Thomas Phelps and Robert Wilensky, for their project, Robust Hyperlinks. Traditional hyperlinks are very brittle, in that they are useless if the page later moves to a different URL. This project improves upon traditional hyperlinks by creating a signature of the target page, selecting a set of very rare words that uniquely identify the page, and relying on a search engine query for those rare words to find the page in the future.

Now, using Google search to do this, they give an example link, which is here and is supposed to find Google's page linked above about this programming contest. And it does....as the second link.

But thinking further, wouldn't it be nicer if instead of 404 pages, we webmasters all subscribed to a service that allowed people to include a code in their 'a href' tags for the page? Maybe an md5sum of some unchanging feature of the page itself, so that visitors could ultimately get to a page where that md5sum was searched for in an index and gave some really automated guesses as to where the page went? And if there were none, maybe linked over the Archive.org in hopes that we could find it stored away?

I haven't bought a pre-installed operating system for a very long time, but I'm more comfortable putting together the components myself than I am in allowing a shop to do it for me. Getting a major manufacturer to ship Linux-installed or no-OS-installed PCs is a huge step to allowing individuals and businesses to control the costs of their computers as well as what operating system they choose to use.

One question though...how far are we from Dell PCs installed with an Apple OS?

In short, after seeing all the flaws in this, I went another direction and used the PHP package archive and remote installation software 'pear'....the "PHP Extension and Application Repository". Basically, with that all I had to do was go through a dependency ladder (upgraded a bunch of dependencies, upgraded pear manually, upgraded pear further automatically, then was able to install a SOAP implementation from their 'beta' repository.

The PEAR commands used, in no particular order for the specific installs and upgrades I was doing:

• pear search [package]


• pear install [package]


• pear install [package]-beta


• pear upgrade [package].tgz # From a tarball package I downloaded for the manual PEAR upgrade


• pear upgrade [package]


• pear upgrade [url-to-package-tarball]

I've been there, though I admit that at the time I didn't follow all the steps this guy did. When the server under my control was hacked, it was blatant and obvious, with a bizarre overflow string in the logs and some very funny behaviour of the machine. I needed that machine up and running, and didn't have the luxury of downtime to investigate, so it was wiped and reinstalled immediately.

So I started hunting some solutions. iMovie on my Mac Mini doesn't do conversions like I'd want, so I hunted further and found ffmpegX. It's based on the very mature ffmpeg project, but definitely separated. It allows drop & click operation for conversion and so far has done very well. Conversion was simple and fast...even on the older, low-powered, G4-CPU Mac Mini I have.

For more information: Flash Movie Conversion for Mac OS X

From the man page:

NAME
   rpl - replace strings in files

SYNOPSIS
   rpl [-LhiwbqvsRepfdt] [-xSUFFIX]

DESCRIPTION
   Basic usage is to specify two strings and one or more filenames or directories on the command line. The first string is the string to replace, and the second string is the replacement string.

It doesn't do regexp matches, but if you need something that serious, you're better off with perl or sed anyway. What I found really nice is the dry-run mode (-s), which would have prevented me making serious mistakes in the past.

Of course, for a single file edit with lots of searches and replacements, I'm happy to use vi/vim. But if I want to do a bunch of changes, especially recursing through directories, rpl is the way to go. If you use a Debian-based system, it's likely in apt.