I realized that when I need to use new Open Source project, I add it to Ohloh (if it’s not there yet) to see what they have to say about it.

The only problem is that I’m too lazy to do that every time I need to use some software. I wonder why Ohloh doesn’t just use Google to find all SVN repos to crawl?

It’s first time for me to see somebody else’s project on the domain I registered.

Here’s my feed if you’re interested: http://friendfeed.com/sergeychernyshev

P.S. almost like http://sergeychernyshev.feedfriends.com/ ;)

I spent some time this weekend writing a new kind of application - Facebook application.

It’s quite interesting how they built their platform and they are probably first to do this kind of thing, at least among big players.

Anyway, come and take a looks at Friends Explorer application. Currently it shows birthday time-lines for your friends, but more features will follow (only if you’ll add it ;)).

It looks like Facebook is not the only one to create web application platforms.

I wrote a simple extension to LWP::ConnCache, which allows to somehow resolve keys before putting connections into cache.

Task is not very obvious until you understand that http://www.example.com/ and http://example.com/ are the same server and there is no need to open two connections if you want to get data from both sites.

Moreover http://*.livejournal.com and http://*.blogger.com are the same server.

how you do resolution - using DNS or using some other algorithm is left to developer.

Anyway - her it is: LWP::ConnCache::Resolving.

I’m back from Web 2.0 Expo. I met many interesting people and watched many interesting presentations.

I also started to think about converting TechPresentations.com from blog to Wiki. In any case, I’ll post update it in a coming week to reflect the knowledge I got at the conference.

I’ve been looking at JavaScript recently, especially after watching those courses by Douglas Crockford.

I was doing some bookmark and widget coding I realized that there is no tool to load JSON data from remote hosts. While plenty of libraries provide you with various tools to organize your XMLHttpRequests and other hacks to load data asynchronously from your site, none of them allow you to overcome same-host security limitations and load data from other sites without using some sort of server side proxy.

So to streamline the rest of the development I wrote a library that can be used to load JSON-formatted data from remote hosts without spending your own traffic.

See project page here: JSON Remote Loader

P.S. as part of this project, I got some experience using JSLint, Firebug, Google code project hosting and even wrote Makefile to automate minification and colorizing as well as zipping of the code.

I haven’t been publishing about many changes in Feed Friends and I can’t really remember everything I did but today I finally got tired of link toolbar to only have Feed Friends share button and not to have Digg, del.icio.us and the rest so I changed this processor to include all In-Feed Controls I configured in processing customization page.

So, even if you don’t use Feed Friends sharing functionality, you can still enjoy Feed Preprocessor part of the system and use it to get If-Feed controls to work for you.

P.S. Actually Feed Preprocessor configurator allows you to pick more controls then listed on infeedcontrols.com and they include: del.icio.us, Technorati search, Digg, Feed Friends, Newsvine, Fark, Reddit, Yahoo! My Web and Netscape.

Bre Pettis posted my playing cards top-hat project on MAKE blog.

Alex Tew created his new project - Pixelotto.

I’ve tweaked original MDHP scripts to produce Pixelotto versions of the graphs as well:

Blogging activity graph:

Traffic graph (by Alexa.com):

Sales progress graph:

Animated sales history:

I hope this time I’ll be given some credit over these graphs if they’ll be used on Alex’s site.