Well, most of the work is done by Earthdesk and GeekTool 3.

Earthdesk is set to Natural Color, Equirectangular projection, Natural Color, Real Moonlight, centered on Vienna, Background: Starfield. Zoom 80%, Clouds 80%, Brightness 80%.

In GeekTool, the times and the weathers are all separate Shell "geeklets".

Times are generated by running shell commands like

env TZ=Asia/Tokyo date " %l:%M %p"

every 20 seconds

The weather is the tricky part. The way I am doing it now, if I am not careful, gets me throttled for too many concurrent requests to the wunderground.com API server. It also fails badly if I am disconnected, so I will need to do it differently.

FWIW: I have a PHP script which I run as separate Shell Geeklets. I invoke it with the name of the city I want. It then hits wunderground and gets back an XML stream of the local weather, which I parse, format and echo. (the way I'd change this is run the script from cron, with a 30 second wait between requests, and cache the results locally, which I would then call from the Shell Geeklets)

From there it's just a question of setting fonts, sizes, colors and moving the little Geeklet boxes around as you want them.

Given this new type of information and its value to search, we are very excited to announce that we have reached an agreement with Twitter to include their updates in our search results. We believe that our search results and user experience will greatly benefit from the inclusion of this up-to-the-minute data, and we look forward to having a product that showcases how tweets can make search better in the coming months

When searching for something on the web we’ve all had the experience of clicking on a link in a search engine’s results page only to find that the page no longer exists. If there’s no information on that page other than a default error message, the most likely course of action on the user’s part is to press the back button and try the next search result.

As site authors we can make our error pages more meaningful to our users, so that an error becomes an opportunity to bring the user back into a site and show them content that’s relevant to what they’re looking for. In this article I’ll show you how to do just that.

answer Right. And it results in a type of error that doesn't teach you anything. You have chaotic errors where all you can say is, "Boy, this was really screwed up, and I guess I need to go in and go through the whole thing and fix it." You don't have errors that are proportionate to the source of the error. And that means you can never have any sense of gradual evolution or approximate systems. So, the real difference between the current idea of software, which is protocol adherence, and the idea I'm discussing, pattern recognition, has to do with the kinds of errors we're creating. We need a system in which errors are more often proportional to the source of the error.

We present a system that composes a realistic picture from a simple freehand sketch annotated with text labels. The composed picture is generated by seamlessly stitching several photographs in agreement with the sketch and text labels; these are found by searching the Internet. Although online image search generates many inappropriate results, our system is able to automatically select suitable photographs to generate a high quality composition, using a filtering scheme to exclude undesirable images. We also provide a novel image blending algorithm to allow seamless image composition. Each blending result is given a numeric score, allowing us to find an optimal combination of discovered images. Experimental results show the method is very successful; we also evaluate our system using the results from two user studies.

One of the new things in Solr 1.4 that I am particularly excited about is the new document and search results clustering capabilities. This is an optional module that lives in Solr’s contrib/clustering directory and was added via SOLR-769.

We generate these deep links completely algorithmically, based on page structure, so they could be displayed for any site (and of course money isn't involved in any way, so you can't pay to get these links). There are a few things you can do to increase the chances that they might appear on your pages. First, ensure that long, multi-topic pages on your site are well-structured and broken into distinct logical sections. Second, ensure that each section has an associated anchor with a descriptive name (i.e., not just "Section 2.1"), and that your page includes a "table of contents" which links to the individual anchors. The new in-snippet links only appear for relevant queries, so you won't see it on the results all the time — only when we think that a link to a section would be highly useful for a particular query.

GeographicLib is a small set of C classes for performing conversions between geographic, UTM, UPS, MGRS, geocentric, and local cartesian coordinates and for solving geodesic problems. The emphasis is on returning accurate results with errors close to round-off (about 5–15 nm). In addition, various properties of the Transverse Mercator Projection are described and an accurate algorithm for Geodesics on the Ellipsoid is given.

This project is about running test cases automatically on several versions of the main Web browsers available today. See the StartPage for more information. Tests and results are visible on http://www.browsertests.org.

One of the 8 works I created for the Data Art Show at the Pink Hobo Gallery in Minneapolis.

All these pieces are a pun on the new craze for data visualization. The goals of data visualization as I understand them are to make complicated issues more understandable, to make obscured connections visible and to reveal hidden patterns in the data. After all these tasks have been solved ideally the result should be aesthetically pleasing as well.

But when I look around what is being done in data visualization today I have the suspicion that in many cases the design is more important than the actual information and that the use of data is more an excuse to justify the use of aesthetics.

Since I do not have a problem with aesthetics for their own sake in these pieces I deliberately took the opposite direction. Since I wanted to create something visually interesting I made up my own data which would give me the desired results. All these works are the result of generative algorithms, so all the elements and their connections are actually data and not something I assembled manually in Illustrator.