Slashdot Mirror
Alliance for Linux Set Top Boxes
An anonymous reader noted this article running over at Linux Devices talking about an alliance of companies working together to standardize Linux Set Top Boxes. Bigger names include ATI and Tivo. There are also a bunch of more or less irrelevant companies on the list too so the hype about 24 companies isn't really worth noting. But in the end, I'll believe it when I see products actually taken to market.
Linux set-top boxes have already been classified into two categories: "vapor," and "discontinued." They've innovated on the previous business models by skipping the "product" phase and going right into "discontinued."
Pah. SLS may have been an early market leader, but the first distribution was MCC. My first distribution was MCC because I didn't have the bandwidth to download the (huge by the standards of the day) 65MB SLS distribution, nor the disk space to install it. MCC fitted onto 3 high density 5.25" disks, and installed nicely onto a 20MB hard drive. Before MCC, we just used a root and boot disk that came straight from Linus...
"The invisible and the non-existent look very much alike." -- Delos B. McKown
What happened with the oh-so-great plans of a Nokia-Intel-Linux box with DVB and MHP plus more that everybody talked about 'round 99?
:wq!
The page loaded, and what did I see?
OpenTV is the biggest platform in use in Europe, Liberate are real big in Europe (where, I think, digital-TV usage is biggest - UK, France and Germany have come a long way).
Pace builds loads of boxen, and who said Motorola and Sun (who also owns OpenTV) are insignificant!?
I've worked with developing digital-TV applications, and the current platforms suck so bad it aint even fun. MHP (http://www.mhp.org/) seems more interesting though, than OpenTV and Canal+'s platforms.
What's more, OpenTV development is based off of GNU stuff (libc, gcc et al) and they won't give away the source. After getting my story about this rejected on Slashdot a couple of times last fall, I went to RMS and had a conversation with one of their lawyers about it. Haven't heard anything since.
:wq!
Sounds fairly substantial, when you get down to it...
Agreed - the mention of licencing does seem rather odd, although if it results in the existence of cheap hardware with Linux driver support than that'd be a good thing (presumably GPL's software would soon follow - no need to use the software licensed by this consortium).
I'd also note that Linux already has the V4L/V4L2 video/TV API's, and that other stuff such simultaneous record/play via buffering was just patented by Tivo (hopefully it'll be overturned, but who knows), and the most obvious channel x timeslot TV guide format is patented by TVGuide / Gemstar... I've got to wonder what exactly they're looking to licence...
BTW, QNX is not a linux clone. Unix clone, kinda-sorta-maybe, but not linux.
Best Slashdot Co
I write code all day for the Motorola StreamMaster (tm) (r) (whatever) line of set top boxes, and I consider this FAR from irrelevant. Getting access to all the nice linux development tools would make me one happy camper. Motorola has a heavy involvement in the actual production of STB's, which suprise suprise run PPC chips :).
Sun makes servers.. but all the backend billing code for those STB's runs on - you guessed it - great big honkin' servers like Sun makes. So if they sign on, you get backend compatibility too, which is important - RPC et al.
ATI makes a LOT of video chips, and they do so cheaply. Their support matters from a driver perspective, and if companies like TiVo and other software producers for STBs (like the company I work for) don't sign on to develop for Linux, then your standard base won't mean much. More company support is always a good thing!
Notice Microsoft wasn't listed.. heh, this game is going to be Microsoft vs. Everybody else :).
..don't panic
I like the part where it says they will 'license' the api. It's neato that they want to use Linux, but it is just a kernel to them, they are unlikely to open any of their software up. Probably will have to join their organization to get access to the code, and, at minimum, buy a license to get access to the full api to develop your apps for it.
-Adam
You've got to admit,
the RIAA has balls.
Specifically, yours...
This sig 80% recycled bits, 20% post user.
The real issue is what the development enviroment will be. Linux would make a great basic box, but the key is what software enviroment is running and what standards it adopts. The biggest people in this area are Europe's DVB (adopted even by OpenCable in the US) and their platform http://www.mhp.org is based around Java. The box underneath is interesting, but at the end of the day the application enviroment is key.
An Eye for an Eye will make the whole world blind - Gandhi
The problem with standards is that they will be designed to support the lowest common denominator, limiting features. This is what I fear, not corruption of linux. The changes to linux will be small, and will probably be GPLd.
,sold to the cable provider, and designed so that you cannot access them. And these drivers/apps do not fall under the GPL.
Cable boxes are also closed by nature, and always will be. Communication happens between the headend and the box, and noone else is involved. Even if it provides internet access, that access will be using 1483 bridging so that no packets can be sent to the box from your home network.
All linux will be providing is the OS. All the driver's and Applications will be written by a dev company
Also keep in mind that these are embedded systems, meaning that most features like audio/video are not handled in software at all, but in a chip from ccube, ibm, vmlabs, etc.
For example your encoder question. It's not a question of what encoder they use for video. They won't even use software encoders. Somewhere there will be a farm of specialized video encoder hardware from Minerva or Pixstream/Cisco converting video into Mpeg2 streams which then gets shipped out over the cable network.
Sig:
Sig:
Barbeque is a noun. Not a verb.
uh, in what sense is QNX a Linux clone? i think maybe you're a little Linux-happy here. QNX is arguably a Unix clone, in the same some of the same ways that Linux is, but i've seen no evidence of QNX being a Linux clone. indeed, it's real-time, embeded nature targets very different things than Linux does, with very different priorities.
oh, and many of the important standards for the Internet or modern computing were contributed largely to by major companies. it's not so much the who's involved that's often problematic (although it certainly can be), but more often the nature of the process that's broken, resulting in broken standards.
i speak for myself and those who like what i say.
Just thinking back to the wars between distributions, Red Hat, who invented Linux
We didn't invent Linux, or even the concept of a Linux distribution. Red Hat Linux was the first distribution with a decent package system, and overtook the previous leader, Slackware. And Slackware wasn't the first one either... it replaced SLS as the leader earlier.
and brought it to the marketplace, has their own proprietary code contained within the source
That's not correct - our kernels don't contain proprietary code. The only proprietary code in Red Hat Linux is netscape 4.x, which we intend to replace with mozilla (already included, the question is when we can drop netscape 4.x without users complaining too much). Our kernels come with full source code - and if you look in the SRPM, you'll find all the patches nicely separated and categorized.Why do you think that they give pre-compiled kernels?
So we can be sure that the kernels are working and tested, and compiled with a known good toolchain, to name two reasons.
Net loss last quarter: $49 million.
Net loss a year ago last quarter: $23 million.
Cash and equivalents last quarter: $72.7 million.
Cash and equivalents a year ago last quarter: $124.5 million.
So, even assuming that their burn rate does not increase (though, the current trend is doubling from last year to this), they'll be broke by next January. Hmm, right about the same time as Webvan.
The only "intuitive" interface is the nipple. After that, it's all learned.
"The question of whether a computer can think is no more interesting than that of whether a submarine can swim" -EWD
So, none of that contradicts the historical earnings data I pulled (from TiVo's investor relations site, from their quarterly report).
Don't be so quick to call wild speculation "facts."
The only "intuitive" interface is the nipple. After that, it's all learned.
"The question of whether a computer can think is no more interesting than that of whether a submarine can swim" -EWD
QNX has been in reality BEFORE first Linux went out from LT hands... watch your facts - it helps...
this field has been intentionally left blank
Sort of like the god-and-nazis Usenet rule, as soon as the phrase "Set Top Box(es)" is uttered in a headline...
(Well, some of the subsections already have the WiReD color-scheme-cum-retinal-damage thing going on.)
:-) Kill your TV, and the box sitting on top of it.
--
News for geeks in Austin: www.geekaustin.org
News for Geeks in Austin, TX
although if it results in the existence of cheap hardware with Linux driver support than that'd be a good thing
That is, unless the application interface to those drivers is not V4L but instead a proprietary interface "protected" with DMCA-level (i.e. at least 8-bit XOR) encryption at the syscall level.
I've got to wonder what exactly they're looking to licence...
DMCA rights to the drivers may be part of it.
Will I retire or break 10K?
That nothing will ever come of it, because there are too many companies with too many of their own agendas. Do we need a standardized API of the sort their discussing? Not need, as such, but it would be a great thing to have. Will it get done by all these companies? I seriously doubt it. That's what happens when you get this many companies together... corportate politics comes into play, and everyone's trying to get "their standard" adopted so that they can sell more of their video chipsets, sound or modem chipsets, or cheese graters. Sad but true.
There are two major products that come out of Berkel
Oh, I wouldn't put it past Scotty Boy. One does get the sense that he'd love to see Sun move out of the market it's built for itself, into the consumer market -- how else do you explain Blade, which IMHO is a pretty blatant attempt to test the downmarket waters?
I think Sun wants into the consumer market very badly but so far has had to take it slow. (Heck, for all I know that was their intent with the failed Apple merger in '97...) They do have an in with Linux (LI members for how long? A couple of years anyway...) and roughly the same interest in promoting it as Apple does Darwin (i.e. selling the hardware to people who don't care for the OS), so that's no shock there. Buying Cobalt gave them some expertise in condensed systems design as well; I doubt it's that much of a stretch to go from a Qube to a set-top box.
So I think we'll be seeing Sun making noise about moving into the consumer market Real Soon Now. I doubt they'd go after the desktop, but the TV room or thin client (think SPARC-based iOpener)... hey. Could happen.
/Brian
Actually, that's precisely how Mandrake and TurboLinux (and Yellow Dog, for that matter) started out; they were all RedHat clones in the beginning. They've all grown in different directions, though.
I actually don't think there's anything wrong with this, though -- that's why the GPL exists. They didn't all copy RedHat, obviously, but a lot of them did.
/Brian
I'd say and/or is more like it, but y'know...
Still, like I said, look at Blade. Who is buying this? I'd bet it was at least intended as much for amateur techies as it is the target SPARC market. We know what Scott McNealy thinks of Microsoft, and I'm sure he'd take a piece of that pie if he could get it. I submit that it sure as hell looks like he's going to give it a try.
/Brian
Linux that is a vessel for JVM? Java? Linux in
itself is very capable crossplatform OS. Sticking
something that a corporation owns, into Linux
standart is herecy. java is great tool, however
making it part of specification begs for trouble.
Having Notice how the only definitive software
component of the whole thing is JVM. The rest
exists already, like Nvidia drivers...
Specification is statement of the obvious...
Companies like Lokigames donating SDL and alike
packages thus making their own standarts, being
proactive, rather than formalizing stuff that already exists there.
Such constrains will ultimately tie linux down,
in the future, just like it did Windows.
Windows could not been rewritten, because of the
large set of applications was depending on API
bugs went unfixed for years. Same is to come for
Linux as corporatoids demand higher profits with
less investment, that would be part of these
nonsense political groups swaying enduser crowd.
What made the linux are hackers, donations of
software under GPL, or BSD licences, and clear
documentation if code is not, same goes for *BSD.
In the end these groups will lay things out how
things are ought to be done, and thats bull, cuz
if I do code for my enjoyment, I will not listen
to *no* corporate head, part of the group of
people who like sitting around and telling
everyone what is the right thing to do. I will
just write code.
There's been no less than 5 Slashdot articles on this new box:
Nokia and Intel to make Linux-based Set-Top Box
Nokia and Loki Together on Linux Terminal
Nokia Media Terminal
Nokia's $400 Linux Terminal For The Masses
Nokia's Linux Based Xbox Competitor
It does lots of cool stuff: PVR (Personal Video Recorder, a la TiVo), MP3, web browsing, even games, and it'll probably be easily hackable too. It should be out sometime in the fall, and I'm really looking forward to it.
- Motorola - Major semiconductor corporation. Large range of CPU's
- Pace Micro Technology - Major STB manufacturer.
- STMicroelectronics - Huge semiconductor giant who make internals for many types of STB
- Sun Microsystems - A corporation that has server operating systems as a primary focus.
I wouldn't really say these companies were that irrelevant. They are certainly more important than TiVo and ATI.It's not surprising that companies see the advantage of not having to pay license fees to Microsoft. But so what? If this initiative is successful, it will mean more profit for the companies involved. It won't mean that you'll pay any less for your cable box or cable services. I'm sure the companies will appreciate all the free work that Linux developers do for them. It sure beats hiring programmers.
I dunno.
Why I dunno?
Because the article didn't say shit.
If the picture is reliable (yeah, right), then all they've standardized is the upper interface to the drivers. The right third of that is generic driver API; i.e., open, close, read, write, yadda, blah, etc.
So, putatively (and I ain't puting until I see it on 8x10 color glossy photographs with circles and arrows and a paragraph on the back of each one), you could add any HW driver you want. Provided you know the HW. And how to write drivers and alter the initialization scripts to turn off OEM drivers that try to open the same HW.
The definition of the interface strongly implies some definition of the function to be performed when you call across that interface. But no, that's not an interface spec, that's a software functional spec. Side effects. Some of which you'd hope they'd standardize. There's a big difference between a standard that says "calling int power_off(void) returns 0 for success and 1 for failure" and one that says "calling power_off() turns the power off". Big enough that it makes me think that the picture in the article was typical bogus hack-journalistic Not Getting the Point.
But. I still expect the standard API to be there. And the source code for non-proprietary drivers. Or else Limux might just have to start migrating certain things at a faster rate than the "standards" people can keep up with.
--Blair
Specs are a way to standardize a platform so that multiple efforts can strive for a common goal. Specifications, however, do not impede on proactive donations of GPL/BSD software.
Convergence Integrated Media, one of the companies involved, is contributing to Free Software via DirectFB which seems to be quite impressive.
In any case the corporate adoption of GNU/Linux as a viable platform should help spread copyleft fever. This is a Good Thing.
I can definitely see the danger with influential companies taking an "interest" in Linux, however people like myself that use Linux for our own purposes outweigh companies.
Suppose for a minute that one notable contributor decides that Linux should only run on a cerain platform. They shmooze kernel developers to drop other platforms and Linux 2.4.5 is the last Linux as we know it. Are we doomed? Nope. We have the kernel source so we can do with it as we please.
Another thing to consider is that here we are talking about an effort to produce an API, something which could benefit users and developers. X Windows has many of them, and nobody is forced into anything. I would much rather companies add APIs than fork Linux into things like: LinuxTV or LinuxIBM.
I don't know if you know but...
PII 350 (bargain price) 50$ (with mobo)
HDD 40Go 155$
A Shitty TNT2 30$
Hollywood + 40$
a Sound card : 25$
Linux (free for some, very expensive if I charge my hourly flat rate as a consultant, rate yourself)
=> 300bucks for a DVD-MP3 station.
You add 50$ for TV In and you have a Top line Tivo-Mp3-OogsRipping-PrOn station.
That you can upgrade as you wish
Please stop bothering me with Top Boxes that will come in 5 years, and of which we speak for 3 years.
I made my own out of despair 8)
It takes 40+ muscles to frown, but only four to extend your arm and bitchslap the motherfucker
Is "PeplayTV" actually a misspelling for Replay? I assume it is, but then I wondered if it was some unknown startup trying to ride on the coattails (and mis-typings) of Replay's users...
Does anyone know?
Also: what's with the "I'll believe it when I see it" cynicism? This is a *good thing* -- especially when TIVO, PePlay, and ATI are involved.
What, you want Microsoft to come on in with their odd UltimateTV and stomp all over anyone who's not Microsoft?
I don't see MS looking for a "standards-based" television platform. (Although I wonder if their .NET technology will someone play into Ultimate TV and XBOX. That might be somewhat exciting ...)
Whenever I see a bunch of large-ish companies getting together to discuss "standards", I get worried. A big conglomerate like this could have the marketing power to impose any standard it likes, even if it is not a good one. For example, what if they decide the "industry-standard" set-top box should use a closed-source encoder as standard? Or a GPL-free Linux clone like QNX? Too much power in this few hands is worrisome.
I'm the stranger...posting to
SLASHDOT: When News Breaks, We Give You The Pieces
-d.
--
Slashdot: When News Breaks, We Give You The Pieces
FreeBSD for the impatient.