Endarkened from Kaliptus on Vimeo.
Warning: very graphic cartoon content.
Thanks, LSDEX!
Archive for February, 2010
Why the internet was invented
Sunday, February 28th, 2010Pic of the moment
Saturday, February 27th, 2010
Thank you, ImgBit.com (and Stumbleupon)
Another amazing talk by Bruce Sterling
Friday, February 26th, 2010Ignore the damn title; it’s interesting anyway.
One of the most amazing things about Bruce Sterling is how he can be such a great author and insightful speaker and sound like such a smarmy jackass. I don’t think it’s really a reflection of his personality; I think it’s just his voice. Maybe it’s even just me. But his voice surely does grate on my nerves.
Here’s a totally different talk by him to demonstrate what I mean about his voice:
Bruce Sterling from Innovationsforum on Vimeo.
(Bruce, if you read this, please don’t take offense! I love all of your books, and you are brilliant. My issue is my issue and is totally superficial. And I’m pretty sure in a personal conversation awe would overcome any ookie voice-based feelings.)
A Thing Someone Needs To Build
Friday, February 26th, 2010I have this vision of A Thing. Right now it’s a cloudy vision, but I’d like to ask your help clearing it up.
The time seems particularly ripe for a Thing that helps People start Projects to make useful Stuff. For example, people could collaborate to make a story. Or get an idea patented. Or start a distributed business making handicrafts, and connect to customers. Of course, I say the time is ripe because of current employment market, or lack thereof. The unemployed could be spending some of their time creating valuable stuff or connecting valuable stuff to people who need it – hence becoming employed.
The qualities this Thing would need (IMO) are:
- easy and interesting to get started using
- gives a comfy feeling that if the Stuff you’re working on generates revenue, you’ll get some kind of reasonable cut
- helps People find Stuff that needs their talents
- helps Projects find the talents (or other resources) they need to generate revenue
- possibly include a system of currency so People with Stuff can trade it with other People for Stuff they need
- helps People leverage other resources for their Project (e.g. instant messaging, wikis, etc.)
I kinda sorta started a wiki some time ago about this sort of Thing, but it never went anywhere.
Whaddaya think? If someone built it, would they come? Or is that impossible to answer without knowing a lot better what it is?
Oh. My. Gods.
Monday, February 22nd, 2010
Jones is making a limited run of D&D themed sodas. Made with pure cane sugar.
I’m so gonna get me some of that!
BTW, next week (Feb 28 – March 6) is Read an RPG Book in Public week. Be there and be square!
Random pic of the moment
Sunday, February 21st, 2010
Thanks to Anorak News (and Stumbleupon).
Website optimization gold mine
Friday, February 19th, 2010Best Practices for Speeding Up Your Web Site
A bunch of things I didn’t know, or didn’t know the importance of. It looks as if they’re listed in order from the ones with the biggest payoff to the least.
If you host your own website, read it!
Major employment improvement under Obama
Friday, February 19th, 2010
Someone show me the counter-argument to this compelling graph, please.
New compression tool (and so much more!)
Friday, February 19th, 2010Another nerd alert. This is really the context for the last nerd alert.
I have the first (buggy; see last post) version of my rabin chunking tool done. The main things that it’s good for are:
- compression
- identifying the parts of a file (text or binary) that have changed
- minimizing the data needed to store multiple versions of a file
- lots of other things. For example:
- you can identify which binaries are infected with a virus by knowing the hash for some of the chunks in the middle of the virus and checking the binary to see if it also contains those chunks
- you can automatically detect viruses by scanning your outgoing traffic for the same chunks going out to many different machines
As an example, I took a tar of the first 100mb of /usr/bin on my Debian box. Gzipped, it’s about 31 meg. If I run it through my tool first, it gzips down to under 25 meg – slightly better than 20% smaller than without using my tool.
The tool runs in several modes:
It can do simple compression of a file. Longish sequences of data that are repeated in the file are replaced by references to the first time that data occurs, no matter how far away the repetition is from the first occurrence.
It can decompress what it compresses. Duh.
It can split a file out into chunks and throw all the chunks into a directory as separate files. Chunks are identified by their Rabin hash (currently; I may switch to another hashing algorithm). If a chunk occurs multiple times in the file, it only appears once in the directory.
It can list all of the chunks, chunk lengths, and chunk ids in the file. It can do this in combination with creating the chunk files.
I have a script that can reassemble chunks into the original file based on the chunks + the list of chunks; I will put this functionality directly into the tool.
I have started a github project for rabin-tool. You can download, make and play with it from there. Run it with no arguments to see all the possible arguments.