[besson3c]

I've been doing a lot of playing around with MythTV and lirc (Linux software that allows you to associate remote control signals with keyboard shortcut commands, among other things), and it has had me thinking about the potential Apple has to develop something better...

Here is how I envision such a system:

- Silent Mac Mini/Apple TV like device with built in digital and/or analog video capture in order to function as a PVR

- A source of TV listings for every cable/satellite system that can be grabbed and stuffed into a database. MythTV does this with Schedules Direct. This seems to work fine

- A new remote with PVR/TV controls

- The complicated part: Apple will need to figure out what to do about Satellite systems that do not use RF, but instead use IR signals. Right now, I have a serial IR blaster installed that sends IR signals to my satellite set top box for when I want to change channels via lirc (lirc also allows me to control VLC via my remote control). Apple would need to provide this hardware, as well as an up-to-date database of satellite set top box remote control commands. They will need a setup app where people can select their setups from a list, perhaps an advanced interface where people can add their own TV service or remap their remote control buttons. There may be some legal restrictions with remote control reverse engineering.

- This may never happen due to Apple's interests in cozying up to TV execs, but like MythTV recordings should support commercial flagging so that commercials can be stripped out of recordings

- The hardware should support multiple capture devices like the Hauppauge PVR 350 so that users can record on a separate channel while they watch TV on another

- Like MythTV, this system should allow users to schedule recordings via the web, or via a lightweight client that can be run on an PC so that users can do this remotely

- An SDK so that third party vendors can add on things (e.g. there are telephony, weather, games, etc. extensions for MythTV)

- Support to connect a web cam to use as a security camera, maybe other home automation stuff


Clearly I'm thinking of MythTV here, but I certainly think that Apple can create a more just-add-water solution - the iPhone of PVRs.

It seems like there is some significant money making potential here, and Apple is already sort of stepping into this area with the AppleTV...

[mduell]

That's what, now the question is why? They can make more money at $2 a pop in iTMS.

[besson3c]

Originally Posted by mduell 

That's what, now the question is why? They can make more money at $2 a pop in iTMS.

You think so? The potential to capture a huge part of the TiVo market is less than their sales of TV shows on the iTMS?

[Laminar]

Well, with so many companies offering DVR solutions built in to the cable/satellite service, it seems that way is a lot easier for the consumer.

[wallinbl]

The whole concept of scheduled airings on cable/broadcast will eventually die off. Downloading seasons and episodes or streaming them is the future. Tivo is simply a patch to hold us over until the future fully arrives.

[besson3c]

There seems to be a lot of political stuff playing out with regards to time shifting (I think is the term they use for managing when you watch your favorite TV shows) too... I'm sure Apple has their own political interest within this mix too. My list of features is probably more fantasy than something Apple will, in fact, do, but politics aside I think Apple could build a successful money making product for themselves.

Then again, addressing what Laminar said, I don't know how common it is for PVRs to be bundled in cable/satellite packages.

[Chuckit]

All the cable companies I know offer DVRs, and DirecTV does as well.

[Kevin]

Cool for the bleeding edgers and geeks. But the average joe wont care. IHMO.

But cool idea non-the-less.

[moonmonkey]

IR blaster is a must, I always thought apple should make a universal remote control with IR blaster capability.
Then produce a range of plug / sockets / light switches which it could control.

That way my Mac could control my TV, Stereo, DVD player, lights in my home, AC and anything else without messing about with home automation systems - just plug and play.

Unlikely to happen but it would be cool, so much could be done with well designed IR blaster (or some RF control system) for the Mac to control household appliances.

[MacosNerd]

I don't see this having a huge market appeal, and the R&D costs to implement mean its a money losing proposition catering to the geek market or those who like to show off their high tech toys.

[besson3c]

These PVRs that some cable/satellite companies provide... are they pretty nice, or Mickey Mouse, restricted and locked down? Is there anything you'd really like to be able to do with your PVR, or something about it you'd like to change?

[lpkmckenna]

Right now, AppleTV complements a DVR quite nicely. AppleTV brings personal and internet content to your TV. The DVR brings cable and recording functions to your TV. I don't think Apple needs to reinvent the wheel and absorb the DVR into AppleTV.

If you want an Apple DVR, a Mac mini + EyeTV hardware is great. From what I understand, CableCARD has been a nightmare.

[besson3c]

I always had the impression that the EyeTV PVR stuff was pretty basic. Does it do all of the stuff I've described above?

[vmarks]

No, EyeTV does not do all that you describe.

There's a good reason for this.

MythTV has the benefit of open source to enable its developers to make it run on a wide range of hardware. It has the curse of open source in that every developer can do their own thing with it, and its up to the maintainer of the project to keep it focused.

MythTV suffers from an extreme lack of focus. Allow me to elaborate.

- Silent Mac Mini/Apple TV like device with built in digital and/or analog video capture in order to function as a PVR

- A source of TV listings for every cable/satellite system that can be grabbed and stuffed into a database. MythTV does this with Schedules Direct. This seems to work fine

Schedules Direct is just reselling the Tribune data. Apple would push it through iTunes, but I almost wonder if Apple would be their own aggregator rather than pay Tribune.

- A new remote with PVR/TV controls

- The complicated part: Apple will need to figure out what to do about Satellite systems that do not use RF, but instead use IR signals. Right now, I have a serial IR blaster installed that sends IR signals to my satellite set top box for when I want to change channels via lirc (lirc also allows me to control VLC via my remote control). Apple would need to provide this hardware, as well as an up-to-date database of satellite set top box remote control commands. They will need a setup app where people can select their setups from a list, perhaps an advanced interface where people can add their own TV service or remap their remote control buttons. There may be some legal restrictions with remote control reverse engineering.

Not complex. USB IR Blasters aren't hard at all. You just need to pick a good one for some range on the receiving end. The blasting isn't hard at all - it's just not elegant, what with the blaster IR leds having to be taped in front of other set top boxes. Not Apple like at all, extra wires.

There's no need for remote control reverse engineering.
It's simple. You either do the learning mode stuff (press 1, press 1 again, ok, now press 2, press 2 again, ok...)
Or you license the database from universal remote control MFRs.

- This may never happen due to Apple's interests in cozying up to TV execs, but like MythTV recordings should support commercial flagging so that commercials can be stripped out of recordings

It's not about cozying up to TV execs. It's about the legal repercussions of making such a feature commercially available. Remember that ReplayTV had to fight lawsuits over their 30second skip forward feature. Where is ReplayTV now?

- The hardware should support multiple capture devices like the Hauppauge PVR 350 so that users can record on a separate channel while they watch TV on another

Sorry, no. If Apple makes a system, Apple makes the complete widget, so you don't need any other hardware pieces to make it work. You can buy accessories and peripherals to enhance it, but it has to work without Hauppauge, Miglia, Pinnacle, etc., if it is to be an Apple system.

If Apple were to do this, they would end up (a) requiring customers to spend more money, make two trips to the store, etc., (b) end up absorbing support costs from these other companies. Bad.

Apple does not want to get into making TV Tuners. TV is hard. NTSC, PAL, ATSC, QAM, DVB, and DVB that uses h.264 instead of mpeg2 -- Apple makes one piece of hardware that works everywhere in the world. Apple has no intention of making all these different devices to try and suit each country's idiosyncrasies.

- Like MythTV, this system should allow users to schedule recordings via the web, or via a lightweight client that can be run on an PC so that users can do this remotely

Easy enough. Personal web sharing, etc.

- An SDK so that third party vendors can add on things (e.g. there are telephony, weather, games, etc. extensions for MythTV)

Macs have SDKs for other developers. There is no SDK for AppleTV or iPhone, and you have to work to get in on the Made for iPod program to be able to develop for iPod.

The ecosystems are closed because Apple thrives on ensuring what they see as a stable, consistent and consistently good interface.

- Support to connect a web cam to use as a security camera, maybe other home automation stuff

Here's the lack of focus I'm talking about. If you want to make home automation stuff, fine. If you want to make security systems, fine.

Neither of these are PVR applications, except in MythTV or LinuxMCE. If you want to make a home entertainment system, fine. Don't confuse it with other junk that it's not.

That's the trouble. It's easy to pile on features. It's hard to pare down to the essential features that express the totality of what the device or product aims to be, make those features work amazingly well, and then ship based on that. Simplicity is hard and hard to get right.

Clearly I'm thinking of MythTV here, but I certainly think that Apple can create a more just-add-water solution - the iPhone of PVRs.

A just-add-water PVR solution is absolutely the opposite of what you described when you started adding telephony and weather, games, security, home automation.

Take all the features you named.

Now decide which three are the most important, that you have to have.

Get just those three correct, and you'll have your just-add-water PVR.

Now step back, and look at the TV standards problem. You have customers who buy in different countries all the time, even when you tell them not to do so (iPhone exports and unlocks). You've got a support nightmare based on SD, HD, antenna, cable, satellite.

Currently if you're Apple, you have Elgato and Miglia both making these solutions, and dedicating themselves to dealing with all these issues. You're busy pushing iTunes and AppleTV and letting people watch what they want, when they wish, and if they happen to use Elgato or Miglia to get a TV broadcast into iTunes for AppleTV, you don't care.

Do you really want to kill that competition and take on all their difficulties and costs? Not if you're Apple, and you'd have to make tuners for every region.

Additionally, you have to compete with the cable and satellite companies who will be happy to sell you a DVR box for an extra 10 bucks a month.

So, the reasons why this won't happen:
(1) Bad business to require selling non-Apple hardware with the Apple product.
(2) absorbing support costs of non-Apple hardware mfrs? No.
(3) open ecosystem that will mean lack of consistently good interface? No.
(4) lack of focus. Security is not a PVR application.
(5) dilutes messaging on AppleTV and iTunes

The good news is this: You have the sources to MythTV. You can decide on the essential features. You can cut out the rest of the junk. You can fix the inconsistent interface. You can make it into just-add-water. You don't need Apple to take on this task.

Let me know when you start, I'll be happy to talk with you about it.

[besson3c]

Originally Posted by vmarks 

Sorry, no. If Apple makes a system, Apple makes the complete widget, so you don't need any other hardware pieces to make it work. You can buy accessories and peripherals to enhance it, but it has to work without Hauppauge, Miglia, Pinnacle, etc., if it is to be an Apple system.

Agreed, that isn't what I meant.. I didn't mean that the system should literally use Hauppauge hardware, I was just using the PVR 350 as a model.

Apple does not want to get into making TV Tuners. TV is hard. NTSC, PAL, ATSC, QAM, DVB, and DVB that uses h.264 instead of mpeg2 -- Apple makes one piece of hardware that works everywhere in the world. Apple has no intention of making all these different devices to try and suit each country's idiosyncrasies.

Good point, I hadn't thought about that.

Easy enough. Personal web sharing, etc.

Yes, it would be easy...

Here's the lack of focus I'm talking about. If you want to make home automation stuff, fine. If you want to make security systems, fine.

Neither of these are PVR applications, except in MythTV or LinuxMCE. If you want to make a home entertainment system, fine. Don't confuse it with other junk that it's not.

That's the trouble. It's easy to pile on features. It's hard to pare down to the essential features that express the totality of what the device or product aims to be, make those features work amazingly well, and then ship based on that. Simplicity is hard and hard to get right.

Yes, it definitely is hard to get right. What seems to inevitably happen with projects like this is once they become big and successful enough, there is a fork.

A just-add-water PVR solution is absolutely the opposite of what you described when you started adding telephony and weather, games, security, home automation.

True, but these pieces can be added over time once a focus and the core functionality has been established. I think that for the most part, this works with iTunes (even though it is now poorly named for what it does). What I was laying out was more of a long term picture type of thing rather than the blueprints for a version 1 project.

The good news is this: You have the sources to MythTV. You can decide on the essential features. You can cut out the rest of the junk. You can fix the inconsistent interface. You can make it into just-add-water. You don't need Apple to take on this task.

Let me know when you start, I'll be happy to talk with you about it.

I just finished building myself my first complete MythTV box (I was sort of using a FrankenMyth setup before under FreeBSD), although I'm still refining a few things and I haven't setup MythWeb yet. I take it you have experience with Myth as well?

[C.A.T.S. CEO]

There were rumors of the TV and Mini becoming one, as both of them are, seemingly, not doing as well as they could. I believe you can run full OS X on a TV so its only a matter of time till Apple combines them.

[vmarks]

catsCEO, the AppleTV will gain an iTunes store and not become a full fledged computer. That's outside of its focus.

[vmarks]

Originally Posted by besson3c 

Agreed, that isn't what I meant.. I didn't mean that the system should literally use Hauppauge hardware, I was just using the PVR 350 as a model.

My point was that for an Apple solution, you can't expect them to ship non-Apple hardware. The days of Apple printers being Canon underneath the covers are over.

Yes, it definitely is hard to get right. What seems to inevitably happen with projects like this is once they become big and successful enough, there is a fork.

Forks are good in that they allow people to do what they wish. They're bad in that they can distract focus and energy from the project.

True, but these pieces can be added over time once a focus and the core functionality has been established. I think that for the most part, this works with iTunes (even though it is now poorly named for what it does). What I was laying out was more of a long term picture type of thing rather than the blueprints for a version 1 project.

I would say that the iTunes Music Store isn't exactly named appropriately anymore. I would say that iTunes could have a different name to better represent its focus as the organization of all non-photographic multimedia on the computer.

Still, I'm okay with the name having stayed the same even as the purpose expanded from music to moving pictures.

I just finished building myself my first complete MythTV box (I was sort of using a FrankenMyth setup before under FreeBSD), although I'm still refining a few things and I haven't setup MythWeb yet. I take it you have experience with Myth as well?

I've done Myth on Fedora core 2 and ubuntu, as well as knoppmyth, using PCHDTV3000 cards (linux compatible conexant chips for HDTV) and ATI's HDTV Wonder. I've got some other cards and usb devices around as well, along with the Hava and HDHomerun.