Bleezer: An offline blogging tool is born

Blogging software has been around for years. There are a number of platforms to choose from, most of them with the standard features that are required to get a blog up and running.

So why did I decide to design, develop, and launch my own blogging tool, Bleezer?

Because I could!

I'm sure those words have been the genesis of many software projects, but there were actually other motivations. I saw needs in the marketplace that weren't being satisfied by existing tools, and thought that I could bring my own development experience to bear, and create something better.

Let's back up a few years to late October 2005, to the point where the idea for Bleezer was born. At the time, I wasn't particularly happy with the available tools for offline blogging. They didn't have the features I wanted such as the ability to cross-post to multiple blogs or the ability to create multiple posts simultaneously. Also, I use Windows, Mac, and Linux, and there were no multi-platform tools, and certainly nothing that ran on Linux.

So I decided to write my own. I am an experienced programmer, and creating an offline blogging tool was within my abilities.

My first decision was the choice of language to write in. My multi-platform requirement limited my choices, especially since I wanted the user interface to be consistent across the platforms. For ease of distribution I prefer a compiled language so that users don't have to install anything else. To meet these goals I chose Java, and Swing for the interface, assuming that at this point Java virtual machines were available everywhere.

I didn't have the slightest clue how to start posting blog entries, so I spent a couple of days researching how blog APIs worked. I found that they rarely work as documented. XML-RPC seemed to be a common protocol, though each API offered a different set of functionality. Fortunately The Apache Software Foundation provided an XML-RPC toolkit.

I would come to know Apache very well before I was done -- many of the components I required were available there, especially when it came to parsing the XML that was returned by the blog engines.

My first test was to be able to either read from, or post an entry to my own blog, which publishes to the Internet using Movable Type. It took me an evening to manage to post a test entry to the blog. Compatibility with other blog engines would come later, too.

And I was off. I spent a couple of hours each morning at a nearby coffee shop fleshing out the user interface to do a bit more each day. I added features such as the ability to get and edit previous posts, to preview the article before posting, and a toolbar for quick access to functions. About two or three weeks in I decided to switch to Bleezer as my primary blogging engine -- I figured that eating my own dog food would force me to add the features I needed to be productive. I also enlisted three friends I'd met through blogging to test it out as well.

But it still needed a name. My wife and I poured through dictionaries in different languages until we stumbled on the name Bleezer, and since the domain name was available, that's what it became.

At this point, the tool only allowed users to create HTML posts. There was no "what you see is what you get " interface (WYSIWYG) -- I just wasn't sure how to do that. Early feedback from users suggested that a WYSIWYG editor and spell-checking would be very valuable. So I set out to add those features. I found an open source spell checker called Jazzy and implemented it. WYSIWSG turned out to be a bit more complicated until I found an open source WYSIWYG editor called Kafenio. SourceForge turned out to be another very useful repository of open-source components.

But the open source integration wasn't without its drawbacks. Something I learned very quickly is that (a) open source documentation is usually poor to non-existent, and (b) some open source software doesn't work as advertised and you end up spending time fixing it. Of the two to three months I spent writing the code, about 33% was spent trying to figure out incorrect APIs or fixing open source code.

I packaged up Bleezer appropriately for Windows, Mac, and Linux, and released it to the world in January 2006. I was immediately noticed by folks like Om Malik, and even got a half-page article in the business section of my local paper. I probably couldn't have achieved this without the reach of the Internet.

People have been generally complimentary about my little software project, and full of helpful suggestions. I have learned a few other things, too.

For example, people are not generally fond of the Java look and feel, especially Mac owners. They point out that Java does not adhere to Apple user interface guidelines. Linux users are the most complimentary -- they often appreciate the fact that they have an editor to use. While the Linux version seems to be fine, it might be better to write in the dominant language for each platform (i.e. Objective C for the Mac), though this would substantially increase development time and effort.