Labnotes

This thread starts innocent enough, typical loon invents the perpetual machine equivalent of a compression algorithm and solves the unsolvable:

The FAQ has been pointed out to me where it “proves” that not all files can be compressed by any algorithm, but it seems to me that it has some mistakes. They aren’t serious, but they are oversights that prevent it from applying to fractal bits, permutations and sorting the bytes of the original file.

Said loon immediately encounters resistance from a likely source, your typical “between kill filing you and being nasty, I choose door no. 2“:

> Good day everyone,
Well it was a good day…

The discussion quickly evolves into theories of compressions, fractals, perfect numbers and a bunch of other complex statements, none of which I can validate for myself, and arguments regarding sexual practices, none of which I can validate either.

So far, a typical day on Usenet.

But you want to keep reading, because half way through the thread suddenly takes a turn for the better — or worse — when it turns out our new compression wizard is motivated by more than the basic desire to create a better compression algorithm.

Can’t say, though, whether it’s winning the Darwin awards, or pulling out a brilliant prank:

Well, Jim, I don’t know “most people” are about, but in this case I really would like the decompressor. What motivated all this was a lack of space on my Hard Drive where none of the off-the-shelf compressors could save enough space. The fractal compressor I wrote was perfect (and I could prove its correctness too). I was able to squeeze down several Gigabytes of data to fix my space issues for now.
The files in question are a lot of personal correspondance, research, and coursework archives. It’s nothing terribly critical, but still it’s stuff I’ve saved for years and which I’d really like to get back : (. The decompressor must be out there, it *has* to be. I don’t know what these other people you’re talking about did, but no one will try harder than I will to write it.
I’m very optimistic (I still have all the compressed files sitting right here, the data is in them in some form), but I’ll keep you guys posted no matter what. You deserve no less, I suppose.

In other words:

You deleted the originals without testing a working decompresser first?

Hilarious.

The pessimist in me is appreciating the fact that none of that research material will see the light of day. The optimistic in me is wondering what brilliance is required to pull the collective leg of an entire Usenet group.

Of course I keep reading further, mostly for the train wreck effect, but seriously hoping to find it’s a prank after all, not expecting it to turn into a brilliant piece of performance art about the state of scientific discourse in the US:

Thanks everyone, for your polite comments. I think the takeaway from this for all of us is that voices have been heard on both sides in this thread illustrating the healthy debate among experts on whether random data can be compressed.

Brilliant.

Where's my comment? I get too much comment spam, so I have to moderate comments. Damn those spammers. If you don't see your comment immediately, be patient. I'll approve it the minute I see it. Want to know when your comment shows up, or check if anyone responded? Track it.

Aug 27th, 2007

Gabe

Honestly I don’t see how anyone could take this seriously the moment he said he compressed his own files (and probably long before that). It’s just too hilariously juicy. I read all 50 messages in the thread and no one called him out for trolling (that I noticed).

As far as I’m concerned this guys are just a little too gullible. I guess it’s possible that someone would honestly be this batty, but it just doesn’t seem likely. It’s one thing for an ignorant individual to spew a bunch of random mathematical terms in an attempt to seem smart, but this seems to be going much farther to bait people.

I think the comp.compression people should be embarassed that they fell for this one. The only thing really brilliant about was picking his audience–a group of people so obsessed with math that they become enraged at the first hint of a crackpot. After that it seems to be like shooting fishing in a barrel.

Assaf

All you need to pull such a prank is a Usenet group with people who take themselves too seriously. Like, something starting with comp, that would be a good start, better if it involves some mathematical intensive computation, like maybe compression?

Aug 28th, 2007

Crosbie Fitch

The human brain is one of the best compressors and decompressors we know about.

We don’t yet have the source code to it, but it’s on its way.

The RAM required to hold the program and data is still going to dwarf most of the text files that can be losslessly compressed.

Unfortunately, not only is the compressed data non-portable, but decompression is necessarily optimised for internal processing rather than external streaming.

Aug 29th, 2007

I also noticed that legacy data tends to suffer from lossy compression and long retrieval times.

Performance art on comp.compression

Leave a Reply | Trackback | Track with co.mments