The problem is they keep appointing people to run Telkom that don’t seem to know how the comms industry works. Look at AkC when he got to CellC – first thing he did was lower prices and try and increase the customer base. That’s the right thing. Telkom have wasted their gigantic war-chest and their advantage of being first in the market, where they should have been pursuing a single aim relentlessly: getting more customers on board and lowering fixed line costs. It’s not too late for them to do that, but they don’t have many more chances to do it right, before it is too late.

The last stat I saw, Telkom had around 4mil fixed line (voice) users. The infrastructure is mostly there. They need to get all those people on to ADSL. Give them a dirt cheap 128kbit or 256kbit uncapped product, like they have in India, for example.

“And uncapped is a necessary precursor to video on demand and other “over
the top” services where operators like Telkom are going to have to
compete.”

In markets that have popular services (e.g. Netflix), video-on-demand and over-the-top applications are actually bringing an end to uncapped offerings. In the U.S., most broadband providers have been introducing tiered pricing, or limiting video-on-demand on their uncapped packages.

“Given the scale of the investment involved, and the time it will take to
complete, it’s worth asking whether Telkom would have been better
advised to be more aggressive in future-proofing its network by being
more aggressive in rolling out high-speed fibre directly into homes.”

I think you need to do a bit more research on the MSAN roll-out and other aspects of Telkom’s network upgrades. Fibre-to-the-home is still quite costly, mostly due to the per-premises capex for the actual fibre from the street to the home (in the region of R20 000). Telkom’s network upgrades are, however, bringing fibre much *closer* to the home than before.

“Is it really worth pouring billions more into a decrepit copper network?”

Actually, the current problem with fixed-line is that it doesn’t make financial sense to invest in the copper network (especially as the indecision of the regulator means no-one can do a return-on-investment calculation), and the quality of existing copper (that which isn’t replaced because it was stolen) is quite poor, so fewer customer get the speeds they should be able to get (based purely on distance from exchange/DSLAM/MSAN).

“Telkom needs to be playing a long-term game, and that means simply fibre, fibre and more fibre.”

Who has the most fibre?

” Fibre-to-the-home is still quite costly, mostly due to the per-premises capex for the actual fibre from the street to the home (in the region of R20 000). ”

I’ve seen this written so much in ZA forums, and tried to do some research but can’t find much reliable information – do you have an idea of cost breakdowns? Usually when you ask someone to explain this figure they clam up. I based my research on Verizon’s FiOS architecture, and the road->home components there are the optical splitter ($5-$10), cable (dirt cheap, probably cheaper than copper), and the passive optical router to break out into Ethernet or wifi ($25-$50). Of course, this is a simplistic view, but much of the rest of the installation/wiring is the same as installing copper (ie. strung overhead).

I agree completely with you that the MSAN route Telkom is taking is the right one for them. There’s nothing wrong with the quality of the copper lines themselves, it’s the aging equipment and connections in between which is the problem. Upgrade that, and the last mile will come alive again.

Correct me if I’m wrong, but that’s the total cost per-customer in the network build, including all the back-haul and infrastructure? They’re usually pretty vague when they put these figures out. If that’s the case, then the majority of that cost would be the same on any other network, and it’s safe to assume the actual last mile portion is a fraction of that. Google is charging $300 to connect to their Google Fiber project, which is more in the ballpark of what Link Africa are saying, if my interpretation is correct.

I think there is a super simple solution to the fibre question: Just roll out a pilot project over two months and measure it up. I will take a bet that the project will show that fibre is viable and needs a bit of guts and ‘go’ to start the fibre ball on its way. Scores of small guys could be employed in this industry too.
By the way, where I am in Bedfordview there must be hundreds of kilometers of fibre under the roads…so just when the hell is it going to be SWITCHED ON?????????

I for one would love to have ADSL but even after all these years there are no exchanges for me. I stay in a suburban setting and my National provider can not supply me. What does that say!

Telkom doesn’t have any line connections in the Fouways area I live in. Been hassling them since 2005 to get a line in so I can use their fixed line…2013 and still waiting. There are many more individuals that would sign up if the idiots at Telkom would catch a damn wake up.
Mobile broadband is so damn unreliable at best the max download speed I’ve managed to achieve is not even close to what the mobile operators claim the speed should be…I track downloads at kb, as in maybe 100 kbs on really good days. Yes, really.