J. Carlos Navea – loconet

Day after day Google is releasing new projects that have this geek constantly interested. This latest one is called Google Code. Google Code is a website hosted by Google for discussion of Open Source projects related to Google and its services. This is a great win for both Google and the Open Source movement. Kudos to Google for being able to see the benefits in Open Source.

SpeakEasy is embracing Firefox by recommending their customers to use it. They have released an extension for their customers which basically adds a SpeakEasy homepage icon and a menu with interesting links. Nothing that hasn’t been done before but a very interesting move from this ISP. Hopefully their enlightened customers will spread the word to their families and friends.

Click to view full size:

I’ve been wanting to beef up the blog section of this site with a proper administration area for publishing entries, allow for comments, backtrack urls, rss/atom feeds, etc. I finally got around doing it by integrating an already existing package. The challange was to integrate the blog package with my existing code with minimum modifications to the package so that upgrades can be done without major problems later on. I played around with Movable Type but it didn’t quite allowed me to do what I wanted. After a bit of research, I noticed several bloggers are moving away from Movable Type to WordPress. WordPress is an open source, php based, MySQL back-end blogging system which is freely distributed under the GPL. The administrative tools are great with the exception of a few quirks like not being able to edit existing users and lack of page navigation on the posts listing. I was able to integrate WordPress’ front-end with almost no modification to the code itself other than some css and adding one or two php lines. The WordPress code itself is bit messy and not very well organized but it does the job for now.

I’ve added a new section which includes Firefox extensions I’m developing along with other Firefox/Mozilla.org related stuff.

Yesterday Google released a very nifty upgrade to their search tool, which allows you to choose from a list of popular hits depending on what you are typing on the search box.

After taking a quick look at the JavaScript code and sniffing my network to see what my browser was doing I was able to somewhat figure of the basics in the technique. Everytime you type something the key event is captured by the browser which in returns sends a query to a special url at google using a XMLHttpRequest object. This object essentially allows the browser to get data from a server in the background using JavaScript without leaving the current page. Each character you type in the search box, produces a query which then spits out the array of results used to populate a floating div used as the list of suggestions. The resulting code is really small (~20KB usually) and it’s compressed. At the present time this probably barely tickles google’s boxes but it would be interesting to see what will happen when this tool gets out of beta and more people start using it.

Anyways, this new toy was too cool to leave aside so I spent most of the night hacking together a proof of concept Firefox extension to add Google Suggest like behavior to the browser’s search tool. Just to show how great XUL/XBL/XPCOM is.

I’ve also created a small search plugin for Google Suggest so the extension only runs when this custom search plugin is in use to avoid any conflicts with the main Google search plugin.

New Google’s Suggest

Firefox GSuggest Extension

Obviously Google’s is much better hehe.. and there are still some issues with GSuggest’s results
list (ie: lack of keyboard navigation) but you can test it out by following the steps below.

NOTE: Use this extension at your own risk – It is only a proof of concept. The author takes no responsibility of direct or indirect damages caused by this program.

Known Issues:

• Mainly tested on win2k ff1.0!
• The result list lacks of keyboard navigation, you need to use your mouse to select result to query
• On Linux FF 1.0 the result list is not updated properly.
• Depending on how fast you type, there might be some lag in getting the result list to refresh properly
• Obviously if google decides to change delivery format, script location, etc. this will break..

I finally decided to get a proper domain for this website – loconet.ca
It’s also hosted on a much more flexible and faster box:

loconet@~/loconet.ca $> cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 2
model name      : Intel(R) Xeon(TM) CPU 2.80GHz
stepping        : 9
cpu MHz         : 2784.383
cache size      : 512 KB
...
processor       : 1
vendor_id       : GenuineIntel
cpu family      : 15
model           : 2
model name      : Intel(R) Xeon(TM) CPU 2.80GHz
stepping        : 9
cpu MHz         : 2784.383
cache size      : 512 KB

I’m going to start working on some side projects that I hope to put up very soon for download.

Many of you may know about the new attempt by Microsoft at a search engine (http://beta.search.news.com) to compete. Besides the fact that it’s obviously a ripoff of this other popular search engine with some eye candy added in, this thing is truly irritating me.

Note that 207.46.98.33 is registered to Microsoft so lets assume it’s the msn bot. Notice that the damn thing blindly keeps on downloading the same file! The requests are just a few minutes from each other. The of ther ip address is from google’s bot. This can be dangerous for those of us who have transfer caps if we don’t take the precautions of limiting the bot’s requests using robots.txt or through other more secure means.

Yesterday, on the anniversary of the falling of the Berlin Wall, Everyone’s favorite little browser that could finally reached 1.0.
Some have started calling November 09, 2004 – Web Independence day!

For those of you who don't know, Firefox is a free (as in beer and speech), compact, extensible, fast and relatively secure web browser developed by the Mozilla Foundation, a non-profit organization dedicated to the development of open source cross platform applications.

Firefox has various excellent end-user features such as tabbed browsing, built-in pop up blocker, download manager, live RSS bookmarks, among other nifty tools. Besides the great end user features, Firefox is built based on open standards, reason why as a developer, I love the program.

1.0 launch day was a very insightful day. I logged onto irc.mozilla.org to see what people were saying regarding the Firefox launch. Things felt like a party in the various Mozilla related channels, even my girlfriend joined us the geeks to discuss the launch. I got to see system administrators of the various mozilla.org servers scramble to solve the incredible surge in traffic on the servers. After all, they were getting 50,000 downloads per hour! I also got to hear live interviews with key developers for Firefox, including chief architect Brendan Eich, the creator of JavaScript. Shortly after the interviews were over, I witnessed the Mozilla community take interest in keeping this form of communication going and decided to maybe start an online radio station for Mozilla related content. Within minutes, an IRC channel was created to plan the project, technical requirement decisions were being made, people were being recruited, bandwidth and hardware was being donated. It was an interesting experience to witness the possible start of Mozilla Radio .

As I was reading /. today, someone pointed out quite a curious bug submitted to bugzilla regarding Mozilla not adding time to the day, something which I totally agree with. Along with this bug, it is mentioned the request for the so useful feature of Mozilla displaying pages before they are downloaded; now _that_ would be innovative.

It’s great to see developers and testers of one of my favorite applications have a good sense of humour.

I’m back in Canada and working. In a way I’m glad I’m back because I’m completely broke, and in another way I’m depressed because my trip is over =(.

I’ve put up some photos of the Europe trip, including a little map with a rough idea of my trajectory through the 6 countries I visited.

I’ve made some changes to the photo gallery to support albums. I’ve also upgraded the navigation and display of images. You need to make sure you use the latest version of a w3c compliant browser. I’ve tested these pages using Firefox 0.9, Opera 7.53. IE 6.0 renders well for the most part. Note that in some versions of Firefox earlier than 0.9, the css I use on the gallery was breaking the loading of the larger sized images, this seems to be fixed in the current version of Firefox.