The XORBlog

XORBlog has finally moved to its rightful place at www.xorblog.com. This transition, however, would have been a lot harder had it not been for all the following help:

1. How to get Typo running on Dreamhost

Here are the instructions.

There were no additional steps. The instructions on Aiden's blog were spot on.

2. Get Jroller to generate full RSS 2.0 feed with comments

The instructions came from Null Pointer, but I made a minor change to add the IP tag in comment section with following line: <ip>$comment.remoteHost</ip> .

The full template is here.

3. Import with Typo converter script

Typo comes with a migration script called feed.rb in db/converters directory. Unfortunately, this script doesn't support import of comments, so I extended the script to also import the comment and publish the entries and comments upon import.

Here is the script.

Note: If you don't want the articles and comments to be published. Make sure to comment out following lines in the script:

a.published_at = a.created_at
a.state = ContentState::Factory.new('published')
a.publish!

Run the script:

ruby fullrssfeed.rb -a <blog_author_name> -u http://jroller.com/page/<username>/<template_name>

That's all. Just go to your Shiny New Blog and it should have all the articles and comments.

After being hosted with jroller.com for over two and half years, the XORBlog is finally moving to its own host. With this migration I also hope to blog more frequently. The migration process itself (installation of typo on dreamhost and content import from jroller) will be subject of my next post as soon as I find some time to put it together.

Again, thanks a lot to wonderful folks at jroller.com for keeping it running for such a long time.

The idea that small web 2.0 startups need to start integrating their applications to form larger "application suites" has been floating in my head since Kiko decided to put themselves up for sale on eBay. Had Kiko integrated its service with an email provider or other service, they would not become irrelevant the moment Google released GCalendar. There really is not much reason for a standalone online calendar to exist. In other words, it's a feature and not a product. But does it mean that you should not create an application that might be more suited as a feature? In my opinion, the answer is that you should create such a product but partner with other services that need this feature. It's not just a great distribution plan but also helps you compete with bigger players.

And this is also what Forrester is now proposing in their report at CIOs Want Suites For Web 2.0 (via R/W Web).

I previously blogged on this topic at: Integration is the killer app (even in Web 2.0).

Long time no blog. I have just been too busy with my new job. But the experience I had today I thought was worth sharing.

A while ago someone sent me a link to a David Blaine paroday on YouTube. I really liked it and after about 2-3 weeks wanted to show it to my friends. But I had forgotten the name of the magician. The only thing I remembered was that it was on YouTube. So, I go back to YouTube and try to find it. I use keywords like magic, magician, trick, crazy, funky, freaky and so on. The things I thought described the video. YouTube always came up with a search result of over 200K videos. I went up to 10th page and then gave up. There is no way you can browse through all those videos and find the one you are looking. So, I give up on YouTube and head to MetaCafe.

The same search query on MetaCafe results in far less search results (they simply don't have those many videos). I thought it would be easier to go through these search results (the assumption I made was that the video was so cool that it was bound to be found on other video sharing sites). Indeed the video was found around 4th page.

I wonder if there is a lesson to be drawn from this. As the content on a site grows it becomes very complicated and difficult to find it. When the web was small you practically "knew" all the sites on the Internet, then came directories and search engines. It seem to work fine for text, but video is a whole different ball game. One of the solutions to this problem was supposed to be tagging, but it's definitly not enough. So far, it seems that the assumption that if something is good enough it would be also found on smaller sites holds true. What makes it easier to find things on smaller sites? The fact that you can browse through all the content much faster. The size does bring some disadvantages with it.

The video in question is here.

TechChrunch has a new post on this topic at: Top social media users getting paid; is the balance shifting?. There seem to emerge some economics of social networking (for participants). Recently, there has been a lot of talk about most social networks (Wikipedia, Digg, YouTube, ...) being driven by a small group of users. Roughly under 20% of the participants. What motivates this small number of very active users? Some companies are adding money to the mix and some of these top users are actually getting paid. Those not getting recognized (paid?) seem to feel underappreciated. That means social or moral incentives alone are not enough to keep users interested. Revver.com is playing this card. Their punch line is: Revverize your videos. Share them with the world. Make money.

It seems that social network/citizen media is a great way to start a new user driven web community, but might not be enough to sustain it in the long run. Of course, it's a developing story, and we will see how it unfolds. Perhaps we can draw some analogy from the open source software. Most projects are started by inspired hackers, but once they hit the critical mass they end up becoming commercial one way or another. Linux, Apache, Eclipse, JBoss, etc. all seem to have followed that path.

And as always there are notable exceptions, so far reddit.com seems to have not fallen into the trap.

OK, so here goes the link: Technorati Profile.

I have been wondering for a while about why people participate in so-called social network websites. If we can figure out what makes people participate, we can of course, provide the right incentives for participation. Some of the reasons for participation are:

• Bragging rights – Slashdot karma, Reddit karma, Digg ranks/home page submissions
  
• To promote you version of truth – Wikipedia
  
• To troll
  
• To meet people online – Find people with similar interests
  
• To learn and spread the knowledge – Wikipedia, discussion forums
  
• To make world a better place – Product reviews (if unpaid and honest)

Of course, that is not a complete list.

Pepole respond to incentives and most incentives (in general, not only for participation in social networks) can be classified into:

• Social
  
• Moral
  
• Economic

Most reasons we saw above fall into the social or moral types, but recently it seems some new social network sites are trying to bring in the economic incentives for participation. They are downright paying the people who post the content, comment and generally make it look like there is some activity. Many, specially the existing players, question the morality of such incentives. But is it really so bad? I don't think bringing money into the equation in itself is wrong. Just as having advertisments along with editorial content is not in itself wrong. If you are able to clearly mark the paid participation against voluntary participation, it could work. But trying to simulate the traffic and activity by paying people without telling others is wrong. Actually the term for this practice is astroturfing.

But I think the established players need not worry much about this. It takes a lot of traffic and active users to build a successful social network website and no matter how deep the pockets of some of these new players are, they will never be able to pay enough users. All you need to care about is building a system where users participate because they want to and find the experience great.

Here is the washington post story about social network contributers getting paid: http://www.washingtonpost.com/wp-dyn/content/article/2006/08/25/AR2006082501308.html.

This is interesting: To google or not to google? It's a legal question (via Slashdot).

This reminds me of an old marketing lesson that, you want your brand to be popular but you don't want it to become synonymous with the product. Then you may even lose the brand. This happened to "thermos", which was originally the name of the company that made vacuum flasks.

Some other brands that became synonymous (Genericized trademark is the technical term) with the products: Escalators (for well, escalators ), Aspirin (acetylsalicylic acid), Xerox (photo copiers), Surf (a washing powder in India), Hoover (Vacuum cleaner), even Colgate (for toothpaste, in parts of India).

Linux (the trademark) itself was at the core of some such legal battle (I can't find more info on that, was it in Australia?). Someone please post a link if you know more on the Linux trademark issue.

In a previous blog post I had stated that SLED had some edge over Ubuntu. As I continue to update my Ubuntu install it seems to be getting better. Here are some things that have improved over last few weeks:

• Compiz update: There was a recent compiz update that makes 3D effects way faster.
  
• SLAB for Ubuntu: SLAB can be installed for Ubuntu and works as well as any other distro.

Most people seem to think that entrepreneurship is about stumbling upon a great idea, funding it, executing it and eventually profiting from it. Now, that's a very logical and time proven process, but does it really work like that, specially the part with discovering a great idea. A great idea is, no doubt, essential to creating a great business but, as I have observed, most businesses don't necessarily start with the killer idea. They start with some idea, a business plan (how are they going to make money) and a lot of passion. It's extremely rare that an entrepreneur will stick with the "original idea" for a long time, of course, there are notable exceptions.

Once you are in the process of building a new business, the ideas come from all over the place. Everything seems to inspire new ideas, and that's not because you are brilliant (though everyone would like to believe so ) but because you are in the process of looking for the ideas. The key here, I think, is to pick something reasonably good and start elaborating, and just doing it. Over time you will figure out what works and what doesn't, you improvise and eventually you succeed. Some of the great businesses actually started this way:

• Wrigley‘s found out that chewing gums were more popular than baking powder, after they started giving them away free with baking powder to increase sales
  
• Intel was a memory chip company long before it became a microprocessor company
  
• Sun Microsystems was a workstations company before it became a server company
  
• Flickr started as game-never-ending and still uses the same technology
  
• Skype is built on Kazaa architecture
  
• 37signals was a web development shop before they got real and built backpack, basecamp, etc. for their internal needs and released rails in the process

And this pattern repeats over and over. Of course, there are times when an idea/product/service has run its course or the industry is in transition. An example would be IBM reinventing itself as an IT services business. But that's more of an innovator's dilemma thing (large companies hesitating to innovate in new niches), my point is that the founder's "original" idea rarely ends up becoming the killer app. To put it in philosophical terms: Entrepreneurship is a journey not a destination (cool ain't it ).

So, if you are not in the process (at least mentally) of entrepreneurship, chances are slim that you will have that great business idea. More than anything it seems like you should stay in the game for as long as you can and iterate till you discover that killer idea. It also depends on what motivates the entrepreneur, the money or just the joy of creating something new and useful. If it's the latter the results are generally more innovative and even more profitable. That's probably what separates an Enron from an Apple.