So what about the GPL license you ask? Yes, the Helix DNA Client (the FOSS media framework which supports any format and any operating system) is licensed under the GPL license.
Sure it is, but none of the codecs are. So it's 100% worthless.
There are zillions of "frameworks" avalailable already. It's the codecs we need, and Real still requires their commercial license for those.
Not likely. Since the Gmini cpu (CalmRisc) uses a harvard architecture, it cannot support loadable plugins. All code must be compiled in and I'm afraid RockBoy won't be very high on the priority list.
The wikipedia entry is correct. Just because you don't know about it doesn't mean it doesn't exist.
One of the more famous examples is Dr. David S. Touretzky's "Gallery of CSS Descramblers", which contains more than 20 different examples of code that is (assumed to be) illegal under the DMCA.
The page also prominently displays Dr. Touretzky's name, email address and a photograph of him. It was explicitly created to draw attention to the absurdity of the DMCA law, through civil disobedience:
If code that can be directly compiled and executed may be suppressed under the DMCA, as Judge Kaplan asserts in his preliminary ruling, but a textual description of the same algorithm may not be suppressed, then where exactly should the line be drawn? This web site was created to explore this issue, and point out the absurdity of Judge Kaplan's position that source code can be legally differentiated from other forms of written expression.
Changing the definition of time_t is an API change, which naturally carry consequences. You didn't think the 2038 issue could be fixed entirely without effort, did you?
A possible and rather simple solution to the time() issue is to change it to no longer return time_t, mandating the use of the 't' pointer parameter, thus freeing up the return code to be used exclusively for error codes. Then you can grep for calls to time(NULL), change those and be done with it.
You want a fix to all computers that does not involve actually changing anything on these computers? That sounds pretty difficult.
The point is there is one single thing we need to do to fix the issue. Y2K was a nightmare where each system had a different set of problems with a different set of solutions. 2038 is much much simpler.
2.2.2.77 seconds since the Epoch: A value to be interpreted as the number of seconds between a specified time and the Epoch.... If the year 1970 or the value is negative, the relationship is undefined.
Apart from these open source license issues you also have to pay attention to the involved patents of the AAC and MP4 formats that are handled by the licensing administrators Via Licensing.
Who needs to license MPEG-4 AAC patents?
An MPEG-4 AAC patent license is required for manufacturers or developers of complete (or substantially complete) end-user encoder and/or decoder products.
What kind of weird world do you live in where a patented algoritm requiring royalties equals "an open standard"? These patents mean Apple is free to harass and sue anyone who implements and/or distributes AAC without paying the Apple tax. This completely excludes all free and open source software. Why do you think FAAC are afraid to distribute binaries?
While we appreciate positive attention for our work, this story is a bit early. We have just begun to look at the iRiver iHP/H1 hardware and are quite a bit away from having anything of significance to show (such as running code).
I'll try to preemptively answer some common questions:
- No, we are not violating the DMCA or any other intellectual property laws. We are only distributing software written by ourselves and we run it on our own hardware. Our software does not circumvent any access control or copy control mechanisms.
- We are not doing this to "expand our market share" or any other weird corporate-style reason. We are doing it because our old Archos hardware is becoming obsolete and hard to find so we need to find new hardware to run our software on. The fact that the iRiver has a large user base is a bonus though, since it means more potential contributors.
- We are not looking at the iPod or Rio Karma since they contain a chip made by Portalplayer that you have to sign away your firstborn to see the docs for. That is a silly practice we do not wish to encourage. The iRiver contains hardware with published docs.
Feel free to drop in on irc if you have any questions.
Be careful where you throw those stones. There's glass nearby...:-)
The original expression "I couldn't care less" is British and dates back to WWII. The distorted version "I could care less" started floating around the US in the 60s. It is very illogical and prone to misunderstandings and nobody really knows why or how it happened.
"It's actually a very interesting linguistic development. But it is still regarded as slangy, and also has some social class stigma attached. And because it is hard to be sarcastic in writing, it loses its force when put on paper and just ends up looking stupid."
Guess what? EVERY USB Mass Storage Device uses FAT for storage.
Nonsense. USB Mass Storage is the SCSI layer and has nothing whatsoever to do with the file system. How do you think USB DVD burners could work if they had to use FAT? And do you really think all USB Harddisks have only FAT on them?
USB Mass Storage devices can have any file system, just like any other storage device. The fact that many (but far from all) manufacturers choose to use FAT is for obvious interoperability reasons, not because USB defines it so.
Actually, gapless mp3 playback has worked in Rockbox for a long time if your mp3 data doesn't contain gaps.
Use the --nogaps option in lame to create gapless mp3 tracks. Of course you also need a player that actually plays the frames back-to-back, and doesn't insert artificial pauses between tracks. I've heard conflicting reports about the Ipod in this regard.
Not future tech - my ISP offers it *now*
on
150 Mbit/s DSL.
·
· Score: 2, Informative
Why are you slamming ATI for releasing binary-only drivers, while hailing Nvidia? Nvidia does exactly the same thing. What do you think the 1MB 'Module-nvkernel' file in their NVIDIA_kernel-1.0-nnnn.tar.gz is?
NVIDIA_kernel-1.0-2960> file Module-nvkernel
Module-nvkernel: ELF 32-bit LSB relocatable, Intel 80386, version 1 (SYSV), not stripped
You didn't seriously think the few snippets of C code in that package was the complete driver, did you? That's just a kernel wrapper for their binary blob.
Reverse engineering means examining a product to find out how it works. Disassembling the firmware is merely one tool used in that examination. Oscilloscopes and logic analyzers are other tools we have used.
We have written every single byte of the Rockbox firmware. But we could not have written the software without first researching how the hardware was put together, i.e. reverse engineer it.
All of this is completely legal.
If you define "from scratch" as writing software without first researching the surrounding interfaces, then no software has ever been written from scratch.
The difference is that the purpose of the CSS scrambling is to prevent you from reading the files on the DVD and save them on another disk (or play them in an "unauthorized" player). CSS is thereby a copy protection scheme, which DMCA forbids circumventing.
The purpose of Archos' scrambling is obviously not to prevent copying, since the files are very much copyable, and Archos themselves freely distribute the files on the Internet. Thus, descrambling it does not circumvent a copy protection scheme.
Car CD changers are very simple to handle in software, but we naturally can't do the hardware ourselves since we don't have the head units.
SID support is perhaps possible. The main bottleneck is that we must encode the sound data as mp3 before feeding it to the sound chip. Unless, that is, someone manages to write a new DSP codec to play uncompressed PCM data.
We can play any format if only we can write code for the DSP to decode it. The MAS 3507 (and 3587) are generic DSPs that simply have MP3 codecs in ROM. We can download new codecs in them and I'm the first to hooray if we can get OGG or FLAC or anything into these DSPs.
Unfortunately, we have no docs or tools for writing new MAS DSP code and Intermetall is very secretive about it. If anyone can help, please get in touch!
The 12 MHz SH7034 CPU is unfortunately much too slow to handle decoding in software.
Slashdot Mirror
Sure it is, but none of the codecs are. So it's 100% worthless.
There are zillions of "frameworks" avalailable already. It's the codecs we need, and Real still requires their commercial license for those.
Not likely. Since the Gmini cpu (CalmRisc) uses a harvard architecture, it cannot support loadable plugins. All code must be compiled in and I'm afraid RockBoy won't be very high on the priority list.
One of the more famous examples is Dr. David S. Touretzky's "Gallery of CSS Descramblers", which contains more than 20 different examples of code that is (assumed to be) illegal under the DMCA.
The page also prominently displays Dr. Touretzky's name, email address and a photograph of him. It was explicitly created to draw attention to the absurdity of the DMCA law, through civil disobedience:
Changing the definition of time_t is an API change, which naturally carry consequences. You didn't think the 2038 issue could be fixed entirely without effort, did you?
A possible and rather simple solution to the time() issue is to change it to no longer return time_t, mandating the use of the 't' pointer parameter, thus freeing up the return code to be used exclusively for error codes. Then you can grep for calls to time(NULL), change those and be done with it.
Read what I wrote: Negative time_t is invalid and you should be spanked if you've used it in violation of that.
You want a fix to all computers that does not involve actually changing anything on these computers? That sounds pretty difficult.
The point is there is one single thing we need to do to fix the issue. Y2K was a nightmare where each system had a different set of problems with a different set of solutions. 2038 is much much simpler.
2.2.2.77 seconds since the Epoch: A value to be interpreted as the number of seconds between a specified time and the Epoch. ... If the year 1970 or the value is negative, the relationship is undefined.
Wrong, that's the last second in 31-bit unix systems!
The 2038 limit is way overhyped. The only thing we have to do is change the definition of time_t from:
typedef long time_t;
to:
typedef unsigned long time_t;
And we can merrily keep using it on our 32-bit systems until 2106.
POSIX disallows negative time_t anyway, so if you've used it you deserve to have your system break.
(This rant is a dupe since I said the same thing here four years ago.)
Correction: Dolby owns the AAC patents, not Apple.
That doesn't change the issue though, Dolby is no stranger to harassing open source developers: Dolby Standard tolerates no OpenSource implementation
Then ask Via Licensing:
What kind of weird world do you live in where a patented algoritm requiring royalties equals "an open standard"? These patents mean Apple is free to harass and sue anyone who implements and/or distributes AAC without paying the Apple tax. This completely excludes all free and open source software. Why do you think FAAC are afraid to distribute binaries?
http://www.rockbox.org/playerhistory/
It may surprise some people to see that the iPod was announced a full two years after the first harddisk-based mp3 player.
While we appreciate positive attention for our work, this story is a bit early. We have just begun to look at the iRiver iHP/H1 hardware and are quite a bit away from having anything of significance to show (such as running code).
I'll try to preemptively answer some common questions:
- No, we are not violating the DMCA or any other intellectual property laws. We are only distributing software written by ourselves and we run it on our own hardware. Our software does not circumvent any access control or copy control mechanisms.
- We are not doing this to "expand our market share" or any other weird corporate-style reason. We are doing it because our old Archos hardware is becoming obsolete and hard to find so we need to find new hardware to run our software on. The fact that the iRiver has a large user base is a bonus though, since it means more potential contributors.
- We are not looking at the iPod or Rio Karma since they contain a chip made by Portalplayer that you have to sign away your firstborn to see the docs for. That is a silly practice we do not wish to encourage. The iRiver contains hardware with published docs.
Feel free to drop in on irc if you have any questions.
/Björn
The original expression "I couldn't care less" is British and dates back to WWII. The distorted version "I could care less" started floating around the US in the 60s. It is very illogical and prone to misunderstandings and nobody really knows why or how it happened.
Suggested reading is Michael Quinion:
Does Starbucks have a license for this groundbreaking innovation from the patent owners at FPDC inc?
"System for reproducing information in material objects at a point of sale location"
And before any europeans in the audience start chuckling, it's patented in Europe too.
That is not true any more. It used to be like that, but the law changed July 1 2003.
Nonsense. USB Mass Storage is the SCSI layer and has nothing whatsoever to do with the file system. How do you think USB DVD burners could work if they had to use FAT? And do you really think all USB Harddisks have only FAT on them?
USB Mass Storage devices can have any file system, just like any other storage device. The fact that many (but far from all) manufacturers choose to use FAT is for obvious interoperability reasons, not because USB defines it so.
Use the --nogaps option in lame to create gapless mp3 tracks. Of course you also need a player that actually plays the frames back-to-back, and doesn't insert artificial pauses between tracks. I've heard conflicting reports about the Ipod in this regard.
My ISP, Bostream, offers this to customers already. Here's their service info page (in swedish)
In essense it says depending on distance to your switch, you get:
<300m: 26 Mbps full duplex
<1000m: 13 Mbps full duplex
>1000m: 8/1 Mbps (down/up)
Price: 399 SEK/month (~50 USD)
Another swedish ISP, Bredbandsbolaget, is also offering VDSL but currently "only" up to 10Mbit.
Then get the Recorder20, it has USB 2.0.
The Archos Jukebox 6000 is a $199 6gig MP3 player and USB harddisk that has an open source linux driver and
open source firmware.
Why are you slamming ATI for releasing binary-only drivers, while hailing Nvidia? Nvidia does exactly the same thing.
What do you think the 1MB 'Module-nvkernel' file in their NVIDIA_kernel-1.0-nnnn.tar.gz is?
NVIDIA_kernel-1.0-2960> file Module-nvkernel
Module-nvkernel: ELF 32-bit LSB relocatable, Intel 80386, version 1 (SYSV), not stripped
You didn't seriously think the few snippets of C code in that package was the complete driver, did you? That's just a kernel wrapper for their binary blob.
Don't confuse the terms.
Reverse engineering means examining a product to find out how it works. Disassembling the firmware is merely one tool used in that examination. Oscilloscopes and logic analyzers are other tools we have used.
We have written every single byte of the Rockbox firmware. But we could not have written the software without first researching how the hardware was put together, i.e. reverse engineer it.
All of this is completely legal.
If you define "from scratch" as writing software without first researching the surrounding interfaces, then no software has ever been written from scratch.
The difference is that the purpose of the CSS scrambling is to prevent you from reading the files on the DVD and save them on another disk (or play them in an "unauthorized" player). CSS is thereby a copy protection scheme, which DMCA forbids circumventing.
The purpose of Archos' scrambling is obviously not to prevent copying, since the files are very much copyable, and Archos themselves freely distribute the files on the Internet. Thus, descrambling it does not circumvent a copy protection scheme.
Why not help us?
Car CD changers are very simple to handle in software, but we naturally can't do the hardware ourselves since we don't have the head units.
SID support is perhaps possible. The main bottleneck is that we must encode the sound data as mp3 before feeding it to the sound chip. Unless, that is, someone manages to write a new DSP codec to play uncompressed PCM data.
We can play any format if only we can write code for the DSP to decode it. The MAS 3507 (and 3587) are generic DSPs that simply have MP3 codecs in ROM. We can download new codecs in them and I'm the first to hooray if we can get OGG or FLAC or anything into these DSPs.
Unfortunately, we have no docs or tools for writing new MAS DSP code and Intermetall is very secretive about it. If anyone can help, please get in touch!
The 12 MHz SH7034 CPU is unfortunately much too slow to handle decoding in software.