Mathew Carley Talks On Hayai Broadband

From there, I researched for about 3 months EVERYTHING I could find regarding operating a telco in India (and I still know only a fraction of the overall situation) – legal issues, last-mile, cablewalas, regulations, existing networks, the topography of the city including areas of interest and eventually I was able to cobble together a rough spreadsheet detailing how much *I* might be able to sell the same product for – and even with a wired service, it was less than Reliance, so I decided to pursue the idea.

Q. Tell us about Hayai India (company ownership, licenses, operation areas, infrastructure, tariff plans )

Company ownership:

Myself and 3 other directors, 4 primary shareholders and some other details I can’t make public at this time with regards to how we work with our upstream providers, licenses and funding. Because of the regulations, I can’t be CTO of the company, even if I wanted to be (must be an Indian citizen). I’m not much of a desk-jockey, so CEO and CFO roles are going to others, and so I’ve put myself in the COO position, though naturally, I still call the shots.

Operation areas:
Once the launch has happened in Mumbai and that is all actually working, there is strong demand for us in pretty much every major centre – but in addition to that, we would focus a lot on building to areas that are underserved and/or have currently got little competition. While the idea started out as being a Broadband operator, we will be including Voice and IPTV services of course, so ideally what will happen is that we’ll end up offering triple-play services to take advantage of the network we’re putting in.

Infrastructure:
The idea has been, from scratch, to build fiber all the way to the user. Since everything is all-new, the capital expenditure as compared to DSL isn’t much different once all things are taken in to account, but most importantly, fiber allows us upgrade paths which copper does not: VDSL does up to 100mbit/s through extremely short amounts of copper – and even though there are some new DSL technologies coming out which are giving around 600-700mbit/s, we can already push 1Gbit/s over fiber with potentially even 10Gbit/s later on if we want to.

Considering the costs of digging, we are trying to come up with new ways to get fiber to people more cheaply – in NZ we’ve just started exploring the possibility of using the sewers, since, we can lay fiber without any interruption to things that are happening above ground, all thanks to a few specialized robots. They’re using this technique in Italy, Japan, France, Scandinavia, Australia, UK and the USA that I know of, and I’m hoping we’ll be able to use it in all urban areas (even in India) because I believe it will substantially reduce the cost of building the network.

Tariff plans:
Our aim is to be the fastest ISP in India – this is probably what I would call “the easy part”. Doing that while maintaining a reasonable price – that’s a bit more tricky. Tariffs are changing everywhere, so naturally we’re keeping an eye on all of our competitors – large and small – to ensure that we can respond to their offerings and maintain our competitive edge. Keeping this in mind, we haven’t launched yet in India and as such I’ve recently taken the proposed tariff schemes offline until we are properly ready to launch and we can set the prices in stone, but part of my aim for this continues to be that we should like to see European Broadband pricing in India: we’re almost there, but not quite – for reasons I’ll explain in the final question.

Mathew Carley

If I had a chance to travel back in time and speak to myself 2 years ago, I would probably tell myself I was insane, as there seem to be enough regulations surrounding the running of a telco/ISP in India for Indian citizens, let-alone a foreign citizen.

It hasn’t been the most fun experience of my life, and as much as I hate to say it, I probably would have had an easier time of starting in Pakistan for a variety of reasons: company formation is quicker and easier, ISP license acquisition is quicker and easier, the local loop is unbundled and so on.

Q. Tell us about Hayai Broadband in other countries.

Hayai NZ was formed in August 2010 after I returned to Newzealand to get my Indian paperwork done.

The reason I’m tacking NZ as well is because, in NZ, everyone uses the same last-mile loop which is leased from Telecom. Everyone’s pricing is basically the same, the speeds are generally all the same and there is not much differentiation between providers. Now that unbundling has started happening, some providers are building their own last-mile, but even then it’s still basically an extension of the existing infrastructure.

There is limited work going on in the fiber arena, although now the NZ government is going to be building a nationwide FTTH network over the next 10 years – we expect to be using and becoming a part of creating this new infrastructure, but we are still discussing with the consortium building the government network if and how we can go about getting services ready NOW, since the government project won’t start until next year.

Hayai NZ may also get in to mobile if it can, because the situation with mobile services (voice and data) is that, even if the quality is good enough, mobile tariffs in NZ are extremely expensive.

As for Hayai Pacific, this has only been incepted in the last month due to my discussions with the Ministry of Foreign Affairs and Trade in Vanuatu – but we’re hoping to move in to that region as well. Eventually, we expect to become a worldwide company, but I expect to be focusing on “the other 3 billion”. While there are a few companies focusing on these markets in different parts of the world (including but not limited to Digicel, i3 Networks, LIME, and what used to be Zain), based on preliminary research, the services are not as good as what we see in (for example) Europe – so our USP then is to bring the type of services you’d find in Europe to these people for the same sort of pricing that existing players are offering – or less.

Q. What is the expected date for Hayai launch? Tell us rollout plan in other areas after Mumbai.

At this point, I have to get a few approvals (financial, security etc) so all I’m waiting for is to return to India: the plan then is to give India about 17 hours notice of the launch (well, it might actually be more like a week, since I do want to get a launch party organized).

Once Mumbai is up and running, we’re expecting to expand quickly in to other centres in Maharashtra and other states within quite a short timeframe (funding permitting, of course: this stuff ain’t exactly cheap!)

Q. What’s your take on broadband scenario of India, present and future?

India’s Broadband future is extremely bright – if only we can remove the barriers which are, in my view, holding it back.

Aside from regulatory issues, everyone seems to fight with everyone else, and it’s not in their best interests nor is it in the best interests of the consumer. Two major glaring things that need to be fixed: LLU (or at least agreements to use each other’s infrastructure) and Peering.

LLU (Local Loop Unbundling):

Even though we don’t really plan to use it, we would push for LLU in India because we think it would greatly improve the way things operate: no more cablewala issues and no more wondering if I can get service based on my address (the situation I had was, even though I wanted Airtel, and I could see Airtel fiber under the footpath less than 100m from my window, only Tata services my building).

Peering:

I’m a big fan of peering. There can never be enough peering, but despite NIXI having been set up and operating, it’s terribly inefficient due to the stupid tariff structure: whoever decided that peering should be charged per GB should be taken out in to the street and have very bad things done to him. It is literally one of the worst ideas I’ve ever seen, because you simply can’t expect to maintain a healthy domestic peering ecosystem if the per GB charge is double that of my equivalent charges for International bandwidth. My bean-counters would surely prefer I sent the traffic to Singapore and back because although it’s less efficient and worse, it’s cheaper

In addition to this, you pay up to 3 lakhs per year for a 1Gbit/s port, but frankly, I’d be more than happy to pay 3 lakhs per month if I didn’t have to pay per GB, because if you do the math, at a 3:1 in:out ratio, that could cost me as much as 44 lakhs per month. In NZ, I pay less than NZ$500 (INR15k) per month – including transport – for 1Gbit/s with no traffic charges, and in Finland, peering is 250EUR/month for 1Gbit/s, or 1050EUR/month for 10Gbit/s including the membership fees.

My proposal to fix these problems:
Even if LLU doesn’t happen, my expectation is that we will open our fiber to other providers so that they can lease our last-mile fiber to provide higher speeds to customers, and/or perhaps form a bi-lateral agreement so that we can use their last-mile copper: perhaps as a “lite +” offering if we were to offer some plans based on DSL.

Because of the tariff structure, as far as we’re concerned, NIXI is a joke, and unless that tariff structure changes, broadband in India will continue to suffer. I’ve put the proposal out that we should form our own peering infrastructure and charge sane fees for interconnection – even something like Rs30k per month per 1Gbit/s port with no traffic charges: under a full load, this works out to under 10paise per GB – standard IX rules apply (in other words, you can’t send your traffic to networks not connected to the IX through the IX) and so forth.

There is a small organization in Sweden which is comprised of some NRIs which is trying to start up an alternative to NIXI, but they haven’t got very far yet, so we’re hoping that we’ll be able to work together to achieve this, because, peering will help in several ways. First of all, if it’s done at a reasonable cost as we’re proposing, that traffic doesn’t cost the ISP much. In essense, any ISP who chooses to peer with Hayai would more-or-less be included in the Hayai zone, and his customers would see (to some extent) similar benefits to what our customers do with the “Hayai Zone“.

Downloading from our caches and mirrors would be fast and inexpensive which reduces their overall expenditure on international bandwidth, transferring information between our customers and theirs would be faster (and again, cheaper), and it would allow new services to arise in India that currently couldn’t work very well – from cloud-based services to game servers (lower pings!) to locally hosted video streaming sites – and if some media company joins in, then that media company could offer a library of TV channels and/or movies for legal viewing at very low cost – a-la Netflix.

Of course, the challenge would be up to the millions of Indian entrepreneurs both in and out of India to come up with services to take advantage of this new eco-system, and I expect that we would also be able to bring a lot of India-central websites currently hosted in the USA back to India. Data centres in India are already technologically sufficient, they’re just limited by bandwidth and the fact that they’re quite expensive by comparison. Our hosting partner, E2E Networks (based in Delhi) is already trying to do just that.