h a l f b a k e r yLike you could do any better.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
The traditional aluminum CD is thought to be perhaps the best example of a "write once" medium - to my best knowledge, no one has ever written on such a disk after it came out of the stamping press. However, the aluminum layer is very thin, and there is no reason why a sufficiently powerful laser could
not write on the disk. It may even be that existing CD-R drives could do it, at their sustained maximum power level. However, the hardware aspect of this idea is not the most interesting one. Given the desire, one *could* design a CD-R drive which could pit an aluminum disk. The most interesting problem is a software riddle - how does one write to a disk with a relatively chaotic mixture of ones and zeros already present? I propose two solutions. Both involve creating new formats of compact disk data storage. The first, simplest one is to utilize the unused space present on most disks (AOL CDs come to mind.) As a marker for the machine to recognize the file system, record a large block of consecutive one's at the beginning of the disk. (remember, in this system we can make a Zero into a One, but not vice versa.) Then, a second format is utilized - the chaotic data found on the disk is written over in such a way that each logical "one" is converted into a huge sequence of consecutive physical "ones", and the same for 0's. The small block of data written in this way serves as a pointer to the beginning of the free space left over from the unwanted conventional recording. The *second* method, one which is appropriate for disks with no unused space, is to use the consecutive block format for the entire disk. This would result in a capacity perhaps one tenth of what is normal; however, it is arguably better to store 65 meg. on an AOL CD-R than to microwave it, etc. *** For those who are confused, the best analogy for understanding the software aspect of this idea is the fact that one can indeed write legibly on a sheet of paper covered 50% with black dots, if one uses a very thick marker.
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Destination URL.
E.g., https://www.coffee.com/
Description (displayed with the short name and URL.)
|
|
Why not resurface the bottom? Just slap on two plastic stickers - one opaque, to act as a new write surface, the other a transparent protective layer. Might work. |
|
|
_Could_ existing CD-R lasers really pit a CD disk? |
|
|
Anyway, I heard this mentioned somewhere else -- might even have been here -- so I can't claim it as mine, but: shouldn't AOL et al be required to send out only CD-RW discs? |
|
|
That way one could at least re-use the disc rather than tossing processed polycarbonate straight into landfill. |
|
|
Think of it as a return to the days when they sent out floppies... |
|
|
JKew: Yeah, that would be good, a law requiring anyone sending unsolicited CDs to either pay for recycling or offer reusable media. |
|
|
I just realized that the "monstrous blocks of consecutive digits" algorithm could be used to re-use *non-RW* CD-Rs, without any changes to existing hardware. |
|
|
You could use the unused free space to hold "error corrections" (eccs) to the data out in the already-written portion. You would then modify the already-written portion to have its bits as close as possible to what you really want to store. The ecc would have to be variable length since the worst case is that you want all 0's and the already-written data contains all 1's - that would require a lot of ecc since it's so much different. |
|
|
Maybe this wouldn't have to be done on a per-byte or per-word level - maybe you could ecc the entire written portion all at once after modifiying it first to be as close to possible to what you want. |
|
|
"Monstrous" blocks of consecutive digits aren't necessary. For example, represent a logical '0' by an odd-size run of physical '1' bits, and a logical '1' by an even-size run of physical '1' bits, separated in both cases by one or two physical '0' bits from the next logical bit. You could get even more sophisticated than that; it's an interesting problem in information theory. You ought to be able to do much better than 1/10. |
|
|
Unfortunately, the whole project will be complicated by the many layers of error correction present in compact disc formats. You'll have to reproduce that error correction, and do so without the luxury of knowing how many physical bits a given logical bit will take. |
|
|
That's non-trivial, and even worse, that error correction is usually done in the hardware of the CD-ROM drive, which means that your rewritten CDs wouldn't be legible on standard CD-ROM drives (even with special software). |
|
|
Excellent idea. Bun for you. |
|
| |