ately, having spent a good deal of time in the mobile web realm, I have increasingly become more of a performance freak. For mobile websites one of the biggest performance killers is doing HTTP requests. It takes a lot of time to setup / tear down the connection. The transfer rates tend to be rather terrible and they chew up battery life. I noticed that images were the biggest culprit of excessive HTTP traffic. Not only from the HTML itself, but CSS background images, which many web developer tend to forget about, can really slow things down. One easy solution around this is to Base64 encode the images and either deliver them in the style sheets or HTML itself.

So

o consider myself to be an equal opportunity coder. I like investigating different tools built in different languages to solve different types of problems. Recently at work I had the opportunity to play around with [elasticsearch](http://www.elasticsearch.org). Elasticsearch brings the whole distributed / non relational data craze to the world of search. It is actually a rather impressive piece of tech. If you are at all interested in search as a problem, I recommend you give it a look. In any event, I need to get some data into the search engine to play around with. Where could I get a good deal of data without doing the heavy lifting myself? Twitter! Sure the twitter activity stream.

eb fonts are pretty hot topic as of late. With the up rise of CSS3, people are trying to do more with pure HTML & CSS rather than relying on images to accomplish similar task. Logos, banners and things of the sort tend to be the most popular use cases, but it isn't uncommon to see custom fonts used throughout an entire site.

Although it is a more recent craze, support for web fonts has been around for a long time. You could render custom fonts in browsers as early as Netscape 4 Internet Explorer 4. As is usually the case with many things in the css realm, the problem with fonts has always been differences in browser implementation.