Wednesday, August 29, 2007

Yet another controversy for Orkut.

Yesterday our news channels somehow managed to squeeze a bit of time out from more pressing matters; such as the visit of Samlan Khan’s family to the jail on Rakhi or the public display of affection by the constabulary towards Sanjay Dutt, to show us the recent controversy about Orkut. Apparently, someone in Orkut has created a community called “I hate Mayawati” and the UP chief Minister has taken umbrage (if that’s what people take when they are offended) to some of the postings. Our overenthusiastic and extremely responsible media even flashed the name and hometown of the person who has created the community. One of the senior government official commented that “the matter has been forwarded to the cyber crime department”. Surely our governments have got there priorities right. This is exactly what the cyber crime department should have been doing all along. Now all they need to do is to call the army to hunt down that poor soul who took the “freedom of speech” phrase a bit too seriously.

However, I’m glad with the reaction of Google. They have not removed the community yet. They should treat this just like any other abuse report. It took us a lot of effort to remove the “I hate India” community from Orkut. It’ll be interesting to see if Google changes its review procedure to soothe the government officials.

This brings me to a more generic question, i.e. should Google (or any other website) be responsible for the user generated content. If the answer is “yes”, then is removing the content after someone has reported for abuse good enough? Would you trust the sites to act immediately, if it’s clear that they stand to gain by not removing the content?

Brijjes and Tribes

In the past, I’ve written about how it’s difficult for the new players in the social networking market place to “steal” users from the incumbents. So out of curiosity I decided to check out how the one month old Brijj is doing in the market.

In the absence of any other reliable source, I turned, a bit reluctantly, to Alexa. Some of us, who have been keeping track of Alexa rankings for different sites, usually view these stats with the same degree of skepticism typically reserved for things like Saurabh Ganguly’s call for a quick single or Munaf Patel’s declaration that he is completely match fit; i.e. things that have let you down in the past.

The Alexa traffic graphs for Brijj showed that there was an initial peak and after that it followed a downward slope (This is obviously along the expected lines). But, of course these graphs in isolation didn’t make any sense. I thought it would be more interesting to compare the graphs of Brijj with those of the other players in the market.

A direct competitor for Brijj is techtribe; both are professional network sites targeted specifically to the Indian audience. Techtribe is in market for over a year now; so I expected it to have a greater reach and higher page views than Brijj. But I was in for a surprise. The Alexa graphs showed that Brijj has almost 4-6 times more reach and 3-8 times more page views. Techtribe is ranked 2045 in India; as opposed to Brijj which has a very impressive rank of 371.


Now, I’m not in a position to do a feature by feature comparison of the two sites. I’ve never used techtribe and the time I’ve spent in Brijj would make a typical Sehwag innings look longer. But on the face of it, techtribe looks more content reach (and understandably so). Techtribe has communities, knowledge sections, job postings and so on, which are as of now not present in Brijj. So the question obviously is how Brijj managed to generate so many page views. Is it due to the marketing efforts put in by Infoedge? Is it due to the naukri brand name? Is it due to the fact that Infoedge had a much larger employee base to start the word-of-mouth campaign (seems like a trivial point, but some of my “sources” have told me that Orkut was launched that way)?

I think it’s just a matter of when and not if Brijj would introduce features like job referrals, communities, answers etc. It would be interesting then to see if Brijj can challenge the bigger players like linkedin.

[For some reason the Alexa webservices are not returning any result. I thought of using their widgets to put the graphs directly in the blog, but in stead of the graph I only got a blank space. If you know what's the issue, please let me know]

Thursday, August 16, 2007

Cross the Brijj only when you come to it

[This post is partly a review of the new professional networking site Brijj; and partly (or rather, I should say, mostly) a response to another review “A brijj too far” by Shyam.

I must clarify upfront that my views may not be considered completely unbiased. I’ve worked at Infoedge in the past, so might have developed a soft corner for the team. I’ll try to keep these personal biases out of the post
]

Let’s start with Shyam’s review “A brijj too far”. Once we have ignored the first three paragraphs, which tell you how Indian firms are only good at copying stuff from the west and how only the developers’ lovers use their product, we reach the point where the author starts the review of Brijj. The most admirable part of the review is the discovery of the redundant Meta tags and the invisible div tag. Hats off to Shyam for discovering these things, which have supposedly escaped the notice of the QA team. But my question is how do these things affect the user? May be I’m an exception, but when I’m surfing the net I hardly ever go through the source code of each page to check the meta tags. So what Shyam is trying to tell us is that the code is not polished. Big Deal! Isn’t that what you would expect from a beta version? When the time to market is so important I would much rather have my technical team and QA team focus on product features than spending their time in refining the code.

Next, Shyam complains about the absence of privacy settings for the users. Interestingly, even LinkedIn doesn’t have such a setting (some of the granular settings were added much later; for features such as your public profile or Q&A section; features which are so far not present in Brijj). So the only “privacy” setting that is present in linkedin and missing in Brijj is whether you allow others to see your contact list or not. I’m sure the team must be having this on their priority list. But as far as privacy is concerned I think Brijj does a much better job than the personal networking sites like orkut, myspace etc. Just like LinkedIn, it tries to handle most of the privacy and spam issues by itself rather than asking users to specify these settings.

Then Shyam talks about Infoedge’s “critical mistakes like not having a common registration database among its properties”. Mind you this line comes from a guy who was just complaining about privacy. The point Shyam is missing is having a common user database is actually more beneficial to the company than the users. Most of the times it’s the user who doesn’t want to have his data aggregated. In any case, I don’t think it makes sense for the products Infoedge has; 99acres hardly requires a registration, most of the users would use jeevan saathi for a very short time (at least I hope they would), and in naukri although there is a repeat usage, the frequency won’t be very high.

My favorite line in the entire post is: “I am assuming that at some point Info Edge will roll into Brijj, the muscle of Naukri’s database, but they have again erred gravely by not having it on from day one”. Networking is a social activity and job searching is mostly a private activity (by this I mean most of us don’t want to broadcast the fact that we are looking for a job switch, we might contact some selected people for this). And what naukri database is Shyam talking about? If he is referring to database of resumes and candidates; I’m not sure how that can be merged with Brijj. Invite all the naukri candidates to Brijj? They might as well send a mail to my HR manager and my boss saying that I’m looking for a job change. And if Shyam is talking about the database of job postings, that I’m sure would come much later in Brijj, but in a different form. Brijj would help companies identify candidates mostly through referrals (that’s the benefit of networking, isn’t it?).

So what’s my verdict on Brijj. I think it’s way too early to write a review for the product. The true value of a networking site can only be gauged when it has gained a substantial user base (“positive network externality” guys, remember I talked about this here?).

Brijj definitely has some cool features to start with. The UI looks pretty decent (I generally don’t care much for UIs, but it matters to most). They provide a type-ahead functionality for company names and educational institute names. The importance of this type-ahead functionality would be clear when you look at the “People you may know” section in home page. Allow me to explain. Even linkedin has a “people you may know” section, which tries to match the company name/ institute name to find your ex-colleagues and alumni. But linkedin misses a lot of potential contacts because the company name is entered in a free text field. If user A enters his company as “Infoedge” and user B enters it as “Infoedge India Pvt. Ltd”; chances are linkedin won’t recognize them as colleagues. This problem would be addressed in Brijj to a great extent, since most users would select the name from the type-ahead list.

I’m sure what we are seeing is a skeleton version of what’s planned for the site. Not because the team accidentally missed implementing those features; but because most of these features only make sense when you have build a large enough user base. So let’s give Brijj sometime before we give our verdict.