Moving to WordPress

Blogger is too limited in features...I was just trying to put an HTML table designed with Frontpage, looking perfectly fine when it was offline...it looked horrible on blogger... :(

I am moving to WordPress...let's see how it compares...

Startup	What they do
Mobile
Burrp, JustDial, Mkhoj, Onyomo	Local search. Burrp is more lifestyle services oriented. All are getting into mobile/SMS based services. JustDial is more into call centre interface and certainly have a head-start.
MGinger, Myzus, Moveo	Mobile (SMS) advertising
Ticket booking & Movie rental
Bookeazy, Movietell, BookMyShow	Ticket booking online. Bookmyshow is recent entrant from the house of Media giant TV18. Others will have a very tough time competing with them.
Seventymm, Madhouse	DVD rental online (Indian Netflix). Madhouse has been acquired by Seventymm. They also have very good funding. With a clear-cut revenue model, they look set for a big success.
Travel
MakeMyTrip, Yatra, TravelGuru Ezeego and so on...	Travel services online. Biggest revenue category, dog fight going on. All are very similar to each other. Here's a small comparison blog.
Travenues, Zoomtra	Travel search engine
Financial services
Apnaloan	Connect borrowers with banks
Real Estate Search
99acres,	Is there some scope for a vertical search engine here? :)
E-Shopping
SpotEazy, ShoppingTadka	Shopping deals search engines
Global players
SlideShare	Share PowerPoint presentations. They now have the one & only, Guy Kawasaki, on board. Look very much set for success.
Zoho (AdvantNet)	Web office suite. Their products are superb, and are certainly ahead of their Google counterparts.
Social Networking
MingleBox, Yaari, Sixer.TV, BigAdda, SERaja	Social networking. MingleBox has early mover advantage. BigAdda is a late entrant from Reliance. SERaja is events-based networking.
Photo sharing & printing
Picsquare, iTasweer	Photo sharing and printing. There are several other players here. Very crowded. iTasweer has struck a deal with Microsoft for having their service integrated with Vista.
Komli, Tyroo, Ozone	Ad networks. Yahoo picked up stake in Tyroo.
Uncategorized
Myntra	User created designs (T-shirts et al)
Maps & GPS
RouteGuru	Road directions
MapMyIndia	Similar to Google Maps.

Marc says, you don't need to plan too much...

In the 7th article of this series...Marc is trying to assure the aspiring entrepreneur, that he does not need to, rather he possibly cannot, figure out exact details of the business plan or revenue model, all upfront...I was slightly disappointed at the post though...as it focuses too much time on the story (How Edison accidently invented the phonograph)...but too little on actual concrete analysis...unlike his previous more elaborate posts.

Scrum Pros and Cons

When I was at Doubleclick Inc, we were part of Scrum rollout. We had Ken Schwabber himself coming in as a consultant guiding us on the rollout and the problems we were facing.

I think Scrum was really helpful...here is how I would summarize its strengths and weaknesses...

Pros:

1. Scrum encourages teamwork and transparency. It helps breaking down hierarchies. It helps even a fresh developer to understand what an engineering manager and a product manager is doing (even though he might be clueless initially on what people are talking about :) ).
2. Scrum focuses on user features. This is its strength as well as weakness sometimes (see below).
3. Scrum is adaptive, and gives a good visibility to the management on development effciencies of all teams using normalized benchmarks (though this normalization would be difficult to achieve in practice).

Cons:

1. Scrum focuses totally on user features. It might be very tempting for the team to get quick-and-dirty rather than slow-but-clean approach. Simple speaking, non functional requirements like clean scalable archiectue, clean consistent programming practices, consistent documentation, security of the system, memory footprint can easily get ignored, beacause they have no direct impact on user experience. It would be really cumbersome to clean up the system later on, when your userbase grows quickly, and hackers start playing with your system.
2. Theoretically, Scrum asks every team member to forget his specialization (Developer, QA, BA etc.) and work as a team member. But, I could not comprehend at all, how a BA would understand about architecture and scalability when he does not know the difference between Java and JSP?
3. It is primarily useful for product based companies. It does not focus on creating empirical feedback data that'll be useful across all projects (needed in services based companies). It does not focus on handovers (which is needed regularly in Indian high-attrition scenario :) ) and documentations needed for handovers. Power of simplicity is ok, but just removing some concerns from one's perspective (as against CMMI, which is a very heavy process pyramid in comparison) does not make them unreal.

Of course, I know very little about management or processes, but this was what my perspective, as a developer...but didn't dare to argue with Ken Schwabber then... :)

HashMap sorted based on values...

Just wrote a simple HashMap class to have entries sorted based on values...i.e. the keySet() method will return the keys based on sorted order of their corresponding values...

Here it goes...


import java.util.*;
import java.util.*;

/**
* HashMap with entries sorted based on Values. (keySet() method will return keys sorted based on values.)
* @author Sumedh Inamdar
*/
public class ValueSortedHashMap extends HashMap {
  
  public Set keySet() {
    TreeSet s = new TreeSet(new MapValueComparator(this));
    s.addAll(super.keySet());
    return s;
  }
  
  /**
   * Comparator class for comparing keys based on values. Assumes that values are 'Comparable'.
   */
  class MapValueComparator implements Comparator {
    MapValueComparator(Map m) {
      map = m;
    }
   public int compare(Object k1, Object k2) {
     if(!(map.get(k1) instanceof Comparable)) {
       System.out.println("ERROR...ValueSortedHashMap could not compare the values...");
       return 0;
     }
     return ((Comparable)map.get(k1)).compareTo(map.get(k2));
   }
   Map map;
  }
}

By the way...does any one know a quick way to post formatted code on site? This was pain in a$$...Blogger just removes my spaces after I do some editing...and for even adding spaces, I had to hack it like this :(

India Tech Startup scene - Social Networking

For last few weeks, I was following indian tech startup scene...started by reading some excellent articles by Paul Graham and Guy Kawasaki...got particularly interested after reading a detailed report in Business Today (3rd June issue)...VC investments in India shot up from $17 million to $166 million in 2006...the scene is certainly red hot...

Here is a brief summary of web 2.0 (or just web...hell with this 2.0 or 3.0 crap :) companies...I will cover it categorywise…

Social Networking: MingleBox would of course be the first name coming on top...especially getting that sumptuous sum of $7 million from Sequoia. Then there is yaari, Jhoom, Fropper, DesiMartini, Indyarocks and not sure how many more…everyone is ‘Web 2.0’ with all the latest features of sharing photos, videos, blogging, yada yada yada. Find Amit Ranjan’s (of Slideshare.net) take here. There are some who are trying to find a niche. Sixer.tv for cricket fans, or saffronconnect for music for example. It will be interesting to see who wins…there should not be more than 2 or 3 players in general category when war is over. Minglebox is tapping into the college campuses, by creating readymade college campus groups…really smart move...that will make the new generation stick on Minglebox after we all being taken by Orkut...
Orkut looks so passe with these agile newcomers…I wonder why on earth orkut was successful in India, in spite being nowhere in US...It will be interesting to see how the newcomers use mobile connectivity to a good effect. That can prove a main driver for people to switch from orkut…right now, the only reason I am active on orkut is that I have my friends there…the fundamental reason behind the term ‘critical mass’.

Don’t know if there are clones of variants like Fanpop (fan-following-social-network) or twitter (find-what-I–am-doing – the original purpose of weblog) in india…may be some people are working on those already…it doesn’t take long to have such clones these days, with internet minimizing geographical distances and spreading knowledge unbelievably fast…one more innovative venture is Ning…by Marc Andersson of Netscape fame…it allows you to create your own social network… of course it won't be much better than creating groups inside a normal social network as of now…you won't be able to really edit it the way you like...beyond adding a few widgets...but they will evolve…Facebook application platform is the latest buzz in this space, and is attracting huge crowd within very few days.

By the way, Marc Andersson is a fantastic read for people like me (startup aspirants)...a series of articles on life-of-startups here. There are 6 excellent articles in this series so far.

Internet Advertising Economics

With web 2.0 startups popping up everywhere like mushrooms these days...it is getting difficult to distinguish real diamonds from the fake ones. Everyone is offering a great free, consumer friendly, lightweight, AJAXified, ultra modern Web 2.0 service, and expecting to make it big with Advertising revenues.

Total indian Advertising industry would be a minuscle 100million$ as compared to app. 16 billion $ market of US. Find an interesting discussion here. Moreover, it's not only about advertising. Advertising is a mirror of industry depth. For example, a business model like RearDen, a company focusing on business travel, would be useless in a country like India as of now, as we are far from being habitual in using internet for day-to-day needs. Moreover, we can afford to hire a person as a secretary, which Rearden tries to get away with. So, the complexity and depth of industry, especially the industry that heavily relies on internet is nowhere near US.

Even among current internet userbase of 20 million, it would be very interesting to know what is the amount of money being spent per person on internet, on e-shopping, or ticket booking. Of course, the internet awareness will increase very fast. Today's college goers are very net savy, even in small towns. They are tomorrow's earners and comsumers. So, picture after 10 years might really be radically different.

Quite a few startups have come up in online ad networks space.

Tyroo sold their stake to Yahoo very recently. They also have an Ad agency, Quesar media, under the same umbrella (Smiler interactive group). So, it might be their key USP.

Komli, is another one, started by Amar Goel, an ex-microsoft. They, I believe, are focusing on optimization of delivery, using behavioral targeting as their key USP. But, statistical methods and optimizations would be helpful when one already has a big portfolio of media and advertisers available. Google can show extremely relevant ads, based on geography and content type, only because it has a huge footprint among advertisers, as they say, 'the long tail' or 'bottom of the pyramid'. By the way, does anyone know what this name, 'Komli' means? :)

There is one more, called Ozone media, but not sure what they are upto.