Joel Hainley

Mike challenged me to a contest, hopefully this one goes better than the last one.

The challenge : write a web-based application and release it to the world on or before November 5, whoever generates the most revenue between November 5 and January 5 wins. The only other requirement is a weekly development blog posting detailing status and the work on the agenda for the coming week.

The stakes : I think it’s either hot wings at Hooters or Wing Stop and maybe a case of Mountain Dew or Jolt…can you even buy Jolt anymore?

My project : ClikClock a time tracking program. I really want to do a good, solid, well thought out time tracking program because the few I’ve used drive me nuts. Everyone I talk to hates the applications that they are forced to use at work. So I’m going to give it a shot. See if I can make something that’s powerful and easy to use.

Update : I have had several false starts on ClikCLock in the past but I want to get this idea out of my system, because I’ve had thoughts about doing it for so long. I have written a couple of partial implementations and was reasonably happy with how things were turning out. However some business reasons ultimately forced a rewrite on a different technology platform. On May 1 I opened up an editor and started the current implementation of ClikClock. Since then I’ve written my own MVC framework and have integrated Doctrine and Smarty to create a pretty nice framework to do development in. I had to build a lot of stuff in the beginning to get things working that I wouldn’t have had to build if I had used one of the existing frameworks but I had a desire to build the framework and quite frankly it only took between 4-6 hours to build out the MVC framework and It’s WICKED fast.

I have also decided to use git for this project so I’ve been learning that as I go along. So far I like it but I haven’t had to do anything too crazy with it yet.

Tasks for the week :

• New Account Setup wizard
• Account Administration screens
• Build out the Account Configuration system

Work on Clik Clock is continuing at a rapid pace, there is still a lot of work to do but this weekend saw a major milestone reached. I am now able to enter/edit data using the UI that I have been designing for the Time Entry module. The other modules are all still being edited with the development UI but it’s getting closer. There’s something magical when the code you’ve been working on suddenly starts doing some small piece of what you’ve been envisioning. When it does enough that it can now support you in your efforts and you are able to “eat your own dogfood” it’s like it has taken it’s first breath and is now starting to take on a life of it’s own. It’s really a special time, at least for me.

Paul Graham ( the y combinator guy ) stated in the book Founders At Work that he wished they had setup an online store and sold SOMETHING so they would have had better insight into their customer experience. I can definitely see the value in this as I am more likely to overlook the bumps and bruises of something when I have my engineering hat on than when I have my user hat on. You get annoyed with things that you might not as a developer, watching fancy animations over and over again because they make the screen transitions cool comes to mind as something a developer might like but a user is going to eventually say “just get on with it.” I know powerpoint does a boatload of wipes and dissolves but I rarely see more than the default setting unless I’m watching the first handful of powerpoint presentations the person has made.

Anyways, I’m eating my own dogfood now and it tastes pretty damned good.

There’s a programmer party down in the valley that I’ve been attending at somewhat regular intervals for the last couple of years called Super Happy Dev House. Mike and I were invited to the party by the elusive Nathan the Kraut ( this is Mike the Quaker’s nickname for him ), it is always interesting because you see a lot of different things and get to talk to the people doing them. There are teams there doing all sorts of things at various times I have met the Dojo Toolkit guys, the Meebo guys, the Flock people, Redswoosh ( in fact I think they bought their domain at one of the first SHDH’s that I attended but it was well into a night of perl hackery and free flowing redbulls so I could be wrong ) and a bunch of other interesting people.

Anyways, Mike the Quaker and I attend these things together and are normally at a loss to have anything interesting to work on, so we actually came up with a project to work on specifically for when we go to SHDH. We called it cackl because cackl is a very web 2.0a sounding name so we wouldn’t stand out, and because we wanted to do a sort of joke database of epic proportions.

The odd thing was that since we had the idea we have only attended one more SHDH and left early so we could go somewhere quiet and work. GO figure.

So…after a couple of restarts, a lot of caffeinated beverages, much pizza, and several nights of questioning whether we really know what we are doing we have launched cackl.

What is cackl?! : cackl is a joke database of massive proportions. It supports tagging/searching/etc. All wrapped in an easy to navigate interface so that you can be sure you are learning new jokes to bore your friends with within 2 seconds of hitting the site. It is social in nature, so if you have a good joke that you’d like to add, just create an account and add the joke.

So if you need a joke, or have a joke you want to share, please go over to cackl and take a look around.

History  

I finished the basic functionality of hamtesting.com in July 2007. I wrote the whole thing in Ruby/Rails in a week and then spent some time trying to deploy it, dealing with a bunch of issues related to inefficient xml processing etc, making it play nicely with apache and whatnot. While sitting at a Super Happy Dev House I finally had enough with trying to deploy Rails when I already knew all of the issues with PHP, so I sat down and rewrote it in PHP.   I launched the basic testing module of the site about a week later after I made it a little bit prettier( all css based ). I then wrote the review module and released that about 2 months later.

Getting The Word Out

I made a couple of lame attempts at getting some visitors to the site. I made some posts to some of the enthusiast groups and told people that I met about it, hoping to generate some interest. However, I naively thought that it would get itself found by the internet since it was vastly superior to most of the free testing sites, and even superior to most of the pay software/sites. Unfortunately it never seemed to generate much interest, I have some guesses at number of visitors in a given month and it’s VERY VERY OBVIOUS that the word ISN’T getting out.

 Using the Brain

Highly underrated, I used my brain to think about why people weren’t using the site. Then I typed in “ham testing” and “amateur radio testing” and “arrl testing” into google and started clicking back through the pages. I never did find my website. So how are others supposed to find it?

The Plan and The Goal

They say you need a goal and a plan to get there. So I have a goal. Hamtesting.com showing up on the first page of results when some keywords I’ve identified as being the most relevant are searched on. The plan is all laid out, I’d like to get triple the traffic of my biggest month thus far by the end of April, and then I’ll set some more goals.

The Baseline

In an effort to measure progress I’m going to put up some very unscientific numbers for the last two months :

january 2008 : 102 page impressions

december 2007 : 65 page impressions

november 2007 :  63 page impressions

So with those numbers as a basis perhaps we can see how effective my efforts are at generating some users for the site. I really believe the product is top notch and can improve people’s chances at passing their tests and learning what they have problems with quickly. It’s top notch, just gotta get the users!

I’ve been busy with school and other things that I’m not going to talk about at the moment *cough* cackl *cough*, however I finally got the question pool review system built for hamtesting.com. With this release, we now offer more functionality for your dollar than any other ham test prep site on the internet.

There are a few things that need to be tweaked, but this was the last major piece that needed to be done so that people had a useful tool for preparing for their tests. I’m sure that there is going to be some tweaks, feature enhancements, that will be done, but none of the other features are going to require as much work as the review system needed.

What happened was that I wrote the practice tests without using a database, everything was done using XML. When I started working on the test review system I realized that I needed all of the data in the database, which meant that most of the classes that I had used for the practice tests weren’t really usable for the review system. In the end, it wasn’t as bad as I thought it was going to be, but it was certainly more work than it should have been.