The convergence Bill is expected to be tabled in the budget session of Parliament. What is the ISP industry looking forward to?
Part of our challenge as an industry is that we have to deal with multiple agencies and multiple ministries-the department of telecommunication, the ministry of information technology, ministry of information & broadcasting etc. Ideally, for convergence to work, we need a convergence ministry, a single body or a window where all the inter-connected issues can get resolved. So, that is the starting point we would like to have.

What is the post-convergence scenario that you foresee, especially among ISPs?
Collaborative industry is what I see-I don't see very many companies trying to do everything everywhere. Companies will focus on strength areas both in terms of services and geography, then collaborate with others to make it a national service. I can't see any one company doing the entire last mile for the whole country. It is not just the question of cash but a of speed and the ability to do so before somebody else does it in the other markets. The word that will be used a lot is what they call cooptition- co-operate in one area and compete in the other.

In what sequence do you see convergence happening?
The immediate convergence that I see is that of voice and data. The Internet is focussed on three areas of communication-data, voice and entertainment. For ISPs also voice and data are the first two priorities and entertainment is the third. This is because broadband pipes and bandwidth are still a little distant, whereas voice and data can be combined in a single pipe on the existing infrastructure. This will benefit the customers a lot in terms of price as well as service.

What is the viability of an ISP, who has the requisite infrastructure?
I don't think there are very many who are well-equipped to carry on. May be regionally yes, but not nationally. I think the idea of revenue sharing on any service is the right way to go about it. What that revenue share should be is the next challenge.

What model do you suggest?
It would be nice if it was 100-0, but it should not be more than 5 per cent. With the current suggestion of about 15 per cent to the government and the rest to the operator-may be it could move to that as the market develops and matures-you do not get enough reserves created to put back into the business. I don't know how many companies are fully equipped to take advantage of the opportunity straight away. It needs a fair amount of investment.

A lot of shake-out and consolidation is expected in the ISP industry. How will it happen?
The opportunity for consolidation is there but is not very viable, with the revenue per subscriber being very low today and the subscriber-acquisition costs very high. In the last deal that was done in the US, $250 to $350 per subscriber was paid, this would mean around Rs 9 crore if you take over somebody with 5,000 customers. In half that amount, I can finish a `C' category ISP through marketing. So, ISPs will consolidate if there is a customer stickiness i.e., if a customer is tied to a service provider contractually for a few years and his ability to migrate to another service provider is difficult. If that is not the case all you will be buying is hard infrastructure and not customers.

Is Sify thinking in terms of buying out smaller players?
Depends on the price and what we are getting for it. For somebody like us who are present in most (220) cities and towns in the country, it does not make sense unless it comes at a very attractive price.

How comparable are Indian bandwidth prices with international prices?
They are four to five times more expensive.

Even after VSNL has dropped bandwidth prices?
Yes, if you compare with trans-atlantic and competitive Asian markets, like Hong Kong, in terms of absolute price, the quality of services and the time to get a link-up. While the quality of service is `best efforts' in most cases, there is no committed timetable when you would get a link-up. This is because of the monopoly status. Once competition comes in, all this will be corrected.

When do you see yourselves completely independent of VSNL?
Waiting. Waiting. I think it will take time as they have a monopoly status till 2003. I am told that there is some negotiation to advance that, but I don't see anybody physically building the infrastructure and bringing alternative bandwidth to this country through fibre at least for the next two years.

Estimates of bandwidth requirement in the country are very confusing. While Nasscom predicts it at 300 gbps by 2005, Frost & Sullivan say it is not more than 35 gbps by 2004. What is the real requirement according to you, and is the supply likely to outstrip demand?
I think it's closer to the Nasscom figures-they are looking at it from an opportunity point of view, rather than from where we are today and taking incremental steps. Today, it takes a long time to download even text files on dial-up mode and one ends up going to a browsing centre which has ISDN lines for speedier access. However, if this can be done faster at home, you will start downloading audio, video and anything else that interests to you. Which means people's personal usage of bandwidth will go up exponentially. For instance, we managed when our PCs had no memory, but today, with 64 megabytes of memory we feel the machine is too slow. We used to have 20 megabytes hard disk, now we have 4-10-20 gigabytes and we still run out of space. We are restricting ourselves today because of lack of bandwidth.

Other than personal usage, which other areas do you see increasing their usage of bandwidth?
IT-enabled services like medical transcription, for one. For instance, think of a doctor in India who can see a CT scan of somebody sitting in the US and write the report here because he's able to see it with the same clarity because of broadband. There is very good opportunity for this and other such services.

But with superior technology, more material can be compressed in the same pipe? Will that not reduce the demand for bandwidth?
I agree. There will be compression for the same application. But what we are looking at is newer usages.

Like?
Movies, music, video-based news, apart from remote health services between rural and urban areas, like what Apollo Hospital talks about with 64 kbps. Imagine if they had unlimited bandwidth. The possibilities are endless.