SEO considered harmful

Companies go to extreme lengths to optimize their web content so that their pages show up on Google search. Showing up on top 10 of a google search will dramatically increases the number of visitors to your site. While this is all goodness, there is some real stupidity in Google's algorithms. For example, if your URL contains the search term, Google will rank you higher. This means creation of funny URLs that match the search terms. Add Long Tail keywords to the mix, you have the internet full of web pages with horrible quality content, but using right keywords in the URL and the text.

Google would argue, if the content is not good, they would see a higher bounce rate. Thats not true, if the webpage has links to other pages that promise to tell you more about the subject or cleverly designed adwords. Overall, the bouncerate is not a reliable indicator of page quality.

We are at a point where the content on internet is becoming mostly useless ( some would argue that it has always been the case). What needs to change?

1. Google should change their algorithm, so the URL words do not have much importance
2. Have human editors, randomly look at the content and aggressively penalize sites that has made up content. Perhaps, Crowdsource this effort.

There is no good fix for this issue though, a better alternative might be emergence of an alternative search engine that has different ranking mechanisms. I would love to see Google lose its marketshare, so the search monopoly is not impacting the quality of the content on the internet.

JavaEE app servers are slow

Not a single day passes without hearing that their JBOSS, WebLogic, WebSphere or GlassFish is runinng very slow. I see tweets or comments or blogs on this topic. One has to wonder why wouldn't these guys move away from Java EE app servers. Here are few reasons:

1. Political reasons - deal has been sealed at a golf course, and avg developer/IT ops guy has zero say or influence in changing that decision.
2. Resume Building - developer believes using a massive app server means more buzz words on resume.
3. Vendor hijack - Vendor tells that they won't support the app unless its running on GlassFish or JBOSS or WebLogic or WebSphere.
4. Leading Figure says Java EE is becoming simpler - I call this bullshit factor - these leading figures rarely write code or have to support production apps.

And there are many other reasons that continue to cause heartburn for developers and IT operations.

Why I do more searching on twitter than on google

In last 6-9 months, most of my searches are on twitter instead of on google or bing. As I stopped to think about why this is the case, few things came to mind:

1. Majority of my searches are for real-time news and/or information
2. For non real-time, research related searches, its useless to use a search engine, instead, I just go to wikipedia.org
3. For real time news, if someone hasn't tweeted about it, its probably not worth my time to read it
4. twitter search is a nice way to determine the importance of an item ( you get to see how many more tweets since you started searching ).

So, while many people think of twitter as something that they don't understand or something that doesn't have business value, if you start using it, you get addicted to it, and eventually this is going to impact the search engine traffic. Not going to see a major dent in search engine traffic in 2010, but I think by 2012, the internet users behavior would have changed.

And, don't be surprised if twitter were to buy google in 2012.

If you are a startup and thinking about product management, where do you start?

A friend recently called me asking for my thoughts on product management, I shared the following links with him. So, here are my recommended courses/blogs/tweets/books for product managers. This is by no means comprehensive list.

Pragmatic Marketing website: http://www.pragmaticmarketing.com/

They offer courses in different cities, would be a good course to check out. A word of caution, following all of Pragmatic Marketing framework for an early stage startup will be a overkill - instead focus on the concepts and the spirit of what they say ( for ex: get out of the building to learn about the market ).

Then, read as much as you can about customer development. It may end up saving you and your company.

Read the following blogs:

Customer Development by Steve Blank: http://steveblank.com/

Lean Startups: http://www.startuplessonslearned.com/2008/09/lean-startup.html ( Eric Ries is a good blogger that covers customer development).

Read this book:

http://www.amazon.com/Four-Steps-Epiphany-Steven-Blank/dp/0976470705/ref=sr_1_1?ie=UTF8&s=books&qid=1265057140&sr=8-1

This book is hard to read, but its not because its not well written, its because, almost every section makes you think hard and you will find yourself thinking about the past mistakes - its like going through an Yoga cleanse.

On Twitter ( If you are not already on twitter, please consider a different profession - I hear there is shortage of roofing contractors ):

Follow @sgblank, @ericries and @davemcclure ( and the ones they recommend from time to time ).

Apache releases Tomcat 6.0.24

After weeks of discussions, testing, bumping up release numbers, Apache Software Foundation has finally released latest release of Tomcat 6.0.

Tomcat 6.0.24 has several important bug fixes, but the one that I am most excited about is the fixes made to stop memory leaks in applications, when they are using shared libraries ( for ex: JDBC drivers ).

Anyone running Tomcat 6.0 in production should upgrade to Tomcat 6.0.24 immediately. My colleague Jason wrote a detailed blog post here: http://blogs.mulesoft.org/apache-releases-tomcat-6-0-2/

Ofcourse, to get enterprise features, you really should be running Tcat Server, which is vanilla Tomcat server. Download it here: http://www.mulesoft.com/download-tcat-server-enterprise-tomcat

Disclaimer: I am the product manager for Tcat Server, so its my baby and yes, I am biased, but hundreds of others told me that we built a solid product in Tcat Server and they are running their apps on it.

What am I doing lately?

I have been doing product management for MuleSoft Tcat Server, the leading enterprise solution for Apache Tomcat . I have been having lots of fun competing against SpringSource's tcserver.

This kept me busy to blog here on a regular basis, but you can follow my blogs on blogs.mulesoft.org.