panel. Watch the recorded version and you can hear me asking about whether iTunes download-to-own makes still sense for them, why they went back and licensed the older ef: A Tale of _____ shows, and what led to the reissue of Clannad After Story with an English dub.


Making Music with Miku

Hatsune Miku Vocaloid 2 softwareAnother panel I went to was a how-to showing how Yamaha’s “Vocaloid” software can be used to create synthetic singers, such as the very popular (and much-cosplayed) Hatsune Miku. What I didn’t expect to find out is that the Yamaha guy who created the Vocaloid software is a big Mac fan, and despite the fact that the Vocaloid products are Windows-only, was shown in a video from last year’s Anime Expo toting a MacBook Air to a panel (it may be this one, but if not, it still shows Ito-san clearly using an Air).

Duly inspired by this, the panelists at ACen showed a MacBook-based workflow that used Vocaloid 3, despite the fact the program is Windows-only, and was only partially localized for English. They played a synth into Logic Pro to lay down a base music track, then played a second track as the vocal line. They exported that track as a .mid (MIDI sequence) in a directory shared with VMWare, where it could then be imported into Vocaloid and sung by Miku. After tweaking the Japanese phonetic lyrics, they exported a .wav back to Logic to complete the song.

Of course, wouldn’t it be nicer to cut Windows out entirely? Knowing that there are Mac fans at Yamaha and Crypton, maybe Apple should make some calls. Having Hatsune Miku as an instrument in GarageBand or Logic would be a hell of a lot of fun, and would surely lead to getting her as the vocalist of even more songs on iTunes (she already had more than 1,500 last time I checked).

I’ve also mentioned on several occasions that the AUSampler audio unit in Lion and iOS 5 is doing a pitch shift more or less equivalent to what Vocaloid does, with the key difference that AUSampler can be played live, while Vocaloid uses a render step and could look ahead to upcoming notes to produce more realistic output. I’ve meant to try hacking up a “Hatsune Mac-ku” with AUSampler one of these days… it’s on the list of experimental projects that could turn into an article/blog/session if I find the time to get it working.


There’s An App For That Anime

Kids on the Slope page on Crunchyroll iPad appOne of the first things I did on the show floor was to finally sign up as a paying subscriber for Crunchyroll, the streaming service that offers many anime shows within hours of their Japanese airdate. Considering I’m currently using it to keep up with Kids on the Slope, Bodacious Space Pirates, and Puella Magi Madoka Magica, I’ve come to feel like quite the freeloader.

Part of the reason I’m watching so much Crunchyroll is that it’s easy to do so on the iPad while waiting in the hallway for my kids to fall asleep. Like a lot of Flash-based websites, Crunchyroll has had the sense to build an iPad app. And one of the upsides of having a real app is that the video can be sent over to an Apple TV for proper wide-screen viewing.

Watching Bodacious Space Pirates on Apple TV with AirPlay
When I look at the “Anime” folder on my iPad — consisting of Crunchyroll, Anime Network, Funimation Free, and Adult Swim — and then look at the icons on the Apple TV screen, I can’t help but wish that these streaming apps could be on the Apple TV itself. With the latest Apple TV update presenting us with a grid of app-like icons for the various services and providers, it sure feels like this is what we’re heading towards. And when you can just subscribe to your favorite sports league as an app and not deal with cable/satellite/local-broadcast hassles and blackouts, it starts to show the promise the cord-cutters have been talking about all this time.

The New York Times had an interesting article on this the other day. After presenting the channel-as-app metaphor, they point out that this could bring about the “a la carte” model that so many viewers have wanted for so long: the ability to just buy the content that you want, and not have to buy a bunch of programming you don’t want. But they also identify the catch that I’ve worried about for years: many of the content providers are vertically integrated with distributors. All the NBC/Universal networks are owned by Kabletown Comcast, which means that they may not want to sell content directly to end-users when that cuts into the parent’s core business of selling cable subscriptions.

And would a la carte make sense for consumers? We might get sticker shock. Think about my anime apps: I’m in for $7/month with Crunchyroll. But they don’t have everything. If I want Lupin the 3rd commercial-free on Funimation streaming, that’s $8/mo, and ef on Anime Network is going to set me back another $7/mo. So just for anime, I’m in for 22 bucks a month! Do I have anything left for sports or news? Suddenly, the DirecTV bundle isn’t looking so bad anymore. [Indeed, Anime Network is also on DirecTV VOD, which is why I’d be highly unlikely to pay for it again as a streaming subscription. Last year, instead of cord-cutting, we doubled-down on DirecTV and upgraded to their “whole home” service. We’ll probably eventually want to get the nomad too. YMMV.]

Anime folder on my iPad
In fact, when I saw the much linked Oatmeal cartoon about I Tried to Watch “Game of Thrones” and This Is What Happened, I had a specific thought when I got to the frame where the protagonist is flashing his credit card in front of the computer, saying he was ready to buy if only they would sell it to him. Here’s my thinking: HBO knows full well that there are people who subscribe to the channel entirely for the sake of one show. That is their business model: you buy it for Game of Thrones or The Sopranos, and as a bonus, you get to see Splash and Independence Day 400 times a month. Which is total crap of course, since we only care about Game of Thrones. So if you’re paying just for that show, what’s the cost? Say it’s $30/month times however long a season runs, plus time to unsubscribe, leaving a little room for customers who don’t bother unsubscribing religiously. Shall we say four months? Then that means a season of Game of Thrones is arguably worth $120 per subscriber.

Now imagine if HBO put out a Blu-Ray set, day and date with the series, at that price point. Everyone would scream bloody murder. But, Mr. Oatmeal, you were flashing your credit card! Did you think that a new production should cost the same as a back-catalog show from 20 years ago that has already paid its bills several times over?

But I’m kind of digressing into old arguments. The point to make about networks-as-apps is that Apple’s treating Apple TV as a “hobby”, the lack of an SDK for Apple TV, the company’s slow movement towards backing it up with content other than the not-terribly-popular iTunes download-to-own and partners like Netflix… it could be that they’re not going to take on the entire cable/satellite industry until they’re confident there’s a real opportunity, if not an absolute certainty, that they’ll win. What I see is a long game, where they roll out technologies like HTTP Live Streaming and AirPlay, see if they take, and let the pieces quietly get into place. Not a fiendish master plan… just preparing relevant technologies and partnerships so they can enter this war at a time and place of their choosing.


Dreaming of Streaming

Speaking of HTTP Live Streaming, one last point is to acknowledge what a tremendous success story this has been? Non-existent 5 years ago, it is now the technology that delivers all streaming video on the hundreds of millions of iOS devices, in a way that satisfies the security demands of the major media companies, sports leagues, etc. As I mentioned before, my iPad is full of video streaming apps, not just the various networks and content providers, but stuff like UStream that captures and streams live video.

The thing to start watching now is what happens with MPEG DASH, which resembles HTTP Live Streaming and similar technologies from Adobe (Adaptive Streaming) and Microsoft (Smooth Streaming), all of which use HTTP (rather than custom socket connections on frequently-blocked ports) to deliver small segments of video and adjust to changing network conditions. Streaming Media reports increasing support for the proposed DASH standard from many companies, but notably not Apple. Which makes sense in a cynical view — why let the competition catch up when content providers already have to go HLS to reach the massive iOS user base? A more practical concern is that DASH seems to want to make everybody happy by just wrapping some existing standards for codecs and manifest delivery, which may end up meaning that it just becomes the “15th standard”:

And speaking of HTTP Live Streaming, I’m preparing an all-new session on HLS for CocoaConf outside DC in late June. The early-bird deadline has been extended until this Friday (May 4), so if you want to see how cool this stuff is, you’ve still got time to save a few bucks.

Now if you’ll excuse me, I need to put this Crunchyroll subscription to work and catch up on the subscriber-only new episodes, now that I just finished Angel Beats! from iTunes last night.

Mid-April progress update

OK, grab bag of stuff:

Author copies of Learning Core Audio

Amazon freaked out people who’d pre-ordered Learning Core Audio by sending, on the book’s release April 9 release date, a “yeah, it’s gonna be another three months” e-mail. Gee, thanks guys. Publisher assures me that thousands of copies have been printed and many of them are in transit to Amazon, or may be there by the time you read this. Amazon’s listing now says it ships on Saturday.

Collecting my kickback-powered affil links from previous posts:

Availability of eBooks outside the US is spotty: iBooks is US and Canada, Kindle is US and UK, etc… Apparently, we’re waiting on contracts with various vendors in various countries. Elsewhere, I think you can just get the ePub+PDF bundle from the publisher, and one or both of those formats should work on the reader of your choice (plus, InformIT’s eBooks have no DRM, just watermarking, so you can be publicly shamed if you seed a torrent of it. And don’t think I won’t.)

Also, if you’ve read the eBook via Kindle, iBooks, or via the Rough Cut on Safari Books Online, Kevin and I would really appreciate it if you could assign some stars and/or write a review on Amazon or iBooks. Empty review sections look lonely. Honestly, I’d rather have a bad review than none at all; indifference is a killer.

On my other book, we’ve just pushed beta 4 of iOS SDK Development, which includes my final chapter, covering “grown up” techniques like source control, submitting to the app store, and working with .crash reports from iTunes Connect. Now we’re off to tech review, print to follow.

I feel like Bill and I have accomplished what we wanted to with this book — instead of a “grand tour” of every API that tickled our fancy, we focused on a solid grounding in the new tools and techniques (Xcode 4, blocks), and best practices (unit testing, multi-core awareness). I think that’s also going to be a little more future-proof than the first edition, which got increasingly out of date as Apple bulked up the iOS SDK.

I do hope that Apple gets /usr/bin/xcodebuild test target-name working before we ship the paper book. I’d like to send this off with a solid automated testing story.

Finally, I’m going to be speaking again, at CocoaConf DC, in Herndon, VA, on June 28-30. This time around, my talks include the iOS 5 Core Audio talk I’ve done a few times now, another round of the Reverse Q&A that went over so well in Chicago, and a new talk on HTTP Live Streaming. HLS is something I’ve shoehorned into many of my AV Foundation talks, and it’s time I just build a whole one-hour talk around it with demos and code.

Early Bird registration for CocoaConf is up now, ends April 27. Hope to see you there.

And if not there, well, I’m also talking with them about possibly doing an all-day Core Audio tutorial at a future CocoaConf. This comes from a couple tweets I sent out mentioning the habit of these smaller iOS/Mac conferences to start out with an all-day tutorial for beginners, and wondering whether there’d be an audience for a full day of advanced material. Given the interest in the Learning Core Audio book and the fact that most advanced developers haven’t touched Core Audio — legendary for being crazy hard and all — it struck me as an interesting option for an experiment. Is there potentially a room-ful of people that would travel to a conference to get a one-day deep-dive introduction to a hard topic like Core Audio? Seems worth trying and finding out. Actually, if you know you’d be interested in this, please post a comment. Thanks.

Presumed Valid

A day after Apple’s third quarter earnings announcements, it’s time for the reading between the lines and amateur Kremlinology.

Here’s one thing that doesn’t make me very happy. Grab the earnings call audio (and notice from the .m3u8 extension that it’s in HTTP Live Streaming!), and scroll to around 24:50 in. The question is from Bill Schultz of Goldman Sachs:

OK, and a quick followup. There’s been a lot of news about, you know, patent disputes of late. Some seem to have gone in Apple’s favor while others haven’t. Can you help us understand, Tim, how to put these events in context… how we should think about your IP strategy overall and perhaps, you know, how all this is potentially impacting the competitive landscape in smartphones and tablets?

The much quoted response from Apple COO Tim Cook:

We have a very simple view, here. And that view is we love competition, we think it’s great for us and for everyone… but we want people to invent their own stuff. And we’re going to make sure that we defend our portfolio fervently.

That seems to speak obviously to Apple’s recent ITC win over HTC. And it’s probably meant as propaganda and spin as much as anything else.

But notice that it ignores the patent cases that have gone against Apple, such as Kodak and Personal Audio. Does Cook’s absolutist stance on patent mean that maybe Apple should have “invent[ed] their own stuff” in these cases? Maybe it’s a more relative morality, like every tabloidy legal case where whichever side wins declares to the TV cameras that “the system works.”

The elephant in the room is the patent trolls’ assault on indie developers, the consequences of which are spelled out by The Iconfactory’s Craig Hockenberry in The Rise and Fall of the Independent Developer. Cook’s response didn’t address Apple’s attempt to intervene in the Lodsys cases, or its current absence from even more galling patent claims against indie iOS developers by MacroSolve and others.

Is Apple going to step into every case that claims third parties violate patents by using Apple’s APIs? Well surely, they’re not going to commit to that publicly — barring a surprise “Thoughts On…” letter from Steve — and with the rate at which this nonsense is picking up, we should watch to see if and when Apple backs off and leaves indie developers to their fate. Apple could probably survive the worst-case scenario, the total collapse of third-party iOS development, considering that’s how they launched the iPhone in the first place (and considering how well Android phones are selling, despite a far weaker app ecosystem).

What’s discouraging is that it’s clear, despite hopes to the contrary, that Apple has no philosophical problem with software patents whatsoever. Indeed, in the Lodsys case, it’s not trying to invalidate the junk patent; it’s asserting that Apple’s license of the patents covers its use by third parties. The instinct may well be in their corporate DNA (and that of other large tech companies) that patents are a critical strategic asset, something to be amassed and protected whenever possible. That MacroSolve could patent electronic form submission, in 2003, seems not to bother them. Patents are, apparently, too important to be doubted.

AppleTV SDK? Please say yes.

I’m writing this in flight, Grand Rapids to Denver, connecting from there to SFO and WWDC 2011. Screen typing isn’t the most fun thing in the world, but the small size of the iPad, combined with its shape (no fold open screen like on a laptop, which would only collide with the reclined seat in front of me) makes it reasonably practical.

Lots of people precede WWDC with predictions and wish lists… I’m going to use this blog for a justification, in other words, why the thing I’m hoping for might make sense.

The surprise I’m hoping for is an Apple TV SDK. Not that I’m the biggest Apple TV fan… not only do I not have an Apple TV, I don’t even have a TV that will accept its HDMI connection.

But man, would I get on board quickly if there were an SDK.

And no, not to play Angry Birds on the TV. If anything, the potential for general-purpose apps on the TV is badly overstated, and has been for years, if not decades. QUBE, WebTV, various cable box standards (what was that Java one that I used to write about at java.net and was roundly ignored?)… the space has been an unbroken string of failures.

But let’s narrow our foucus. What if the SDK was optimized for streaming AV content, and providing just enough UI for browsing and choosing content (i.e., nothing more than Electronic Programming Guides [EPGs]), ignoring most of the power and elegance of the underlying iOS frameworks? What if Apple only allowed apps in this genre, rejecting anything that wasn’t an AV streaming client?

It would be awesome.

Here’s the thing: the iPhone and iPad gold rushes have motivated a lot of content providers to make their stuff available in iOS formats. Movies, TV shows, live sports, niche programming, tons of stuff. More variety and depth than you can get from a cable TV subscription, without the tyranny of the cable company taking away your favorite channel in a carriage dispute, with no recourse for viewers. And it’s a la carte: buying the Major League Baseball app doesn’t require you pay for a CourtTV or Lifetime app.

Yeah, you see where I’m going with this. Using apps as a content delivery platform, Apple TV could grab a lot of those monthly subscriptions that are currently going to cable and satellite television. And with providers taking 70% of content subscriptions sold through in-app purchase (or 100%, if users can be convinced to purchase via the provider’s website instead), there’s huge money to be made in cutting out the middle man.

More content for viewers, just the stuff they actually want, and more money for the people who actually make the stuff. Win-win. The only loser is the middle man – the hated cable companies – who get cut out of the equation because they’re no longer necessary.

So what’s the case against? Apple is careful to pick its battles, and this may not be a hornet’s nest it wants to kick up this year. Let’s war-game it: what would cablers do in response? They’re dreadfully afraid of becoming “dumb pipes” (even though that is clearly their destiny), so they can be counted on to fight this as hard as possible. They might want to block the offending streams, but the nature of HTTP Live Streaming makes that difficult: you can’t just switch off port 80, nor can you block content delivery networks without collateral damage. So they could be counted on to revive the “content providers are getting a free ride off us” spin.

More importantly, cable companies by their nature are whores of local governments, and could get legislative and regulatory attention by pointing out how much cable fees go into local government coffers. An Internet TV world is one without public access channels, without the goddamned useless “city council meeting that lasts forever” channel, and who would want to live in a world like that? Well, pretty much everyone, actually, but good luck getting a bureaucrat to listen to that when their paycheck is partially funded by Charter.

Local cable franchises are a relic, but a lot of parties have a vested interest in maintaining the status quo. Apple would have to be eager for a thousand little battles to pull this off.

Provided, of course, that cable sees it coming. They might not.