...but will it run M.A.M.E. ?

Wow. Impressively crappy example from Born, there. The only reasons the IBM PC was an "open platform" was that other companies cleanroomed the BIOS, opening the door to PC compatibles...and that IBM conceded to Microsoft the right to sell MS-DOS for other platforms. If IBM had had their way, the PC compatible movement would never have existed; IBM had no choice in the matter.

Looks nice and shiny. Shame no inbuilt DVB-T. (Analogue television is so last century) I suppose I could get one and some USB DVB-T tuners, if there is any software out there for them.

Note that the thing that ships now is development kit only, I wouldn't consider it ready for "ordinary" users just yet.

Now, if only they added DVB-T support (it's near useless without it for me, analog TV is being shut down in this country later this year) too, I'd be buying a couple of them right away...

JBURKA, you are certainly right, IBM didn't do it on purpose, but nonetheless they still did it and it changed the industry. The reason that I use that example is that too often "open" gets confused with "open source" which often gets relegated in people's minds to a small niche. The point I was trying to make is that "open" is really significant and can have profound industry ramifications.

And nevermind that Visicalc was originally written for the Apple II, not the IBM PC.

It does seem like analog to digital video encoding is a dwindling market, especially with the impending broadcast switch. And only 720p playback ("up to 1080i with certain content")?

On the other hand, I do like that Neuros provides low-level access to the device. I have a similar set-top box, the Popcorn Hour A-100, and I've been very disappointed and frustrated with the open-source lip service, where the most important functionality (media playback) is handled by a monolithic closed-source app. But I suspect it has to do with the chipset providers: dominant TI is less threatened by openness than small player Sigma is. (Although I understand it took some pressure from Neuros for TI to provide sufficient public specifications; is this correct?)

And I'm curious: what industry ramifications has the original Neuros OSD had?

hmm, I can put together decent Linux media center running XBMC for $250, and it will decode and output Full HD

#3: There are indeed USB DVB-T tuners with Linux drivers and application software support; I've been using one here in Australia for a while now, although I haven't yet tried to connect it to the Neuros OSD I got thanks to the BB Gadgets competition.

#6: Yes, but will it encode and input full HD?

#6 Neuros had to put considerable pressure to convince them to open up some parts of its software. For the OSD1 the open package was missing several significant pieces that remain closed to this day. For OSD2 it's much more open from the get go, however there are still some closed parts (one kernel module and a library). Pressure is still being applied and there are news these parts are being opened up as well (a kernel module that was closed was opened up some weeks ago, so there's hope). The only thing that will never opened is the DSP codecs. But Neuros and TI are still joint sponsoring work on an open Ogg Theora implementation for that DSP.

#7 please describe how would you do that, too

#8 and #3
you can use an USB DVB-T tuner, however i think you need to port the drivers if they are not available for ARM architecture yet. The USB port(s) are up to the task as far as i know.

In any case the device doesn't even have an analogue tuner, so it's not fair to say "it's analogue only"... since it's neither analogue nor digital TV device -- it just happen to record any signal that you feed to it, whatever the source, but it will never pick the signal off of the air anyway unless you add some external tuner.

#8: No, the PCH will not encode video. But I don't see how, without any tuner, video encoding is a necessary feature of a set-top box. If I have a an analog video source, I can just plug it into my TV. If I want to record live video, I can use a digital camcorder or webcam. I could see encoding, say, a VHS collection, but why not just use a video card in a PC?

#7: Part of the appeal of this class of devices is that they are small, quiet if not silent, and have low power consumption. This is harder to achieve in a $250 linux pc-- but educate me if I'm wrong!

#9: So what kind of obvious functionality is precluded by the closed source portions? For the PCH, it's rather extreme: the only way to play audio is through a closed source app which also claims the display, making it impossible to do anything else (like browse through playlists) while playing a song. Although people have made UI improvements, they can't get around this feature, making the PCH nearly unusable (imho) for listening to music.

I'm all for open hardware (often guised as "reference designs" -- which are unspeakably common across different "badging"), but I'd sooner buy a DreamBox for DVB-S2 on my LAN.

#6 regarding ramifications, it's really hard to know, we work closely with a number of silicon suppliers and they are surprisingly attentive. TI has given us a disproportionate amount of mindshare for the customer we represent. Remember they support the likes of Nokia, Motorola, etc the largest names in the business. I have no doubt that many others are watching the phenomenon of open electronics to see what happens, but it's a bit soon to see those effects

#10 there's not a lot of functionality precluded at least for the kind of stuff that you are talking about. The codec and framework is just literally the stuff that handles decoding the video. Even the file parsing and putting together the audio and video and subtitles are open. Things you can't do are things like optimizing the algorithm that does the decoding or encoding to enhance it's capability.

BTW, there is a beta signup list here to apply to become a beta tester for the device when it's consumer ready (or getting close)

Zuzu- As always, I've learned something interesting from your comments. Thanks!

Joe Born- Thanks for the response. I'm interested to see where this leads.