Applied to TechCrunch 50
July 7, 2008 – 2:19 pmI have submitted statzen to be a part of TechCrunch 50.
Hopefully we will be accepted. I will post here as soon as I can.
I have submitted statzen to be a part of TechCrunch 50.
Hopefully we will be accepted. I will post here as soon as I can.
It has been a while since I have posted on the statzen blog. The primary reason is that I have taken a little time for myself. I took a motorcycle trip from Tennessee to Washington, DC and back while camping along the way. It was a great time and now I am back and motivated to continue getting statzen into the hands of people who can benefit from it.
Right now I am sitting on a plane and have spent the flight working on a way to provide reporting in where visitors come from. I currently see the “referrer” report as the biggest missing piece in statzen. Granted, I am trying to do much more than provide a typical referrer report. I want a report on where people came from that follows common sense. Who cares about identical URLs? The real question is which site is sending the most traffic. Some ancillary questions are: Which pages on that site are delivering the most visitors? Are those links recognized by Technorati, IceRocket, Google Blog Search ,etc? Are those referrals from a search engine? If so, what are the most common search terms for that search engine? What about the most common aggregate search terms?
That is a fair amount of ancillary questions. The challenge is to how to keep it simple and useful. The key to simplifying referrals in a way that makes sense is through aggregate reporting. First I need to address polyURLism (www or no www, etc), this way a unique referrer will only appear once. Secondly, providing aggregate information on the domain is the primary information that people need. As long as their is an ability to drill-down, starting with the aggregate provides the most information with the least mental processing required from the user.
This is what I am currently working on. I hope to have this glaring hold in the functionality of statzen into the current users’ hands by the end of the week.

The statzen alpha test has begun. There are some users in the system and we are trickling more in as we have time and know that the system is stable.
The fun has begun!
I have lot’s of opinions; some of those come through in my software. With statzen I am likely going to provide some features that I am opposed to while helping to educate my users why they are not something I recommend. Click tracking in feeds is one of those things.
What is click tracking? Click tracking is a way to collect information about people who clicked through to your website from a feed. The way click tracking works in a service like statzen is that instead of going directly to your blog from your feed, readers will actually be directed to a specially crafted link on the statzen server and then redirected to your blog. The reader still gets from your feed to your blog with no discernible delay, but I am opposed to this method for several reasons that I will get to in a minute.
Not all click tracking is bad. This exact same method is used for most affiliate sales programs like Linkshare and content ads like Google AdSense. These content monetization opportunities would not exist without click tracking. Similarly, the same method of click tracking is used by most email marketing applications. In both of those scenarios, click tracking makes sense; the pros outweigh the cons.
There is really only one “pro” to click tracking. You are able to count how many times a link in clicked and in some instances, where it was clicked. There are several cons. Here is how I see them:
On a philosophical level, I think the permalink is of a higher importance than other links on the internet. A permalink not only says that it will always work, but it also indicates that it is the one true location for that post. Other URLs may work, but the permalink is the one you can trust. Changing the permalink (especially to a different domain all together) seem to me to be a little dishonest.
Philosophical reasons aren’t going to sway all the marketing types and rightfully so. They want to get a message out and measure it’s reach. The problem is that the measurements they are looking at are not what they are trying to capture. Here is how.
If I read a feed in my feed reader and I copy the link and paste it into a post. Unless I intentionally clicked through and got the real permalink, I just published the tracking link on my blog. Likewise, if I right click and “add to del.icio.us”, I just put the tracking link in del.icio.us. That may not seem like a big deal because you are extending the reach of the tracking, but…
Because of what happened in step two, the tracking link is now getting the SEO love for the link from another blog and the link in del.icio.us. The same is true of automated aggregators that publicly syndicate your feed. Likewise, checking how many blogs linked to a post (or your blog) in Technorati or Google will not return the instances of people using the tracking link. Similarly, del.icio.us will not see the correlation and you will not be able to know how many people linked to you in del.icio.us (or ma.gnolia, my linkblog of choice).
To me, the cons of click tracking in feeds easily outweigh the pros; it is just not worth it. This is especially true if you have other ways to capture the data that click tracking is trying to capture. I imagine that I will eventually add click tracking to statzen, but now you know why I don’t see it as a priority and will likely caution people against using it.
Ok, I had hoped to start letting alpha users into the system two weeks ago. That sure didn’t happen. I got swamped at work and just couldn’t get it out the door. Although, if I am being perfectly honest I should admit that I am struggling with releasing an imperfect product even with the alpha moniker which basically means “this software is still in the sucky stage”.
The good news is that I seem to be back on track now. I have pushed out a major release of the code and discovered some new bugs that need to be worked out. I am squeezing in a few more pieces of functionality too. Tonight I created the email to alpha users and figured out the process for creating accounts and granting access. All the while we are still working on ways to get this thing funded so that we can go full speed ahead.
I haven’t sent out the first round of invites yet, but I am getting painfully close. I am hoping to start allowing people to trickle into the system at a pace that I can control to ensure that the system stays stable.
So much to do; so little time. I am loving every minute of it though.
I have been somewhat stalled on statzen development for the past 2 weeks. I have a few things I need to do before I start letting some alpha users in, but my life (aka day job) is not cooperating. I have to find a way to free up my time.
The good news is that the work I have been able to do has been very productive. Many of the things I have been wanting to implement with URL unification are getting checked off the list. The application is getting more useful, but there are still lots of interface tweaks and additions that I need to add before it will be user-friendly.
So many things to implement. So little time. More soon.
Last night it became painfully obvious that I had to do something to speed up statzen before I could go much further. After some quick indexes and a little database configuration tuning, some queries went from 5 seconds to milliseconds.
This is a huge step towards letting people use the system. You may have noticed it is not just the database that is speeding up. Development has accelerated over the past two weeks as well. This is mostly because I have been getting some very positive feedback from people I have been talking to. I have also been very fortunate to be able to work smart, not hard. I think last night I deleted more lines of code than I wrote, but in doing so I was able to add more functionality to the system.
While I have not been making releases public, I have been following the “release early, release often” mantra. Last night I crossed the 150th revision in subversion and I have deployed at least 30 releases to the server over the past few months. Once I start letting people on the server I will probably limit releases to once a week or so for a while.
If you are itching to be one of the early alpha testers, leave a comment here or shoot me an email. The time is getting near.
A few months ago I developed a proof-of-concept widget for statzen. It showed a list of the items on my blog that received the most attention (via the web and feed readers). While I thought that was neat, I also wanted a tag-cloud based on attention. I replaced the top posts widget with a top tags widget.Tonight I revamped the widget to have both the top tags and the top posts. There is a little micro menu in the widget to switch between the views. I am also planning to add more contextual information to the ‘this page’ option with the widget. I think it will be neat to share some of the information statzen has about the page that is currently being viewed in the browser.
I really like the idea of widgets as micro applications. Too many widgets are billboards advertising a service. While I am surely branding the statzen widget, I am more interested in sharing the functionality. I mean, once people know what this thing can do they will want to use it.If you want to see an example of the statzen widget in action check out the right side of jaxn.org.
The time has come to have a dedicated place to share information about statzen. Well, this is it.
If you are subscribed to my personal blog you will automatically see the posts from blog.statzen.com in the feed. (Combining feeds into a single feed is one of the many capabilities of statzen).
The primary purpose of this blog is to be a place to talk solely about statzen. I plan to post about development progress and new features as well as business announcements, etc.
It is not much to look at yet. I am more worried about starting to get information out. Branding and a polished look can come later.