Victus Spiritus


Iter8r: Organic Crowd Shared Startup

30 Sep 2009

Forget What You Know About Startups

This past week I have had the pleasure of learning about by emailing back and forth with the founders Pandu Rao and Harish Kulkarni based out of Redmond, Washington. It all started off with this email from Pandu Rao:

I read some of your articles with interest:
Open Social Media Of the People, By the People, For the People
The Creation of Value
Early Exclusivity Fosters Later Empires
I think the ideas mentioned in those posts can be put into practice. My co-founder, Harish and I have worked to bring some of them to fruition. Its called Iter8r (

Iter8r is a social news/bookmarking website that enables users to:
1. Post interesting links
2. Vote on them
3. Share (or donate) the profits
The part about sharing news and voting on them is pretty straight forward, but the way Web 2.0 companies engage with their users can be improved. The objective of our company is to treat users more equitably (that's it).

Iter8r's salient points are:
1. Full financial disclosure
2. Transparent and inclusive corporate governance
3. Conversion to non-loss cooperative
4. Full ownership and operation by the community
5. Commitment to Free Software and Free Culture

Its described here in more detail:
If you have any questions or comments, I will be happy to address them.
If you would like an invitation, I would be more than happy to send you one.
Thank you for your time.

Best regards,

Once again I’ve been pleasantly surprised by discovering some topics I’m passionate about are already in development. Not only are these ideas shared by other folks, but they had the energy and vision to actually launch a start-up and a functional website.

After a burst of emails back and forth I had both a beta account and a great feeling about the company. What Pandu and Harish are working to achieve is a self sustaining profit sharing web community. Their model is similar to but with plans for their 32% ownership transferring to the community over 8 years following a "trigger event" of 80,000 community members (here's a self plug to my Squidoo lenses).

Harish has been the heavy for web development and coding thus far. Go Harish! I'm growing great respect for those with the skills and will to build functioning and useful web apps.  Pandu pitches in on the software end,  while working the business, marketing, and writing side of things. Truly they have an inspired sense of business innovation, and as you'll see it's infectious. They are prepared to open source (GPL V3 license) not only their web software, but their framework.

*Clarification from Pandu*, they will place under Free Software Licenses all the software they have writen (and will write in future) for Iter8r. And not just that which suits them.

Question and Answer with the Founders

After reading the Iter8r about page and using the software here are some of my specific suggestions/questions and Pandu's and Harish's answers:

Me: How is Iter8r developed and using what platform?

Pandu: As for development and platform, we use Linux, Python, Django, Postgres and Pinax. That set provides the bulk of the components and libraries we need.

Everything that is Iter8r-specific has been developed by Harish (to a large extent) and myself (to a smaller extent).

Also, the web-programming part has turned out to be rather straightforward.

As for programming help, we'll take all we can get.

Here's a short list of things (in no particular order) that we can use help with:
1. Derive practical ranges for the parameters governing the profit distribution engine
2. Security analysis of our web-site code
3. Security analysis of our data-security applications (krypton and kryptonite)
4. Improve the commenting component
5. A formal game theory analysis of Iter8r profit sharing

Me:  It looks like we need subject areas as users. Are there sub communities like subreddits?

Harish: We debated between reddit-style categories and tags and opted for tags because it gave us greater flexibility to adapt to different usage patterns.

Other things belong to the work items list below include improving the tagging module backend and rethinking and redesigning the tagging interface.

Pandu has already addressed the other questions in this thread. We will respond to the other questions in the relevant threads.

Once again, thanks for reaching out and thanks for your support.

Me: How will post moderation work within your crowd sourced linking site?

Pandu:  On moderation - This part will almost completely defined by the community that Iter8r attracts. We'd also like to see how far we get with a self-governed community.

We have to consider how this will work on a large scale:
1. We can have category/tag curators
2. We can have moderators (even rotated periodically)

Its safe to say that we'll adopt a mechanism that's simple and equitable.

Me: Might I suggest a stipend for board members as we want the best people pushing the concept forward in the future?

Pandu:  The stipend idea is a very good one! In fact, our model does take into account the provision of a reasonable stipend for the board members. This is subject to us generating revenue, of course :-)

Me: Also, I know you are working to gather open source support. But will it possible to get funding to hire additional programming power? I have a cyber security expert friend who is looking for work. If there could be a position down the road (or profit sharing) he may be able to provide excellent guidance on the security of the site/linux server.

Pandu: In our opinion, traditional VC Funding or even angel funding may be very hard (if not downright impossible) to obtain for something like Iter8r.

We have, however, designed it to be self sufficient when it generates ad revenue (We are nothing if not optimistic :-))

This does not preclude us from other sources of funding (e.g grants). We were pleasantly surprised to hear of the Knight News Challenge Grant (this a few days before we launched).

You can learn more about it here:

We entered the challenge and have requested a grant to ensure that Iter8r achieves it full potential sooner. But grant or no grant, we'll get there.

Iter8r's entry is here:

I must point out that our vision for Iter8r is not subject to us receiving the grant. In fact we entered the contest to receive user feedback to refine the idea. We wish we had received more feedback from the users visiting News Challenge.

As for your friend, we plan to bring developers and system administrators on board (in paid capacity). But that's a little further in time.

However, I will categorically state this- users and developers who take a contributing interest in Iter8r will be nominated for board elections. More on this later.

I even got some great feedback on my own project:

Me: In a small project me and a few friends share interest in, I ran into some challenges setting up user profiles and a good database (besides an attractive front end). If you are curious that concept is a contextual search/advertising tool driven by social media. I setup a Twitter extracting example at (very rough idea). I wrote up the concept here: Notional Framework for Monetization Web2010 and the post it refers to Monetization Web2010

Pandu: I went through the links you mentioned. If I understand you correctly, you are suggesting a mechanism (central or otherwise) to provide context for a link. This context is distinct from the data that the link points to. And that this context can carry the history associated with a link (e.g who forwarded it, who forwarded it as an affiliate). Its a way to annotate the relationship between users and links they forward.

One way that I envision this being done (and maybe I am wrong here) is  to do half of what a url shortener does. i.e do not resolve completely, but resolve to a page that shows the annotation associated with a link.

I also think that there are a few challenges associated with adverts for microblogs/tweets. e.g Not enough context to determine appropriate ads. I have some ideas on how this can be overcome. But that's for another time.

Can We Use Their Framework?

The answer is a resounding YES! from Pandu and Harish. My personal curiosity has lead me to explore if their design can be applied to a crowd shared programmer model. I gladly mention that this idea is inspired by David Semeria (@hymanroth on twitter) and some of the ideas he has hinted at with his LM framework which I'm eager to see in action.

The idea is that contributors can write source code to support a very high level framework. This framework could have simple to connect modules, such that domain experts (who may or may not be coders) can use the components to build new aggregate (even higher level) applications. In this way coders can have outside folks use their modules in innovative ways, potentially generating revenue somewhere downstream.  Then profit is split with other developers who support the final money generating web service (the module builders and the domain experts with the vision to put it all together). The key points being profit splitting and fair attribution, and both Pandu and Harish are keenly aware of these issues with their business design.

More to come on the background of Iter8r after I have a chance to talk with the guys.