Technology » Internet

Updated: February 25, 2010 18:03 IST

Facilitate hosting content locally

D. Murali
print   ·   T  T  
Shivkumar Jagannath, CTO, Zylog Systems (India) Ltd.
Special Arrangement
Shivkumar Jagannath, CTO, Zylog Systems (India) Ltd.

Do people lose sleep, worrying about wireless connectivity (such as that it can be insecure etc.)? Shivkumar Jagannath, CTO, Wi5, Zylog Systems (India) Ltd, Chennai ( concedes that there are concerns regarding wireless connectivity in the minds of people; but the good news, he adds, is that the apprehensions are all addressable.

“In fact, much of the research and development in wireless networks is in the realm of security,” says Shivkumar, during the course of recent interaction with Business Line. Newer standards for encryption of the client to base station connectivity are being developed and companies are adopting multi-tier security to ensure that their networks are more secure than ever, he informs.

“The Government of India has also instituted measures to ensure that wireless networks are not misused by anti-social elements. There is a directive on how customer verification, authentication, legal intercept and other such activities need to be undertaken by ISPs (Internet service providers) and telcos (telecom companies) to prevent misuse of the Internet using wireless access.”

Excerpts from the interview.

What do you see as the major challenges to the growth of connectivity in the country?

In India today, we have more than 200, 000 route km of fibre optic cable laid by different agencies/ companies such as the BSNL, Railtel, Reliance, Airtel, Aircel etc. This indicates that there is a very robust and high-speed backbone available for most part of the country.

The flip side to this is that there is not a very good last-mile delivery infrastructure for optimally utilising this fibre. The largest outdoor plant of copper is owned by BSNL and MTNL and the Government in its own wisdom has not allowed “unbundling of the local loop” which means that this copper can only be used by BSNL and not by other ISPs to deliver high-speed/ broadband Internet to customers.

The only option is for ISPs to lay their own copper if they choose to deliver broadband using wireline technologies. This is a very expensive and cumbersome process and is not viable unless the ISP also offers basic telephony services as in the case of Airtel. This is one of the primary challenges to the growth of connectivity in the country.

The second challenge is that the Government has not allowed ISPs to benefit from any customs/ import duties that would allow them to pass on the benefits to the customers; thus the high capital cost combined with very high import duties makes it unviable to offer broadband Internet access services below a certain price point.

And, going forward, what will be the key drivers of wireless connectivity?

For the reasons mentioned in the previous question, the only technology that proves to be viable, scalable and can be deployed in very little time is wireless. This could include Wi-Fi, WiMax, 3G (HSDPA/HSUPA and EVDO).

Most companies have come to this conclusion and have identified their technology of choice. This choice of technology stems out of the existing models that they already have in place.

For example, a GSM or CDMA-based telco would naturally gravitate towards 3G since that is the natural evolution of the technology for data connectivity over mobile networks. Hence it is felt that higher speed connectivity can be offered to mobile customers at an incremental cost towards upgradation of the base stations.

Other companies, which are pure play ISPs are looking at WiMax and Wi-Fi as the technologies for delivering high-speed connectivity. The reason for this is that these ISPs have no prior asset in terms of copper or cellular base stations. The other reason is that ISPs tend to build pure data-only networks against voice-only networks with an overlay of data as in the case of cellular operators.

One of the key drivers for wireless connectivity is the proliferation of wireless chips in almost all the new devices coming out in the market. This is dominated by Wi-Fi and 3G (for phones), Wi-Fi for laptops/ netbooks and WiMax in some niche markets.

The availability of portable computing devices with embedded end-equipment makes the owners hungry for speed connections to enable them to extract value from their devices. This has led to most operators and ISPs rolling out wireless networks in a big way.

You speak of having the content locally. Why is that important?

It is widely believed and an established theory now that the number of Internet subscribers in a country has a direct bearing on the economic development. We have seen thus far that the US is the dominant destination of interest for the simple reason that most of the content is hosted in the US.

This makes the cost of Internet access directly dependent on the size of the pipe a company has to the US. Hence the term “international bandwidth” against “local loop”, which is the term for the bandwidth inside the country. The major telcos such as the BSNL, Railtel, Reliance, and Airtel have laid more than 1,30, 000 route km of optical fibre in the country with a capacity exceeding terabits (millions of gigabits).

The urgent requirement is to build high-capacity data centres around the country connected to these fat pipes with all ISPs peering at Internet Exchanges like NIXI so that content can be hosted locally. When this starts happening, most of the traffic will originate and terminate within the country thus leading to more efficient usage of bandwidth and thus bring the cost of Internet access to commodity levels.

Do CTOs of enterprises pay adequate attention to communications?

I believe that CTOs (chief technology officers) of enterprises are driven by business needs rather than technological innovation. I have seen that most of them leave the thinking/ planning/ selection of technology to minions rather than dirty their hands themselves. They would rather tread well-beaten paths than try and discover newer ways to improve productivity.

Can city-wide WiFi capability obviate the need for data cards?

The two technologies serve different segments of the market. City-wide Wi-Fi networks definitely would bring high-speed broadband Internet access to more people than ever. The data card is more suited for professionals travelling to various places around the country outside the cities. This segment is driven more by the need for any type of connectivity rather than high-speed connectivity.

It is fair to say that the two are complementary in and we have already seen that in the US most cellular operators are now pushing data traffic off 2G/2.5G/3G networks onto Wi-Fi networks in dense environments essentially because of the huge drain on expensive resources including spectrum, capacity etc.

Your views on the impact that 3G and VoIP can have on the telecom market.

3G and VoIP (voice over Internet Protocol) are exciting technologies that have a potential for game-changing in two main areas.

3G brings broadband speeds to handheld devices which themselves are becoming more and more powerful. Operators have realised that with declining ARPU (average revenue per user) based on voice alone, they need to create powerful differentiators for their subscribers to remain loyal. This platform opens up a multimedia delivery platform including video and data.

So people will be able to switch from pure voice calls to video calls which inherently have more value. Also, high-speed data availability opens up other opportunities such as powerful hungry applications running on the phones.

One such application that comes to mind is augmented reality (AR), which is somewhat of an amalgamation between mapping/ geo-tagging/ photo blogging/ location-based services that enables the user to navigate and find information in real-time at the location that he is currently at.

VoIP is another game-changer we have seen evolving over the years. Now that everyone has moved into SIP (session initiation protocol) based platforms as against H323, we have already seen that the cost of making long-distance calls has come down by as much as 90 per cent in some cases.

It is widely believed that local calls can be free with VoIP. Most countries tend to treat VoIP as a threat rather than an opportunity for the sole reason of protecting the strong long-distance calling lobby of incumbent telcos. They believe (somewhat incorrectly) that there would be a loss in revenue if VoIP was made freely available. This argument does not hold true as can be seen in the US where Vonage has around 2 million customers compared to an AT&T or T-Mobile with more than 100 million customers.

Any other points of interest.

It is now an accepted fact that the measures adopted by the Government in the years leading up to 2007 (the year of broadband) haven’t succeeded in increasing broadband penetration levels to beyond 6.5 million today in 2010. Some of the measures/ policy changes that could be undertaken are as follows:

1) Unbundling of last mile: BSNL and MTNL between them have the largest copper last mile deployment in the world today. This outdoor plant has been built over a period of more than 20 years. All over the world, this copper last mile is unbundled which means that any ISP can rent out the last mile from the telco and install their DSLAMs (digital subscriber line access multiplexers) and CPEs (customer premises equipments) at the exchange and customer premise and deliver Internet access.

This has not been allowed by the Government of India due to various reasons primary amongst them being protectionism. As a result, the pace at which these companies themselves are able to roll out broadband services is too low to cater to the latent demand; and due to pricing, quality of service etc., the customer uptake is also not too good.

It is believed that only 10 per cent of the existing copper last mile is now converted to broadband customers. The Government should seriously reconsider and “unbundle the last mile”. It is also ironic that BSNL is entering into franchising agreements for rolling out its WiMAx networks by allowing the franchisees to ride on BSNL’s WiMAX spectrum; which in essence translates into the same thing, with the difference being that these franchisees themselves cannot be ISPs.

2) Allocation of 700 Mhz spectrum for broadband wireless access (BWA): It is known that the lower the frequency of operation, the more efficient is the propagation thus translating into a wider area of coverage. Most countries such as the US, Australia and Canada that have a large landmass with low population density have discovered that BWA in the 700 Mhz is a cost-effective way of delivering broadband access to the rural areas which are hugely under-served areas.

In fact, the FCC (Federal Communications Commission) in the US has asked terrestrial broadcasters to vacate the 700MHz spectrum to make way for BWA. India too can follow a similar policy given our large landmass and distribution of population. This will accelerate broadband penetration throughout the country.

3) Easing of duty norms for ISPs: At present all telecom equipment is manufactured outside the country and needs to be imported in large numbers. The high level of import duties levied translates into an increased capex for ISPs thus preventing them from offering low rates of Internet access to retail customers. Also, due to high service tax levied at multiple levels – by the tier-1 provider for bandwidth, and by the ISP for Internet access – the price of the end product goes up by more than 20 per cent.


Pentagon issues new policy on social networkingFebruary 27, 2010





Recent Article in Internet

Citizen Lab, based at the University of Toronto, said UC Browser had more than 500 million registered users and was the most popular web browser in China and India.

Alibaba-owned UCWeb accused of leaking user information

"Chinese and English-language versions of the browser made easily available to third parties personally identifiable information" »