Become a fan of Slashdot on Facebook


Forgot your password?

New Linux Set-Top Project 145

An anonymous reader says "Linux4.TV is the official open-source website for interactive set-top box technologies based on the National Semiconductor Geode? SC1200 integrated processor and SP1SC10 development platform." Looks to be an interesting project, although since they require registration to download stuff, I'll never know.
This discussion has been archived. No new comments can be posted.

New Linux Set-Top Project

Comments Filter:
  • First thing I noticed when I access this site.. "site launched Aug 22, 9:00 AM MDT". It's currently 8:50 MDT, so technically this site isn't available for another 10 min.
  • Linux combined with a there are even fewer excuses for the average lard-ass open source hacker to get outside and away from staring at a screen. Although...being able to watch an episode of Buffy while StarOffice starts up could be useful...
    • Can you name any good reason why the linuxites wouldn't attempt to cross linux with anything else left on the planet? I guess a linux-enabled can opener is a bit off the list of possibilities, then again...
  • Another cool Linux hardware project, some how I think I'll be hacking the X-box before this thing ever hit's the shelve's.
    Note my optimisim, I actually think it might come out.
  • Good God! (Score:3, Funny)

    by Reality Master 101 ( 179095 ) <RealityMaster101 ... m ['ail' in gap]> on Wednesday August 22, 2001 @10:56AM (#2203981) Homepage Journal

    I hear their mailing lists require registration as well. Those bastards!

  • since they require registration to download stuff, I'll never know

    Right, but you'll send thousands of your closest friends over to check it out. Way to stand by your principles.

    On a less serious note, is Linux 4 TV anything like AOLTV []?

  • GPL Violation (Score:1, Interesting)

    by Ex Machina ( 10710 )
    Perhpas we can get the GNU people to scream at these morons for requiring registration, Hello? GPL VIOLATION!
    • Re:GPL Violation (Score:3, Insightful)

      by DeathBunny ( 24311 )
      Read the GPL sometime. Requiring a registration to download isn't a GPL violation. As long as they provide source to anyone who gets binaries and they don't try to add any restrictions on re-distribution and modification they're not in violation.
    • Really? Which part of the GPL does that violate?
  • box mad (Score:3, Funny)

    by gavlil ( 255585 ) on Wednesday August 22, 2001 @11:02AM (#2204010) Homepage
    just imagine the redhat boxes will be tuned to nickelodeon all day, debian boxes to a mixture of simpsons and the science channel and slackware boxes will only work will black and white low res (but fast and powerful) 1920's bmovies. Of course the *BSD version will be hard to find and the MacOSX box will be so user freindly that u never lose the remote.

    an xp version will only show adverts and will crash after excess channel hopping (patch to follow)!
    • The OS X box will never show up -- I don't think anyone wants to relive the Pippin debacle...

    • Nah, u got the Windows XP version all wrong. It will run stable, but every once and awhile your neighbour will take control of it by typing 00000000000000000000000000000000000000000000000000 0000000000000000000000000000000000000 into his/her remote while its pointed at your house.
  • What with the recent proliferation in alternative hardware for future Linux platforms (I'm thinking here of PS2, Xbox) this sounds like a further extrapolation towards the consumer IT device. Surely the end-user expectation is to just plug the thing in and get on with it. Issues such as DHCP, Broadband / PPP etc will have to work very well, particularly in all target areas in which they hope to sell this stuff (e.g. regional differences).

    In what way do you think the Linux.4TV platform will simplify configuration and updates for the end user? With an environment that revs as frequently as Linux, a purely automated system that keeps the kit up to date would be a major boon, and very much at odds with the normal end-user expectation for electrical devices - they're great for a few months but before you know it, they're yesteryear's technology and the only option is to junk them.

    Will these guys be paying attention to security, too? An automated "keep it current, keep it secure" mechanism a la Ximian Red Carpet would be another major selling point.

    • Uh, they don't have to rev the software as often since it's

      1) On a flash
      2) There's not a lot there and few, if any services- it's intended to surf the web and provide a front-end for on-demand media services.
      3) It's not being used in a multi-user context

      Since you've got these three things, it's less likely that little Johnny Haxor is going to root this box by remote, etc.

      You don't need to update things if they're not broken for you- most of the updates are for functionality additions/improvements and for bugfixes. A set-top box or other appliance like device wouldn't need the updates as often.
  • I hear that some sourceforge projects also require that you be registered with them before you can contribute. They even require a login! I say we boycott every site that uses sourceforge for developement, in protest of this trampling upon our rights!
  • Hmmm ... (Score:2, Interesting)

    by jspey ( 183976 )
    It looks like National Semiconductor is trying to sell their hardware by getting Open Source programmers to write software for it. National Semiconductor wants to sell its hardware to resellers who can brand it and worry about bundling software with it so that it actually does something. Resellers will want to do this because they don't have worry about paying for software, since it will all be available for free.

    On the one hand, I have no choice but to applaud National Semiconductor and the people for coming up with such an ingenious way to leverage the open source community to make a buck. On the other hand, it bothers me that they plan to use freely donated software to make their product more appealing. It's one thing for companies like RedHat and Mandrake to charge you for the physical version of their linux distros, but somehow it feels wrong to me for a company to set up a situation where people will want to essentially write their software for them for free. It feels like they're setting a trap for programmers and baiting it with the increased acceptance and use of linux.

    I guess the bottom line is just whether or not they can save enough money buy getting free software to make their product cheaper than everyone else's, without having to use buggy code.

    Random thought: Since all the software is free, I suppose you could just buy a blank, unbranded model and install everythign yourself. Now that's an appealing idea, and truer to the ideas behind the open source movement.

    Mr. Spey
    • The Geode is designed to be modified, hacked, whatever. I've seen comments on slashdot before lamenting the lack of a "cool" form factor for embedded Linux projects. So here is one. I'm not even a programmer but I'm going to buy one just to see how many days I can go without sleep.

      The Linux4.TV website mentions the Geode as the reference model. Doesn't that mean someone can use a different platform if he wants to?

      Killer idea from National Semiconductor.
      • What I think they mean by "reference" is that National Semiconductor would rather sell the basic building blocks of the Geode to mass market companies (Sony, Panasonic, etc.) and have them assemble the final product. This would allow for different shapes and styles, a brand name and a logo in plain view, etc. However, in order to get people to start programming, National Semiconductor will sell you a plastic box with the geode inside of it.

        Mr. Spey
    • Dont assume that the 'open source programmers' are not going to be closely associated with the hardware manufacturers.

      You could view this as a way for the implementors to share code between them so the public gets a higher quality product. More in-house programmers from potentially competing companies pool together to write better code with public peer-review system. Plus programmers can move freely in the industry domain, since they dont have to spend months getting up to speed with code they have never seen before.
    • I think you have a point, but of all the companies you could be sponsored by, I think National is probably one of the better choices.

      National is a good choice because it no interest in any of the current sticky issues surrounding content protection, digital rights management, etc. They are going to make and sell chips regardless of what happens--either way, people are going to want set top boxes.

      They're going to make a killing in the near future off of the embedded processor market. Every digital TV, settop box, DVD player, and a host of devices that are only being blueprinted now are going to use embedded processors. From National's angle, getting a foot in the door now means unimaginable profits later.

      Is National using the open source community as a source of free code? Sure--but that's what it's there for. That is a good thing, because it will ensure that important things related to digital television are GPLed and not proprietary. The more corporations that follow this lead, the more otherwise-closed areas will open up as open-source. At this point in time, National's goals and the open-source community's are complementary.

      -National wants to sell a lot of embedded processors and get into the lucrative consumer television market.

      -Open source software must find a way into digital television and set-top systems, or risk being closed out by proprietary systems and closed 'black boxes'.

      So National provides the hardware, and the open source/free software community provides the software under the GPL. Is it a good thing to get too tied to one hardware vendor? Absolutely not. But is it good to have an ally against corporations like Intel, Sony, and Microsoft? Hell yes.

    • What's wrong with an arrangement where both parties benefit? I thought that was suppoed to be a GOOD thing. They get code, we get docs to write the code and the ability to use the product under Linux. They have an incentive to support Open Source, so that makes them more likely to do it.

      I want more partnerships between hardware companies and Linux, not less.

    • Look,

      Usually it is mostly "Plug and Pray"


      You buy a new DVD set wich can do on the fly VCR, frame grabs, DVD, MP3.

      You also are a White American with an IQ averaging 85.


      'cause most ot them are Morons when it comes to computers...AND it will boost your support costs if 1 out of 3 customers opens the box and fiddles!

      They know how to take it out of the pack (as in Beer 8) plug it to 2 devices (power AND tv) and then laborously decipher the start at "Push START") =>

      SOFT HAS TO BE OPEN, so your 14'old will come and Bash into it 8)

      If the soft is free, it cannot be resold if under some specific licences...maybe here is the thing devs could do.

      Personnaly, It took me 2 weeks, some friends and a brand new Hair Set to get a BAD, FewOptions, No Fast Forward DVD hooked on TV with Linux / PII 350 / DVD Card / Creative Labs / TNT2 Ultra

      To be noted I am not that good in Linux. I prefer windows 3.11 8) (flamebait. Okay. But You know.... you used it too 8)

      Now I am W98 + Power DVD + Winamp. And you know what ? IT WORKS !!! With Cordless mouse / Kb 8)

      Which is all I ask. I can see ripped Divx, DVDs, VCD, MP3, CD + Internet Access...

      Of course stability of 98 rarely exceeds 5 days... But if I can't see it in 5 days, I probably never will

      Then OMS ( will someday work and I will happily load it and cry softly over all those losts Hairs.

      Information Wants To be Free !!! But first you have to be able to read it somewhere...

      • "Now I am W98 + Power DVD + Winamp. And you know what ? IT WORKS !!! With Cordless mouse / Kb 8)"

        I just got the same setup done for my girlfriend except add a sterio, VCR & TV. She loves it too.

        IANAP but I probably will by some version of this whenever some software is available for it. As long as it will talk to the rest of the machines in the house, I'll be happy..

    • Go to (Score:3, Informative)

      by Svartalf ( 2997 )
      There, you can buy the bare machines in bulk...

      Realize, however, that a Geode isn't a speed demon- never was.

      National has yet to sell any set-top solution and most of the set-top vendors have yet to make a dent in things.


      Because the software's nothing like a PC- even though people don't want a PC, they want the Web to look and act like the PC on the box. It's also because they've been using underpowered chips like the Geode to save on costs (The bill of materials on most Geode based boards is ~$100US.)

      That's about to change...
  • Interesting.... (Score:2, Interesting)

    by Kadin2048 ( 468275 )
    Site got /.ed before it even officially opened? I guess that's impressive.

    They seem pretty set on using the SC1200 semi, but never really say why...I mean, there are a lot of chips to choose from, and I would be interested to know why the SC1200 is the one to use. Oh...okay. I get it: National Semiconductor [] is one of the sponsors.

    But that's not bad, though. Personally I've had good experiences ordering parts from National. I wonder, when they get this done, could you choose to get the schematics and parts lists and build it yourself? Or maybe just get the printed circuit boards and key components as a kit?

    I read something yesterday that said that TiVo ( []) used GPLed code in their set-top system, and released the source code to the public. I am looking around on their site right now and haven't found it yet, but I would think that Linux4tv would be silly not to try to use it, if possible.

    Their web site doesn't give a whole heck of a lot of information on it, but I'll give them some slack, seeing as it only launched "officially" twelve minutes ago.

    I like the .TV TLD, though. Wonder if it was a bargain or if they had to pay a premium? And why the heck do you have to register to download?

    • I have a TiVo - IIRC they made some modifications to the kernel which they have released (I believe for better real time support and to support the filesystem) under the GPL; however the main TiVo program is NOT GPL for the simple reason that they need to protect their subscription payments.

      I'm not sure which came first, National support or the choice of National IC, but one reason could be that to make a reasonable set-top box they need a chipset with a lot of the required features of a settop box to keep the need for extra ICs down. One feature of such a box needs to be that unnecessary costs are kept to a minimum.

      P.S. Is the IC chosen a CPU, microcontroller, or just an IC which requires an external CPU ?
      I'm too lazy to RTFM!

      • Here's the info I got on the Geode SC1200, which is going to be at the heart of the Linux4tv. It's actually quite a little chip....more info available here. []

        (This ripped mercilessly from the above URL.)

        • 32-bit x86 processor, up to 266 MHz, with MMX instruction set support
        • Memory controller with 64-bit SDRAM interface
        • 2D graphics accelerator
        • TV controller with hardware video accelerator
        • CCIR-656 video input port with direct video for full screen display
        • PC AT functionality
        • PCI bus controller
        • IDE interface
        • USB, three ports, OpenHCI 1.0 compliant
        • Audio, AC97/AMC97 2.0 compliant
        • Virtual System Architecture (VSA&#153) support
        • Power management, ACPI 1.0 compliant
        • EBGA package
        • 32-bit x86, up to 266 MHz, with MMX compatible instruction set support
        • 16 KB unified L1 cache
        • Integrated Floating Point Unit (FPU)
        • Re-entrant SMM enhanced for VSA

        It's got IDE support, PCI, serial and parallel, video capture. Quite an impressive device. I wonder if once the Linux4tv guys get a distribution for it, whether it would be possible to buy a few and create a low-cost parallel processing array. Of course, I haven't seen a price on it yet...

        Thanks for the info on the TiVO GPLed code, maroberts.

        • Thanks for the info on the TiVO GPLed code, maroberts C'est rien.

          Sounds like the device has the bits and pieces to do everything required. The TiVo (again IIRC) uses a mere 50MHz PowerPC chip [being RISC, this probably delivers similar performance to a 200MHz Pentium, but a lot of TiVo reverse engineers believe the Tivo gets very near to its processing limits at times] with a number of support chips.
          If you're really interested in the TiVo, see the ongoing discussions in the TiVo AVS forum:
          Also Andrew Tridgell (of Samba fame) has done some engineering to get the TivO to take a network cardand other goodies. See
          for more details.
      • A little bit of everything, It has a integrated 32bit X86 compatible with MMX support(but how fast is it?? the pdf @ sc1200 one) says it runs up to 266mhz), named GX1. GX1 has a integrated : 2D graphics controller, a 64Bit DRAM interface, PCI controller.

        Besides the GX1, the SC1200 features: a integrated videoprocessor(blending/overlaying, scaling, filtering, TV encoding) , a IDE interface (ATA33 though), CCIR-656 video input port(kinda cool), USB, Power management, IO contr(IR, parallel, serial), audio features, and a clock!

        Check this info out:

        SC1200 []

        GX1 []
    • From

      "The Overlay subsystem API is used to do hardware overlay of video data on graphics data. This is mainly used to display video coming to the frame buffer through the Video Input Port (VIP) in the SC1200/SC1210 Geode Multimedia Processor. The input to the VIP can be from the TV tuner, or an MPEG decoder. Along with overlay, this subsystem can also do alpha blending of video data with graphics data.

      Quite usefull for a settop :o)

  • The black box shown on their site looks like a PC to me. I wouldn't like to try balancing it on a TV, especially not a dinky European TV. Now before someone points out that the picture is the development platform, this begs the question of what the target STB will be like. Consumer electronics has to be living room and family friendly, not some big black technophile status symbol. Reminds me of the Xbox - try lugging that around to your friends house on a pushbike.

  • Looks to be an interesting project, although since they require registration to download stuff, I'll never know.

    To me, this is a childish and silly reason to not get involved in a project and just shows the pig-headed mentality of Taco and those like him.

    I don't want to drive a car, because that would mean giving my address to the government so I can get a license! Sounds pretty stupid, huh? Yep.
  • I'm sure that SOMEONE will take care of that registration for you so that you won't have to.

    "You've got mail!"

  • Just compare [] and []. Seems like the guys at Century Software are trying to get someone to program software for free so they can sell their set-top boxes later.

    BTW, simliar projects have been around over here in Europe for quite a while - and those require no registration to download software and specs. Just have a look at [] or [] ;-)

  • As someone who developes specifications for hardware and software for the cable industry, there are several issues that are not immediately addressed on the web site, but perhaps are addressed after registering for their various groups and getting the download files. The first is that for cable, interactive set top boxes require that have reverse channels, require a hightly optimised OS that can handle perhaps a couple of hundred different types of interrupts. Additionally, for cable systems, conditional access is an absolute requirement. For retail boxes, this will come in the form of removable PCMCIA devices that are specific to the particular cable operator.
    • I'm looking forward to seeing how they manage conditional access. I haven't been into the groups or register-to-download portions of the site, but from what I've seen so far, you might not have to deal with anything related to the cable system, if the Linux4tv just plugs into the output from whatever decoder you already use. But if you do that, it's basically just a glorified VCR.

      In my area, the cable converter is a large black box the size of a VCR (if you have digital cable) manufactured by General Instruments. Cable goes in, video feed goes out. How would a device like the Linux4tv work in a situation like mine? Unless the L4tv had a way of talking back to the big converter box, it wouldn't have any way of changing channels. And since the converter doesn't have any controls other than the remote (no control ports on the back, I checked), I don't see how you would get a set-top box to work.

      Anyone have any ideas on how they plan to do it?

      • If it's a digital video feed, a la digital cable (DVB format, etc.), they've got DVB cards that will handle everything except the conditional access part; that would be handled by a smartcard reader on board with the decryption keys- just like for DirecTV and Dish Network.
    • A cursory look through the Web site turns up a link to the CableLabs DOCSIS cable modem standard, but nothing about the recently released Open Cable Application Platform (OCAP) spec. Open Cable is the umbrella project which has already set standards for the mentioned PCMCIA conditional access hardware. OCAP attempts to specify a processor- and OS-independent middleware for set top box applications so that the cable companies and their partners can develop services that run on any compliant retail digital box.

      OCAP is currently based heavily on HTML, ECMAscript and Java. Microsoft is understandably lobbying heavily to have Java excluded. One of the more interesting requirements in the spec is that the box must run software provided by the local cable company that can control which other software can have access to low-level resources like the tuner.

  • Looks to be an interesting project, although since they require registration to download stuff, I'll never know.

    C'mon, Taco. These things would probably make a kick-ass MP3 jukebox. They'll sit right by the stereo and use a remote with more buttons than a Macintosh's mouse.
  • Is this complementary to or in competition with the TV Linux Alliance []?

    The latter seems supported by a wide variety of companies.

  • So how difficult/expensive is it to get a hold of one of those SP1SC10 development platforms? Say for an individual interested in hacking with some cool hardware, as opposed to a company developing a commercial product. I didn't really see anything on the website other than "contact a representative".
    • I have a feeling that these chips are probably pretty pricey, and they won't be exactly willing to send a lot of them away as samples.

      If you really wanted to get your hands on a freebie, a good way might be to get involved in coding for the Linux4tv project, and then see if they'd send you one. If my programming experience included anything other than AppleSoft BASIC, I'd probably be over there signing up now.

  • So I guess I have to ask, "So, what's new?" This has been going on at [] for over a year. And they have a link to GCT-Allwell [] who offer a number of links to completed projects.
  • by Kagato ( 116051 ) on Wednesday August 22, 2001 @01:04PM (#2204605)
    Problem is there isn't any 8vsb (Read HDTV) support in the platform yet. Which makes me question the long term vision of the project.

    Right now most of the current crop of 8VSB computer hardware is based on the TeraLogic Janus platform. While the platform is certainly offers nice features and would be easy to integrate with any open source STB I doubt it will never see the light of day in Linux. Two years ago Teralogic was all over Linux. They sent mail to various lists, offered sample cards. Times have changed. Asking about linux drivers or even docs gets curt responces from the devel team. [Insert standard MPAA rant here].

    So, while an actual opensource platform is a cool idea, certainly giving more features than Tivo or WebTV, long term, I just don't know if it's worth while.
    • No one wants it. Cables operators won't be carrying it becuase it costs them more $ and takes up several standard channels. When the dealine hits and 0.01% of all televisions are HDTV capable, the deadline will be put off indefinately.
      • Not to burst your bubble, but the facts don't match up to anything you said.

        First off, many cable carries in the top 15 markets are already catering to some form of HDTV. And not Just HBO-HD or Showtime-HD. For instance the Time Warner cable system in my area made a deal with the PBS station to carry 5 DTV feeds. I personally have DBS, which has a small, but expanding selection of premium content that compliments the Over The Air content I recieve.

        Second, 0.01%... Man, you're on crack. First off most Front Projection televisions sold in the previous 5 years could be cheaply ungraded to DTV. Most medium to high end data grade projecots (LCD, DLP and the like) can handle DTV from low cost tuner box. Direct View DTV units are falling on a monthly basis, and the price gap in Rear projection DTV is becomming fairly slim.

        Sales have been weak, but when you consider that many of the early adopters most likely already have an HDTV compatible display, then consider that major hardware makers have been having to go back to the drawing board to put in place MPAA mandated Content Controls. Yeah, sales could be better, but you're still looking at several hundred thousand units in the channel right just this year.

        Point is this. HDTV is not like the Metric system. The only people with a vested interest in changing was sign painters. Every major consumer electronics maker is banking on DTV. The FCC is banking on selling off the old bandwidth.

        HDTV is here, it's not going anywhere, and dispite your opinion, many of us are already enjoying HD content.
        • Hate to burst your bubble but to my knowledge the whole world with the exception of the US has switched mostly if not entirely to metric. Somehow I doubt we're all sign painters. Of course, Americans like to be different and I applaud that, but when billion dollar NASA projects do dumbass things because people still do calculations in Imperial (which is the name of the system you use in that non-imperial democratic republic of yours) then convert to metric (badly), some changes need to be considered.

          Anyway, my apologies for the rant. I agree with you regarding HDTV though here in Canada we still have no sign of it from cable companies. Ours has said that they estimate 5 years before they even attempt at HDTV support.
          • I don't want to come off as a prick, but you didn't burst my bubble. But you're wrong on both counts. One, the US was not the only country that didn't switch to Metric. For example, the UK still has good old MPH on the speed limit signs. And no, we do not use the Imperial system. For instance, an imperial gallon in the UK not the same size as an US gallon.

            Although Cable in your part of Canada may be behind, the DBS system in Canada is actually ahead of the game. Even better than US DBS systems actually. You can get the big three US networks in HD, as well as showtime, HBO, etc.
        • Sales have been weak, but when you consider that many of the early adopters most likely already have an HDTV compatible display, then consider that major hardware makers have been having to go back to the drawing board to put in place MPAA mandated Content Controls. Yeah, sales could be better, but you're still looking at several hundred thousand units in the channel right just this year.

          Several hundred thousand <gasp!> it's a drop in the bucket. The Content Controls will kill HDTV for good. "What do you mean I can't tape Buffy the Vampire Slayer?"

          The electronics industry may be staking their future on it, but unless people have a real *reason* to HDTV, it just ain't gonna happen. If the FTC tries to turn off the old broadcast stations before 98% of the people are ready, there will be riots in the streets.

          I see all the local electronics stores heavily promoting HDTV equipment. But no one is buying.

          Enjoy your HDTV. I'll get one when I can tape my favorite shows.
          • This is the first year HDTV hit widespread comsumer electronic stores for a wide release (i.e. you could go into a top 20 market Best Buy or Circuit City store and walk out today with a DTV.) Sales of DTV's is similar to year 1 of DVD players. By your logic there is no reason DVD sales should have taken off. The only compelling reason for purchasing a DVD player was quality. You couldn't record with it. The media (although cheaper to make) is not actually cheaper than VHS. The is the only compelling reason for purchasing a DVD player or HDTV is quality.

            Content Control is a problem. Two of the last three HDTV content control articles on slashdot were posted by me. So know this very very well. There are some brights spots to the Content Control problem. Mostly being that it's against the law to encrypt an Over the Air TV broadcast. Thus it comes down to the hardware vendors honoring the flags sent with the stream. Apex 8VSB tuner and VCR anyone?

            Third, it's the FCC not the FTC. And no one is going to riot. Over 80% of the US gets it's Television via cable TV, another 7% get it via DBS. Cable companies are One) Trying to sell digital content. Why? More money because you can force people to lease boxes. Two) they are going to down convert the OTA channels for the customer anyway.

            I used to work retail electronics. Although I'm out of the field I still know people who are selling electronics. And they don't agree with you about "no one is buying".
  • There have been many set top boxes based off of Linux code that have failed. This is nothing new and will prolly fail as well since no set top boxes in development will make it.
  • The box problem (Score:3, Insightful)

    by Animats ( 122034 ) on Wednesday August 22, 2001 @01:19PM (#2204678) Homepage
    What's really needed is a generic set-top PC platform produced in volume. Then there will be something worth developing software for.
    What would you want in such a box?

    • "No User Serviceable Parts Inside". The end user doesn't open the box. This is TV, not PC.
      So no PCI slots. PCMCIA slots, maybe.

    • Several disk drive bays, with disks installable by non-technical end users. (Device Bay, the packaging spec for plug-in 1394-compatible drives, would have been great for this if anybody used it.)

    • Set-top-box form factor. Slim, not PCI board height.

    • Well-shielded tuner section. Must work in fringe areas with poor antennas. (VCRs have this now, but most tuner cards for PCs don't do it well.)

    • No matter what, it always boots up. If a disk gets corrupted, or a hard drive crashes, you can still watch TV, even if you can't record. The base software belongs in a PCMCIA ROM card.
      (OK, developers can use a flash card.)
  • Etherware (Score:2, Insightful)

    by sabinm ( 447146 )
    Sorry. Not going anywhere. No one can do a single thing w/out bandwidth.
    Imagine a flight attendant offering you a inflight movie while you were sitting in an airport with a 5 hour delay. A great service, with nowhere to implement it. ;)

    Remember those cool AT@T commercials, where those cute kids pushed a button on their TV screen and out pops "Star Wars" immediately, on demand?

    Have you ever .... You will! And the company is going to bring it to you is AT@T.

    That was 7 years ago. I don't have any illusions about all these cool new ideas. I know exactly where they are heading w/out the infrastructure to support them. Cool idea heaven, somewhere is Vaporville.

This login session: $13.99