New Delhi Times - Gaurav Bhatnagar reporting from New Delhi, India

It gives me great pleasure to showcase the efforts of the Tekriti intern team in the form of LetsCricket.com LetsCricket is perhaps first social networking community targeted towards Cricket fans. Its built on top of the PeopleAggregator platform, which of course Tekriti develops with Broadband Mechanics. I had this project in mind since a very long time and I had registered the domain name more than an year back.

With the interns coming in and the world cup being just a few weeks away, we felt it was right opportunity to build LetsCricket.

LetCricket is targeted towards the typical crazy overzealous cricket fan (like myself). Indians love to talk about cricket (no matter they might have never played ever in their lives). So clearly disusing cricket is a very social activity. On LetsCricket, users can create groups, participate in discussions, vote for their favorite players, maintain cricket blogs and find other people with common interests.

LetsCricket is very much a work in progress. It has been put together in less than 3 weeks by Pankaj and Santosh (with generous help from Gurpreet and rest of the PeepAgg team). We are adding new features in to the site everyday. We decided to make the URL public in the spirit of open development approach. Do give it a look!

Posted by Gaurav at 06:23 PM | Permalink | Comments (12)

Really good news from Marc finally forced me out of my blogging slumber. IBM is going to offer a corporate social networking solution. IBM's solution, called Lotus Connect, offers out of the box common social networking components for deployment within corporate environments.

The IBM package includes five applications: profiles, where employees post information about their expertise and interests; communities, which are formed and managed by people with common interests; activities, which are used to manage group projects; bookmarks, where people share documents and Web sites with others; and blogs, where people post ongoing commentaries.

This is great news because in effect IBM has validated the stance that social networking is indeed relevant in the corporate environment. The feature set mentioned above is a sub-set of what People Aggregator offers. And People Aggregator has been out in the market for several months now. But selling to corporates is hard, especially for  startups. IBM's announcement will have the effect of making big corporates take notice of social networking. The IBM marketing machinery will make sure that the corporate world gets adequately educated about the benefits of employing the "Web 2.0" and social networking constructs to the workplace.

It is also heartening to know that we were ahead of the curve by some distance! It was almost two years back when the idea of developing People Aggregator as a reusable software download was conceived!

So hopefully next time Marc goes pitching to a big company, at least he wont have to answer the "Why would we want that?" question!

Posted by Gaurav at 11:28 AM | Permalink | Comments (3)

Kanwaldeep of BrainGain Media emailed me to inform me about MeraVideo. Mera Video is a self confessed YouTube clone as of now. The site has very similar features with the main difference being that the content is targeted towards an Indian audience. I had two obvious questions for Kanwaldeep:

1. What are your plans in terms of scalability? YouTube was burning more than a million dollars a month on bandwidth costs. How do you plan to monetize the website to keep it sustainable? Are you raising money for the same?

Kanwaldeep:

MeraVideo can be termed ‘as of now’ the Desi version of YouTube but has been designed to eventually generate monies thru various unexplored revenue streams. Some of the revenue streams include: Syndication of Content, A Television show called ‘India’s Amazing/Funny Videos’ and selling of premium accounts for users with high bandwidth usage requirements. The eventual goal is to be a ‘Media Content Aggregator’ for India. The main source of revenue ‘at this point of time’ is ‘Confidential’ and we plan to unleash it with the rollout of our Beta Version slated to be sometime end of this year. I am not in a position to share the ‘main revenue’ or monetization source at this point of time due to its confidentiality.

I agree, the main barrier to entry when it comes to scalability is ‘the high bandwidth requirements’ for running a video sharing site. To address the same, MeraVideo is raising its first round of funding. A lot of angels as well as early stage VC funds have expressed interest in this venture. We expect to close the first round in the next 20 days. The first round of funding is intended to address the following: Infrastructure, Scalability and Manpower, PR.

2. What is your stance on copyright material? I see lot of copyright material on your site. How do you plan to deal with that issue?

Kanwaldeep:

The real truth is that due to the sensitive nature of the subject it is very difficult to prevent any violation infringements with regards to the ‘intrusion of private space’. The other violation i.e. reuse of components such as Music, Clips etc. can be easily monitored using a skilled set of researchers who screen the videos manually. Apart from this we will always have the standard ‘Terms of Use’ and the ‘Privacy Agreement’ as our first defense mechanism. We screen every video that is uploaded on our site. We have taken permissions from respective producers and studios for most of the copyrighted stuff that you can see on our website i.e. movie promos etc. To further this, we are finalizing contracts with Sahara Motion pictures and UTV to allow MeraVideo to showcase their films/promos as well as movies. Being a part of the Producers Guild in Mumbai, we actually are very particular about copyrights infringement. That is probably the reason why you do not see any clips from Indian TV shows (like the laughter challenge) etc.

My Take

In general I am skeptical about clones of any sort. I think MeraVideo or any India targeted video portal will have to innovate locally in some aspects. My guess is that mobile integration could be the killer feature for the desi audience. The fact that MeraVideo has tie ups with big media houses is definitely a big plus in their favor. At the same time, the popularity of YouTube was to a large extent due to the availability of copyright TV content there. It would be interesting to see how VCs react to Video sharing sites post-YouTube acquisition. This is one space where you can not survive long without VC money! All the best to MeraVideo!

Posted by Gaurav at 08:08 PM | Permalink | Comments (19)

Youtube got bought over by Google for a whopping $1.6 Billion (no links here - its all over the place!). Going by how popular the site has become, it was only a matter of time before one of the big guys bought it out. Congrats to them! Our company, Tekriti, started off building one of the earliest video sharing sites back in Feb-March 2005 for our first client. Ourmedia.org was launched in March 2005. YouTube came along almost 9 months later. Ourmedia had a host of video and media experts on its panel of advisors and it had (and has) a team of volunteer editors to wean out copyright content. YouTube on the other hand has the policy of letting copyright content remain online until the copyright owner objects. Ourmedia allows users to download the original high quality video while YouTube only allows online viewing of very low res. flash versions. So for a site that started with a 9 month lead, it feels like Ourmedia was a missed oppurtunity. Of course, to be clear, Ourmedia is a non-profit and making billions of dollars wasnt even the aim and it still gets significant traffic and has over 100,000 videos hosted. But the kind of popularity that Ourmedia should have seen never came along. At a personal level, it helped us get our company started and gave us the initial credibility any startup needs. So Ourmedia remains very special to me.

Looking back, I think there are some very useful lessons to be learnt:

1. Keep it simple and stupid. A lot of time was spent on defining what metadata should be captured while uploading a video. The result was this huge page (login required) with hundreds of form fields to be filled in. Even though most of those fields were optional, it was daunting for the average user to figure out this form. A simple minimalistic user interface helps!

2. Reliability is the key. Ourmedia uses the Internet Archive for storing media. This is great because it provides for free storage. But there were major problems with getting this integration to work. Essentially the video was first uploaded to Ourmedia and then copied over to the Archive using FTP. This was slow and often the servers would simply not respond. I think this is the killer feature of YouTube. It just works!

3. Focus! In my opinion, Ourmedia tried to be too many things at the same time. The site supported audio, image and text uploads along with videos. It supported forums, groups, user profiles and social networking as well. All in the alpha version. In hindsight (which, of course is 20/20), it would have been much more prudent to have done just one core feature (video upload and view) and make it work 100% of the times. Feature bloat is also a side effect of using a prebuilt platform (Drupal in the case of Ourmedia) since you seem to get so many features for "free" so you might as well add them to your site. In reality, adding a feature to an application should be driven by user demand and not by how easy it is to implement.

4. Virality. YouTube allowed embedding videos published there on other websites and blogs. This contributed in a big way to the exponential growth of YouTube. Same can be said for Flickr as well. Ourmedia did not have that viral effect because the user had to come to the website to view a video.

To be fair to Ourmedia, it came into existence in the very early phases of the "web 2.0" phenomenon. So a lot of things that seem obvious now were very new or unheard of back then. Besides, Ourmedia does have the cleanest terms of use - you own everything that you publish on the site. Very few (if any) other video sharing sites offer such generous terms of use. Ourmedia was also not backed by funding so it was challenging to develop, maintain and upgrade the website on a shoestring budget.

Based on our experiences in doing Video portals, we developed a platform for easily creating video sharing websites - its called TekMedia. You can read more about it here.

Posted by Gaurav at 07:01 PM | Permalink | Comments (16)

The latest version is live now! We got lot of feedback on the old UI and have made an effort to clean up navigation and user interface. A lot more is in the pipeline but 1.1 is a big step forward from 1.0. Do check it out!

Posted by Gaurav at 01:58 AM | Permalink | Comments (8)

People Aggregator V1.1 is coming this week! This release is all about improving the user interface and site navigation. Lots of bug fixes and some new features too! Stay tuned!

Posted by Gaurav at 03:17 PM | Permalink | Comments (1)

(via Dave Winer) Why ugly design is important.

This is one of the best ideas I have heard in a long time!

Posted by Gaurav at 11:04 PM | Permalink | Comments (4)

We are constantly getting asked what is PeopleAggregator and what problems does it solve. So I will summarize my thoughts here.

What is PeopleAggregator?

PeopleAggregator is a term that is used to describe many things:

1. It is a social networking application that runs at http://www.peopleaggregator.net. Yes, it appears like yet another social networking application like Orkut or MySpace. Yes, the UI has issues. Yes, we will be constantly updating it. Yes, you must immediately visit it and create an account there. :)

2. PeopleAggregator is also a do-it-yourself social networking system. Which means, if you want to build your own community, you could create it there at the click of a button. For example, one could create a network for Soccer enthusiasts at http://soccer.peopleaggregator.net. No coding required! Cool, huh? (Everything beyond this point on is geek speak. So you can directly skip down to the "Why I should care as a user" section at the bottom of this post)

3. PeopleAggregator is a development platform. This is a key differntiation that is not apparent when you visit the website. What this means is that the entire source code is available for download (Free for charities, non-profits and developers). So anybody can download the code, and host it else where. They can also modify it, extend the functionality, create a new user interface etc.

4. PeopleAggregator exposes open APIs and supports open standards. So all web applications that support these APIs can seamlessly share data between themselves. We also support open standards like microformats, structured blogging and identity systems like OpenID.

Why I should care?

1. As a developer: PeopleAggregator is a sophisticated yet easy to program to application development platform. It is built using open source technology so it is easy and cheap to deploy. It supports easy UI customizations. Upgrades are easy and automatic! So as a developer of social software, you would find PeopleAggregator saves you a lot of the grunt work involved in developing such applications. That leaves you plenty of time to focus on the specifics of your business logic.

2. As a site owner of another social software: PeopleAggregator exposes data thru open APIs. Which means, you can use our API to build compelling mashups that gel with your exisiting application. For example, a photo sharing site can allow its users to also pull in their photos stored in People Aggregator. And if your software also exposes APIs, then it allows us to integrate with it. That provides better experience to users and also drives traffic and visibility both ways.

3. As a USER! If you are a user of social software, nothing I have written so far would make sense to you. Then why should you care? PeopleAggregator puts forward several radical ideas. For example, the idea of allowing you to use your flickr ID to sign up means you have one password less to remember. By allowing you to export your profile data and content and move it to another application, we do not lock down users. So if tomorrow you move on to using another social networking application, you dont lose your data. By exposing "APIs" we allow you to use your data outside of PeopleAggregator as well. BUT all this starts to make sense only when others are willing to play along with us. Sure, we support a single login system, but what good is it if we are the only ones? Yes, we allow you to take your data with you. But that requires other applications to allow consumption of that data. We want to allows users to aggregate all their data in one place, doesnt matter if it is your photos on Yahoo, mail on Hotmail, or list of favorites on delicious. But we can't do that if Yahoo or Hotmail dont let you take your data with you. This notion of the application owning user data is often termed as a "Walled Garden". Today most web applications are walled gardens. They own your data and lock you in. PeopleAggregator is an attempt to change that. It is all about empowering the user and providing them the choice to do what with their content. This, obvious though it may sound, is radical by today's standards. That needs to change. Business models need to emerge that show that you don't have to be a Walled Garden to succeed. Users have to demand better online experience. And that is why you, as a user, should care! Become part of the effort!

Footnote: Tekriti's role in development of PeopleAggregator. This is another question we get asked a lot. Tekriti is the development partner of Broadband Mechanics (actually BBM was our very first client!) and bulk of the core engine and the application was built by our team in Gurgaon. Phil, Martin, Paolo, MarcS and a host of people across the world work with us on this project. Frankly, the vision is Marc Canter's and it took me months to just "grok" it. But once I got it I realized how important and far reaching it is.

Posted by Gaurav at 08:12 AM | Permalink | Comments (6)

I just realized that there are like 5 dozen people already blogging live from here. And wifi is really really slow. So no live blogging for me. I might as well turn off my laptop actually listen to what is being said out there.

Posted by Gaurav at 12:38 AM | Permalink | Comments (57)

We are at Gnomedex today. The conference has the who's who of web 2.0 space in attendance. Michael Arrington is on stage right now. He has a great sense of humor. BTW I think they should ban laptops in conferences. Because right now more than half the audience seems to be just checking email or reading blogs mostly ignoring what is happening on stage. Or maybe they are really listening. Who knows.

I will be posting more later. I also got my first look at Windows Vista yesterday and I am so happy to see it shape up as well as it has! More on that in another post.

gnomedex

Posted by Gaurav at 11:41 PM | Permalink | Comments (0)

People Aggegator is live now. No more invites needed. Come on over!

Posted by Gaurav at 05:45 AM | Permalink | Comments (0)

Techcrunch profiles People Aggregator.

Posted by Gaurav at 05:42 AM | Permalink | Comments (0)

OK People Aggregator survived for almost a day without going down. Trust me, that is a big deal for an application fresh off the kiln and suddenly getting used by hundreds of people at the same time. Then the phone rang at 6 in the morning. The site was down. Fortunately the team had a good handle on how to fix the issue. In the last one and a half year since we launched Ourmedia, we have gone thru a few releases. My big grouse at Microsoft used to be that we never seemed to ship anything (and what I was working on is still several months away from shipping!). So I always enjoy the end game of a product cycle. Its a race against time and Murphy's law seems to go into overdrive. In the end there is of course the feeling of satisfaction having created something that gets widely used. For the last one week, I have been working till almost 2 in the night and then starting the day early by 6 or 7. But I am so "in the zone" right now that I am thoroughly enjoying it. However, needless to say I am starting to get just a little bit exhausted. So if I have been unresponsive on email or not returned a few calls, please bear with me!

Posted by Gaurav at 01:16 AM | Permalink | Comments (5)

We partially lifted the covers off of People Aggregator today by sending out a bunch of invites. A few hundered people have signed up and the system is starting to look like a real community. It is VERY EXCITING to see people use what you have built. It is also an extremely high pressure time because I am constantly monitoring how people are using the system, what kind of bugs are cropping up and in general how the feedback is. The dev team is working in top gear cranking out bug fixes faster than i can type them out :) We then move the fixes on to a staging server where the QA team validates them and makes sure we didn't break anything else in the process. Then we move the code on to the live server. I think in today's web 2.0 world, incorporating user feedback in real time has become almost a requirement. Users are lot more demanding today and the its a "beta" excuse can be used only so often!

Last few days have been very hectic for the entire team across several continents. Of course we are on the home stretch now leading upto the official launch at Gnomedex

Exciting times these!.

Posted by Gaurav at 12:55 PM | Permalink | Comments (8)

Raju has an excellent post about People Aggregator on his blog. The upcoming release of People Aggregator is what keeps me awake at nights these days! The team is working very hard and it is THE biggest and most complex project most of the dev team has ever worked on. Besides, we are a truly global team with us in Gurgaon, India, Phil in New Zeland, Martin in Germany, and Marc , MarcS and rest of the BBM gang in California. One day I will write in great detail the challenges and advantages of doing such extreme distributed development. But first we gotta ship some code! :)

Posted by Gaurav at 11:32 PM | Permalink | Comments (6)

I was part of a panel discussion around social networking on a BBC radio program called Four Corners last night. You can listen to the audio here

This was the first time I was on radio and it was an interesting experience. It is truly weird to hear your own voice!

Posted by Gaurav at 07:55 PM | Permalink | Comments (6)

Ourmedia has over 20,000 uploads now. Here is the breakdown as of today

Total media uploaded: 20045
Videos: 9633
Audios: 5966
Images: 3902
Texts: 544

We also crossed 30,000 users recently. So 30,000 users have uploaded 20,000 pieces of media. That indicates to me that we have a community that is very active. Its not like a lot of people came, signed up and then never returned. That has been an issue with social networks in general. With Ourmedia, we have a managed to create a social network which is focussed on a specific interest area and hence has retention.

We also did a major upgrade of Ourmedia a few days back. In particular now we support RSS with enclosures and media RSS feeds per user and also allow users to create their own groups. Come, join the Tekriti group!

Posted by Gaurav at 10:22 AM | Permalink | Comments (1)

Last few days were quite a learning experience. On Friday, Ourmedia server went down and the site was inaccessible. At first look, it appeared that the apache logs had filled up beyond the 2GB limit. So we promptly reset the logs and the server was up again. Only to go down within half an hour. Memory usage climbed to 99% of available memory which caused mysql to crash. This led to a few corrupted tables as well. Which caused the apache error log to fill up very fast. Which caused apache to seg fault. Hence the vicious circle. It wasn't clear why we were getting this increased load all of a sudden. We did several restarts but within 30 mins, the memory usage would peak out and mysql would crash. More than a day went by like this with server load peaking within minutes and then requiring a restart.

Finally on Sunday, while skyping with Jeff (lead moderator for Ourmedia), we figured out what was happening. Jeff had freed up disk space day before yesterday by (rightly) deleting the temporary media files. But it looks like Google had indexed those images and/or people had linked directly to those images. In particular, an image of Lindsay Lohan was getting very high number of hits. Now with the image removed, Drupal (the CMS which is used by Ourmedia) started returning 404 errors which were getting logged into watchdog. Watchdog ended up with 6.8 million entries in a single day. I think that was causing excessive memory usage and mysql crashes. As a temporary fallback, I changed the code to stop logging anything into watchdog. Since then things have cooled down and we have a uptime of 30 hrs (and counting).

It was quite an experience in systerm administration - something that I have not done much before. The biggest lesson I learnt was that what seems like the obvious cause of a problem is often merely a symptom of something more serious underneath. We still haven't gotten to the bottom of this but now we know where to look!

Thanks Linday Lohan ;-)

PHM@

Posted by Gaurav at 07:40 AM | Permalink | Comments (2)

There is so much talk about social networking these days and how everybody is trying to figure out a viable business model. Friendster, LinkedIn, Orkut, Tribe all get talked and written about every day. But nobody mentions what I believe is the world's most successful social network. Its shaadi.com Its an Indian matrimonial site and has many social networking features. I heard their ad today on the radio and they claim that 500,000 of their members found their match on their website. That is NOT 500,000 dates, that is 500,000 marriages! If that is not successful, then I don't know what is.

Matrimony is big business in India and for a country with a very large foreign resident population, internet provides an easy way to connect with people back home. When it comes to marriage, trust becomes a key issue. Social networks have trust built into them. When you connect to somebody who is n degrees away from you, there are n-1 intermediaries (whom you trust) who can vouch for that person. Shaadi.com doesn't harness that trusted network. And I think that is an oppurtunity to create an interesting social networking application.

Posted by Gaurav at 06:01 AM | Permalink | Comments (7)

[Via Marc] Google has come up with its personalized homepage ala My Yahoo or MSN. As Marc points out, its beginning of the next generation of portal wars. One thing I really liked about the Google portal is the ability to customize layout using drag-n-drop.

If you think about any of the "my *" services from Yahoo, MSN or Google, they are in the end content aggregators. How many sources of content a particular service can support might become a differentiating factor. Of course in an ideal world with open standards, it would become a non-issue. Another thing of interest is who will take the lead in allowing people to not just aggregate, but create content as well?

On one end we have blogger (which is really google), typepad, Ourmedia etc which allow you to publish content. On the other extreme are all these portal services which allow aggregation. When (and where) shall the two meet?

Posted by Gaurav at 08:01 AM | Permalink | Comments (0)

Ajaxian.com has a good round up of the recently concluded AJAX summit. I believe it was first of its kind and its a clear sign of things to come. I am always intrigued at how fast buzzwords become popular over the internet, and particulary over the blogosphere. The technology that enables AJAX has been around for years. But only now it is getting so much attention. I believe Google has had a big role in this. They have really pushed the envelope with their rich web applications and managed to do what was previously thought impossible.

Around December last year, Ashish and myself created our own first AJAX application. It was a blog indexing search engine (which we affectionately called Bloogle) with a Google Suggest
type interface. It was done over a period of 5-7 days solely to impress a prospective client. I have reasons to believe that our client was impressed :)

Since then, I have come across Laszlo and I feel it has the potential to become the preferred platform for developing Rich Internet Applications. Laszlo was created from the ground up for developing RIAs while AJAX is an innovative use of an existing technology (XMLHttp and Javascript). Microsoft's next gen windows client platform called Avalon also aims to enable rich internet as well as desktop applications. Interestingly Avalon uses a similar XML markup based approach for creating the user interface as Laszlo. It will be interesting to see if Laszlo community (Laszlo is open source) will try to create any migration tools to enable seamless transition to Avalon from Laszlo, once Longhorn is released.

Posted by Gaurav at 05:56 AM | Permalink | Comments (3)

As I was driving to work bleary eyed and half dazed, I had a brainwave. We talk so much about Social Networking and we are trying to build a variety of social networks. But the fact is that today's social networks are less social and more socialist. A social network treats all its members as equals. If Bill Gates and myself both signed up on Orkut, we would both be equal in priviliges. Sure, Bill will have 35000 friends more than me, and maybe a few thousand testimonials more as well (not all of them very kind I am sure :)), but other than that there won't be much to differentiate two people who in the real world have starkly different status in the society. So, essentially, social networks naively follow the socialist/idealist principle of "All men are equal". While that is great - and noble, if I may add, it makes a social network boring. And this is the reason why a Bill Gates or a Sachin Tendulkar will never join a social network. With one click of a mouse, they are reduced to the status of a commoner. Not to mention the barrage of invites, requets and spam they will attract.

On the other extreme are invite-only networks like aSmallWorld which caters to high net worth individuals. It is a safe haven for the rich and famous to communicate, and mingle in general. But by excluding the common folks, aSmallWorld also doesn't simulate the real world.

Of course, one could argue that why would one even want to simulate the real world, imperfect as it is. Well, I don't know. But it sure seems like an interesting idea to me. An anti-social network with all the class hierarchies and realities of the real world. Some would have more priviliges, more powers. Not all would be born equal, so to speak. It might have some niche application also in certain kind of networks where the class hierarchy is apparent. For example, in the world of business, the clients often have an upper hand and vendors clamor for business. In the job market, the employers often have the upper hand (unless its the IT industry - do you know DHTML? Please call me!). So who knows, there just might be a case for anti-social networks.

Posted by Gaurav at 03:49 AM | Permalink | Comments (1)