Compression Formats

Why bothering supporting all formats, when we can use single turing-complete data format and eval would be a decompressor? Then anyone could use it's own private compression algorithm.

I.e. use Lisp's lambda as compression format, then Lisp's eval would be a decoder. You can store any imaginable data inside lambdas (http://en.wikipedia.org/wiki/Church_encoding)

Moreover, let all files be turing complete, so a gigabytes sized file full Pi digits could be represented as a few kilobyte *.lisp file.

The file system is an artifact of the limitations of early processors that had no memory virtualization capabilities and had an insufficient memory address word size. Programs and data had to be marshalled out of and back into a logically distinct store because 64K can only hold so much information. Modern 32 and 64 bit processors with memory management units can treat storage as large, slow, persistent memory obviating the contortions required in conventional operating systems to keep data from going away when the computer is shut down.

A heirarchical name space is still required for users to keep track of where they put things but virtual memory can hold that tree structure just as well as inodes can and the judicious use of separate memory spaces can side-step a 4 gigabyte limitation for storage on 32-bit machines.

>>2
But every nontrivial file would just have overhead instead, and it sounds like you're suggesting that the only way to read a file would be to execute it, which would be silly.

Otherwise, why not just do that, and pass around py.lisp instead of my_fucking_pi_digits.txt? _{(The answer is speed, by the way)} It's not like we can't execute arbitrary files on the filesystem whenever we want, or like we can't bundle README files with datasets saying ``this isn't actually the data you want, but at least it's only a few kilobytes!''

That's already kinda exist.
Rar has its own virtual machine.

>>5
which is why you can write quines!

>>4
you're suggesting that the only way to read a file would be to execute it
you are already "executing" files by the miracle of read() syscall

like we can't bundle README files with datasets
README should be autogenerated, based on the options program supports and it's usage.

>>7
you are already "executing" files by the miracle of read() syscall
For a bullshit definition of ``executing'' that doesn't actually mean that.

"i like cocks" - nerd op

"I like nerds" -- cock >>9

>>20
listen here you

>>8
read() consists of executable, which must be executed in order to obtain content of the file.

>>4
what about /dev/random?

We use LZMA2 at work. But yeah, IHBT

>>3
No sync(), no use.

>>15
turing complete file can implement any imaginable sync

I am the most touring complete.