Tonight when I first read a street sign, I thought it said Lasagna Drive. Why aren't streets named after food? I'd rather live on Banana Ave or Syrup Blvd, than Spring Meadow Creek or some quaint crap like that.

At work (the IS dept at a Christian university) a few weeks ago, our lunch discussion came to whether God could both love and hate someone at the same time. I said that if the person is saved by Christ, God sees Christ -- not the person -- when he looks at him. Drew said "it's like a hard link!"

On the web, it'd be like a permanent redirect. But only when God tries to resolve a Christian's URL does he get redirected to the URL representing Christ. "The LORD does not look at the things man looks at. Man looks at the outward appearance, but the LORD looks at the heart" (1 Sam 16:7). So how would that work? The difference in how the person's URL resolves depends on the viewer, so something has to be done at God's end. We came up with the idea that God must use Greasemonkey with script that applies URL redirections.

I couldn't find any Greasemonkey scripts that did URL redirection (just lame javascript redirects which is not good enough). So here's how I think Christianity could be modeled in HTTP:

Let's say http://brondsema.net/dave represents me (technically, it is an RDF document about me, but I'll avoid getting into RDF). http://brondsema.net/Christianity/Christ will be the URL representing Christ. When God requests http://brondsema.net/dave he needs to get:
HTTP/1.1 301 Moved Permanently
Location: http://brondsema.net/Christianity/Christ
This is assuming that a person once saved can never fall from salvation. If it's the contrary, then instead of a 301 Moved Permanently, there'd be a 307 Temporary Redirect

Probably be the best way for God to achieve this would be with a local proxy that rewrites some responses to be 301/307 instead of their original response value. Of course, only God would know how his proxy works; none of us know if other people are saved by Christ or not.

Some definitions, for the less technically inclined:

hard link

like a file shortcut, but better

URL

Universal Resource Locator; aka web address. It can represent anything, not just provide a webpage

Greasemonkey

A firefox extension that lets users install scripts that change the functionality of certain webpages

HTTP

HyperText Transfer Protocol. The underlying protocol that web browser & servers use to request & transfer webpages

Google indexes about 10 billion webpages. Maybe. It's hard to find a number. But it doesn't matter, because whether they index 10 billion or 2 million pages, I can't comprehend that number. I can read the number, but I don't really understand what it means.

But I'm starting to. I've been researching distributed trust networks on and off for a few years. Lately I've been doing more research in the area and finding tons of stuff that I hadn't seen before. Just by following links around and searching with slightly different phrases is revealing more and more stuff. My reading backlog is growing far too quickly

The most exciting things I've found are a mailing list for applied computational trust, and the T3 group which has lots of papers and event information available. I've been compiling most of my findings that are relevant to Konfidi (my distributed trust network project) at the Konfidi Research page.

Another way to try to comprehend it: if you looked at one page per second nonstop, it would take you 317 years to view 10 billion pages.

Non-nerdy: version control systems are the tools that software developers use to share and track the source code used to build the software; open source licenses permit people to use others' source code in various ways; open development is a community concept about how the software project is managed

John Reynolds writes about Open software pragmatism - Free (as in beer) isn't the point which is along the lines of "Open Development" that has been talked about before. He should've shouted "free (as in speech) isn't the point, either" but whatever.

The practicality is that you can't force open development communities, even a community can't always force itself to do so when it wants to. So what can be done to help? For the matter of maintaining and merging your own set of patches and enhancements all the time, distributed version control seems like a good solution. PlanetApache readers know that Sam Ruby and Elias Torrez have been hacking on PlanetPlanet aka Venus, using bzr. But most projects use SVN or CVS which are centralized version control systems. Has anyone tried svk to maintain a personal SVN repo in sync with the central repo? I really should try that myself. Why don't major projects, like the Apache ones, use a distributed version control system? The only major project I know of using distributed version control is the Linux kernel, which uses git.

So Microsoft Chairman Bill Gates is stepping down over two years, to spend more time at the Bill & Melinda Gates Foundation, working on global health, etc. Why, you might ask? Because Microsoft has reached the limits of humanity. Gates needs to focus on keeping more people alive longer, so he can sell them his software.