Telecom Travails

 Telecom Travails

There are few times in my life when I am at a loss for words and folks who know me will attest to the fact that those moments are exceedingly rare ! Talking about the state of the Indian Telecom industry is one such situation. But writing a blog requires words to be put on paper, so here goes ....

To say that the industry is in a mess is to state the obvious. If as a scientist I were to apply the principle of Occam's razor, the simplest explanation is  that Telco  execs  have strong suicidal tendencies and the state of the companies they run can be said to validate the thesis. But a lot of them are good friends of mine and are intelligent, capable and dedicated people. So while the thesis is not a bad one, evidence suggests the cause lies elsewhere !

As with all human  tragedies, the causes are manifold and as is  always the the case, the principal actors had the noblest of intentions. But you know what they say about the road to hell ...

In no particular order, let us look at the various aspects of telecom in India. The spectrum policy is as good a place as any to begin my diatribe.

1. Multi-spectral mayhem

We got it wrong across the spectrum ! One cannot even begin to count the number of ways we have got our spectrum policy wrong. But instead of doing a post mortem of the causes,  a fresh start with some clear principles as the founding axioms of a sane spectrum policy is a more constructive approach.

To begin with Airtel or JIO or Vodaphone do not consume the spectrum, the end users like you and me do. And since we use it, we should pay for it. No cross subsidy, period. A simple spectrum charge tagged on to call charges is the most logical option.

• This will  give  a clear indication of what spectrum costs. As with power charges or IT rates, spectrum charges can be progressive with a specified bottom rung subsidized. But all of this is transparent to the Telco, it simply collect the charges and transfer to the govt. Presumably the charges will be proportional to the calldata volume. The key is to keep the scheme simple
• This will also result in steady revenue to the govt without the associated hassles of spectrum auctions and AGR issues
• There is of course the possibility of political parties promising reduction of spectrum charges as part of their election promises but the imperatives of  fiscal discipline should mitigate this.Hope springs eternal ....

If spectrum is not auctioned, the question then arises as to how licenses are to be allotted.
While spectrum charges are outside the ambit of the telco in this scenario, companies can still bid for licenses based on revenue share. Learning from the AGR fiasco, the simplest scheme would be a simple percentage on the amount billed to the customer. Other schemes are possible , key is to have a scheme that is simple, not subject to interpretations and ideally the amount should be paid  whenever a bill is raised. 

2. The fallacy of a unified telco model

If the above scheme is taken to its logical conclusion, it will be evident that an ideal telecom network will decouple the front end radio network providers from the back-haul data link providers. Since all networks are packet based now, the distinction between voice and data is arbitrary, what makes sense is QoS for different types of traffic. Some of the QoS guarantees in 5G networks depend on cross layer coordination but that can be formalized to allow radio network separation. WiFi hot-spots are an example of how  network component separation would function. 

The advantage here is that radio operators can be much smaller players. The minimum size of what an auction of a radio network operator tender would entail can be figured out by trading off operational efficiency requirements and minimum number of operators required per area. 


We are not ready for such a scheme now but commoditization of telecom HW, open source stacks and spectral sharing schemes as proposed by IIT-Madras are all needed to make this a reality. Seamless interoperability , especially wrt roaming will be the key challenge.
But it can be done and is very much a solvable problem.

3. The ethos of Intellectual Commons

The above two points are more operational in nature and do not have any major questions about economic philosophy. The key philosophical issue that arises with any technology based infrastructure component that is key to modern society, is the ethos of intellectual commons. It is my position and it seems self evident that that technology foundations of society should be based on the philosophy of commons, hence the term intellectual commons. This naturally is the very anti-thesis of the notion of intellectual property. So what does this mean for telecom networks ? It means that at the very least, telecom networks should be based on open source components to the extent possible. Certainly it should be SW open source and COTS hardware ala Open Compute. But down the line it should evolve to use open source HW too.

The GNU Project, Apache Project and the Linux foundation have shown that the ethos of Intellectual Commons is very much feasible. Asking one of these entities or a new entity to maintain a production ready 5G stack is hardly a challenge. Intel now has a complete set of HW components and so have other vendors. Creating reference HW for standardizing the HW part of 5G infrastructure therefore does not present a problem. The glue components like on-chip, chip-to-chip, board-to-board interconnects are open standards. So are I/O, network and memory components. So no vendor can force the eco-system into a proprietary dead-end. It amuses me  no-end that Intel of all companies has contributed the AIB and CXL standards that makes an open HW market fully realizable. While I am a fan of Gen-Z , I am resigned to it being the  standard board and system level interconnect, with standards like RapidIO completing other parts of the standards zoo. But enough tech jargon, at a policy level it is suffice to know that we have all the standards we need for a vendor lock-free open HW Eco-system.

If the above vision comes to be true then the notion of a Telcom OEM would change dramatically. Service players like the Indian majors can now offer Telco infrastructure
as a service, as MS is already doing with Azure. Google is also making moves in this direction

4. Security

The above point solves the security issue that is bedeviling all the govts these days.
The security certified open source solution provided by multiple vendors sidesteps the whole "do we trust the vendor" issue. Cloud providers also should obviously provide only security certified open source solutions. Approval of vendors is a policy and not a tech issue but presumably govts will  favor local vendors for SW stack support. I prefer to maintain a studied silence on this issue !

India is making noises in this regard but I am a bit puzzled at the reluctance to mandate opens source solutions, either from the govt or by the telcos. Ia m equally puzzled at Silicon providers not pushing this idea since they stand to benefit the most due to value being moved to reference HW. Perhaps it is time for a non-profit  India Telecom Foundation to make all this a reality .

5. Costing of Services

One of the best kept secrets of the telecom industry is that the operators themselves are a bit hazy about what it costs to provide specific services. Telcos are fundamentally utilities. At the core they are bit pipes which provide transport for bit-streams at various QoS levels. It is key that they master the art of costing each service so that there is a clear picture of what it cost to provide a service and hence what should be charged to provide the service. They can provide OTT and other value added services but not as a bundled offering and preferably as a separate entity. Email and messaging are possible exceptions. Content owners definitely should not be part of telco ownership.

Ideally they should be no cross subsidy and service costs should reflect real costs. We are plagued with privacy issues in our email and social network services primarily because that industry could not convince users to go for a paid ad free model. It is time telcos gave us highly secure, ad free basic email and messaging service as part of the basic service offering. I do not think the incremental cost would be significant.