Village Telco Business Model

[Steve Song]

Happy New Year all.

It is possible that the lengthiness of my previous message may have
dissuaded anyone from opening the attached spreadsheet which is our
current best guess for the revenue model for the Village Telco. I'd
be grateful if anyone were willing to take a look, comment, ask
questions, suggest flaws, additions, deletions, reframings,
suggestions for making it more readable,intuitive.... you get the
idea.

Many thanks in advance.... Steve

[McTim]

Hi Steve,

Sorry, I had missed this the first time around!

Where did you get the revenue numbers? How will you terminate calls to
the POTS? Have you gotten the pricing from an operator?

--
Cheers,

McTim
http://stateoftheinternetin.ug

[Steve Song]

The revenue numbers are a thumbsuck although not a completely
uninformed one. USD7 per subscriber per month outgoing and 25% of
that incoming. I can't remember who proposed that figure but I think
it was Rael or Andrew Fletcher who has worked with Rael on Dabba's
business model. Can Rael or anyone else who was there comment?

In terms of call termination, we are assuming that this model is for a
start-up village telco that would connect over IP to a national
upstream VOIP provider like Dabba for instance in SA . Dabba have
already negotiated interconnects with the fixed and mobile operators
in South Africa. We have assumed that a Village Telco in SA would
match the prices that Dabba is charging in its network, effectively
becoming part of that network.

The model doesn't actually specify what the call termination costs
are, only how much people are likely to spend and what percentage of
markup is required. Perhaps this is a mistake?

-Steve

2009/1/8 McTim <dogw...@gmail.com>:

[Antoine van Gelder]

Dear Steve,

A few years ago I did a rough modeling exercise focusing on traffic
and bandwidth scaling - out of curiosity I spent this morning updating
it using your model as a starting point to reflect a Telco rather than
ISP focus.

I hope you find it to be useful.

- antoine

PS
. Bold Italic Green numbers are the input variables into the model
. The financials in my model are really quite a thumbsuck so you
probably do not want to pay too much attention to them beyond a rough
indication of whether the numbers are going up or down!

[Steve Song]

Many thanks Antoine! Processing... back to you shortly.

-Steve

> ------------------------------------------------------------------------
>
>
>
>
> --
> http://7degrees.co.za
> "Libré software for human education."
>

--
Steve Song
Telecommunications Fellow, Shuttleworth Foundation

email: steve...@shuttleworthfoundation.org
work: +27 21 970 1220
mobile: +27 83 482 2088
skype: steve_l_song
blog: http://manypossibilities.net

[rael lissoos]

Steve Antoine

I have looked over plan I will try post comment on our practical
experience soon

Rael

[Rich Bodo]

Since I'm new to the list, I thought maybe I should ask some more
basic questions before getting too deep into the numbers.

I examined the spreadsheet. I would like to cover some of the
opportunities and risks, so I'll start with what a newcomers
impression of a "village telco" business model might as well as asking
about some of the risks. If you think these are interesting, but it's
more appropriate to discuss documents on the wiki, I can start some
work in progress pages there and we can cover individual sections in
their own threads.

VT1 Business Plan
-----------------------------

Summary - product and market
--------------

VT1 is a telecommunications service provider that will receive
revenues from the following service package offered on a contract
basis:

* Internet access - A percentage of total available VT1 bandwidth will
be delivered to users for a flat monthly fee.

* Long Distance phone service - Phone service from the local phone
company will be provided at a fixed markup over cost.

Lease of termination equipment to be included in the price of a
one-year contract.

Customer will recieve a single monthly bill for internet and phone service.

Phone service and bandwidth-limited network connectivity between
village telco nodes will be thrown in with each contract.

Internet phone service such as Skype is available in the region of
VT1, and can be utilized by Vt1 customers at no extra charge.

Market consists of N businesses and homes in a 50km radius of the
proposed VT office.

Y% of the market currently pay Z% more for telecommunications services
than they will when they switch to VT1.

B% of the market will be able to afford the types of services VT1
offers for the first time when

Team - just me:
---------

The VT1 team is composed of a single entrepreneur with the following
capabilities:
* PC installation and administration.
* Phone installation and configuration
* Telephone and data network wiring
* Outdoor installation of solar and radio equipment.
* Strong desire to become expert in computer administration and the
repair of electronic systems.
* Bookkeeping, Accounting, and sales.

This entrepreneur will receive the free support, through the internet,
of the larger internet village telco community.

Options for Growth - a small sampling of fields in the massive matrix
of potential business models.
----------------------------

VT1 will build infrastructure providing it the capability to offer any
of the following valuable products and services within a 50km radius:

Networking:

* Internet access
* Inter-node data networking

Phone:

* Long distance phone service
* Local toll phone service
* Inter-node phone service
* IVR, web integration

IT services

* Setup services
* Data storage/backup
* Security - Antivirus/Antiphishing, etc.
* General IT training/services/support
* Operator services

Client Equipment

* Mesh Potatoes
* Batteries
* Phones
* Phone cabling
* Computers

NOTE: Services that will not be offered:

* Emergency 911-type services - although increased communications
ability will result in areas of deployment, no emergency service will
be offered or is envisioned in the future.

VT can provide phone services on one or more of the following terms:

* Markup on resale
* Traditional tiered models based on cost-basis of bulk purchases.

VT can provide networking services on one or more of the following terms:

* Flat monthly fee for bandwidth limited service, either percentage of
total bandwidth or fixed bps limit.
* Hourly or per-minute pre-paid service.

VT can provide IT services on one or more of the following terms:

* Hourly time and materials service with no gaurantee
* Contract service for well understood projects with no equipment or
software risk variables (i.e. installation of new equipment)

VT can provide equipment on the following terms:

* Lease included in contract
* Lease at fixed monthly fee with contract only
* Up-front outright sale, optionally including support.

Arguments against - playing devils advocate:
----------------------------

This business model, even in it's simplest form, is complicated, and
would challenge even entrepreneurs with highly technical backgrounds
gained through long term exposure to computer administration,
accounting, sales, and support.

The technology is unproven, and even when it is proven, equipment may
need to achieve near-mil-spec-all-weather quality to provide a
reasonable service lifetime. If and when all this is done, these
products must be made extremely cheaply. In addition, all this
engineering, as well as the creation of a reliable supply chain, must
be done for free by an ad-hoc community of disparate part-timers with
no stake in the game or other reason to provide support or even to
continue participating.

When communicating with the goverment and local telcos, one would have
to be able to separate meritless, harmless threats from serious,
business threatening ones, and then know how to deal with the serious
ones! Under most goverments, competition is simply outlawed before it
becomes a problem, and local telcos can simply refuse service.

Can we find any example of a similarly complex business model,
utilizing what was experimental, peer-produced technology, which
became a widespread, highly successful business model?

Cellular is the right technology for the third world. It has near
100% penetration and none of the limitations of village telco. The
per-minute rates are low enough to allow business to get done today.
The monopolies recognize this, and will provide data services via
cellular handsets when businesses can afford client equipment and
internet accesss. The technology is proven and robust and the
infrastructure and distribution network pervasive. This argues
against the need for and likely interest in village telco, which will
have a fraction of the reliability and reach that makes cellular so
valuable.


-Rich

--
-Rich

http://rbodo.blogspot.com
http://www.linkedin.com/in/complete

Skype: richbodo
irc: irc.freenode.net, icltlfatppl
Yahoo IM: rsb_mv

[rael lissoos]

Hi Rich

where are you based

rael
dabba

[Antoine van Gelder]

On 26 Jan 2009, at 16:17 , Rich Bodo wrote:
> Can we find any example of a similarly complex business model,
> utilizing what was experimental, peer-produced technology, which
> became a widespread, highly successful business model?

At risk of being self-referential:

Would you accept the companies and individuals who have built
successful consulting businesses providing support, training and
software development services around libré software (and hardware!)
projects as a valid example?

- antoine

[Cha...@thewybles.com]

Can we find any example of a similarly complex business model,
utilizing what was experimental, peer-produced technology, which
became a widespread, highly successful business model?

RedHat perhaps?
Sent via BlackBerry from T-Mobile

-----Original Message-----
From: Rich Bodo <rich...@gmail.com>

Date: Mon, 26 Jan 2009 06:17:17
To: <village-...@googlegroups.com>
Subject: Re: Village Telco Business Model

[Antoine van Gelder]

Some more thoughts:

On 26 Jan 2009, at 16:17 , Rich Bodo wrote:

> This business model, even in it's simplest form, is complicated, and
> would challenge even entrepreneurs with highly technical backgrounds
> gained through long term exposure to computer administration,
> accounting, sales, and support.

* What is the simplest subset of the VT1 Business Plan that will allow
the entrepreneur to reach profitability?

(optimizing for simplicity, _NOT_ largest one-shot non-iterated ROI)

> The technology is unproven, and even when it is proven, equipment may
> need to achieve near-mil-spec-all-weather quality to provide a
> reasonable service lifetime. If and when all this is done, these
> products must be made extremely cheaply. In addition, all this
> engineering, as well as the creation of a reliable supply chain, must
> be done for free by an ad-hoc community of disparate part-timers with
> no stake in the game or other reason to provide support or even to
> continue participating.

* How cheap must equipment be for business plan to survive a 5% p/a
failure rate ?

* What supply-chain goals (e.g. reflashing hardware, configuration
issues) cannot be achieved by parasitizing off consumer product supply-
chains ? (e.g. Linksys routers)

* What supply-chain goals cannot be achieved by creating an incentive
for third parties to manufacture libré designs ? (e.g. the IP04)

* How profitable must business plan be for entrepreneurs to afford
consultancy services of the ad-hoc community of disparate part-timers
when they want to scale their businesses beyond the base functionality
afforded by the platform ?

> Cellular is the right technology for the third world. It has near
> 100% penetration and none of the limitations of village telco. The
> per-minute rates are low enough to allow business to get done today.
> The monopolies recognize this, and will provide data services via
> cellular handsets when businesses can afford client equipment and
> internet accesss. The technology is proven and robust and the
> infrastructure and distribution network pervasive. This argues
> against the need for and likely interest in village telco, which will
> have a fraction of the reliability and reach that makes cellular so
> valuable.

Everyone in the cellular community I chat with keeps telling me the
same thing.

I don't yet have an answer for them that satisfies me.

One half of what I think is an answer is that in high-population-
density areas ADSL availability does not preclude 3G which does not
preclude WIMAX which does not preclude commercial WIFI which does not
preclude community WIFI.

So in economically-developing, low-population-density areas maybe
there is not so much room to differentiate on the transport layer, but
perhaps there is room to explore differentiation on the customer
service layer?

Awesome questions Rich!

[Steve Song]

Hi Rich,

On Jan 26, 4:17 pm, Rich Bodo <richb...@gmail.com> wrote:
> Since I'm new to the list, I thought maybe I should ask some more
> basic questions before getting too deep into the numbers.

No problem. BTW, for everyone else, Rich is based in California.
He's a telco startup veteran and now works for (the very cool)
Kiva.org. David Rowe and Rich have worked together in the past and it
was through David that Rich heard about the Village Telco.

> I examined the spreadsheet. I would like to cover some of the
> opportunities and risks, so I'll start with what a newcomers
> impression of a "village telco" business model might as well as asking
> about some of the risks.  If you think these are interesting, but it's
> more appropriate to discuss documents on the wiki, I can start some
> work in progress pages there and we can cover individual sections in
> their own threads.

Here's fine for now but as these issues start to take shape, I think
capturing them in the wiki would be great.

> VT1 Business Plan
> -----------------------------
>
> Summary - product and market
> --------------
>
> VT1 is a telecommunications service provider that will receive
> revenues from the following service package offered on a contract
> basis:
>
> * Internet access - A percentage of total available VT1 bandwidth will
> be delivered to users for a flat monthly fee.

The spreadsheet assume a percentage of revenue would be derived from
the Internet but I suspect Internet access will mostly be sold like
phone access i.e. via pay-as-you-go coupons. An assumption to be
tested.

> * Long Distance phone service - Phone service from the local phone
> company will be provided at a fixed markup over cost.

In most developing countries, caller-pays is the rule. Thus,
virtually all calls cost the caller, except in the case of the Village
Telco where on-network calls are free. So, mark-ups are made on:
- local calls to other networks
- national calls on the PSTN network
- international calls

A mark-up is also made on incoming calls from other networks that are
obliged to pay an interconnect fee to connect to Village Telco
customers.

> Lease of termination equipment to be included in the price of a
> one-year contract.

I don't think there is likely to be many contract services for the
scenario we envisage in South Africa. Customers will either be given
a Mesh Potatoes and we assume that the costs will be recovered in
network charges or they will pay for the Mesh Potato upfront or some
variation in between.

> Customer will recieve a single monthly bill for internet and phone service.

Customers will mostly purchase pay-as-you-go coupons generated from
A2Billing for voice services. Can't remember the data equivalent.

> Phone service and bandwidth-limited network connectivity between
> village telco nodes will be thrown in with each contract.

Yes. I think that is key. In fact, thinking about it, it might make
sense to create Village bundles where you could purchase 100 or 200
Mesh Potatoes at one shot, to instantly begin to leverage network
effects. Of course, that

> Internet phone service such as Skype is available in the region of
> VT1, and can be utilized by Vt1 customers at no extra charge.

Yes.

> Market consists of N businesses and homes in a 50km radius of the
> proposed VT office.

I suspect a radius of 50km is pretty big at least in the townships in
South Africa. A radius of a kilometre in a township should provide
sufficient subscribers for a VT entrepreneur to get started. Again,
an assumption to be tested.

> Y% of the market currently pay Z% more for telecommunications services
> than they will when they switch to VT1.

It is a little more complicated than that. Most townships are in what
are designated at "underserviced" areas. This means that service
providers who provide pay phone services via shipping containers that
are kitted out with a dozen pay phones can offer comparatively cheap
calls compared to mobile phones if you are prepared to walk to the
nearest phone container. In South Africa, Dabba (upon which the
Village Telco technical and economic model is based) offer a phone
calls at the same rate as the phone containers. The advantage then in
being a Village Telco customer is that
you can make cheaper than mobile calls from your home. You might
still have a mobile and receive calls on your mobile but at R2.50/min
to call on your mobile versus 90c/min to call from your VT phone, you
have a reasonable incentive. Plus free local calls.

> B% of the market will be able to afford the types of services VT1
> offers for the first time when

I think the kind of pay-as-you go Internet service that the VT will
offer may fall into this category.

> Team - just me:
> ---------
>
> The VT1 team is composed of a single entrepreneur with the following
> capabilities:
>  * PC installation and administration.
>  * Phone installation and configuration
>  * Telephone and data network wiring
>  * Outdoor installation of solar and radio equipment.
>  * Strong desire to become expert in computer administration and the
> repair of electronic systems.
>  * Bookkeeping, Accounting, and sales.

Hmmm... hopefully only to a modest degree in each area. The whole
point of the Village Telco is to turn this technology that any heavy-
duty wireless geek could set up into technology that any Cybercafe
owner could set up.

> This entrepreneur will receive the free support, through the internet,
> of the larger internet village telco community.

Yes, although in many cases it will be a small VT operator getting
support from a larger upstream VT operator such as Dabba.

> Options for Growth - a small sampling of fields in the massive matrix
> of potential business models.

[snip]

I am imagining the scale-up to be more via a kind of federated or
franchised model but this is open to question.

> NOTE: Services that will not be offered:
>
> * Emergency 911-type services - although increased communications
> ability will result in areas of deployment, no emergency service will
> be offered or is envisioned in the future.

Something to plan for the future but not as big a deal in the
developing world.

> VT can provide phone services on one or more of the following terms:
>
> * Markup on resale
> * Traditional tiered models based on cost-basis of bulk purchases.
>
> VT can provide networking services on one or more of the following terms:
>
> * Flat monthly fee for bandwidth limited service, either percentage of
> total bandwidth or fixed bps limit.
> * Hourly or per-minute pre-paid service.
>
> VT can provide IT services on one or more of the following terms:
>
> * Hourly time and materials service with no gaurantee
> * Contract service for well understood projects with no equipment or
> software risk variables (i.e. installation of new equipment)
>
> VT can provide equipment on the following terms:
>
> * Lease included in contract
> * Lease at fixed monthly fee with contract only
> * Up-front outright sale, optionally including support.

The model is going to vary from country to country and especially from
urban to rural environments but for individuals pay-as-you-go services
are pretty much a given. Offering institutional PABX-based VT
services would probably be otherwise.

> Arguments against - playing devils advocate:
> ----------------------------
>
> This business model, even in it's simplest form, is complicated, and
> would challenge even entrepreneurs with highly technical backgrounds
> gained through long term exposure to computer administration,
> accounting, sales, and support.

As I say, if the VT turns out to be technically finicky or complex, we
have failed in our mission.

> The technology is unproven, and even when it is proven, equipment may
> need to achieve near-mil-spec-all-weather quality to provide a
> reasonable service lifetime.  If and when all this is done, these
> products must be made extremely cheaply.  In addition, all this
> engineering, as well as the creation of a reliable supply chain, must
> be done for free by an ad-hoc community of disparate part-timers with
> no stake in the game or other reason to provide support or even to
> continue participating.

Yes, this is not investing in blue chip companies but hey apparently
that is risky too. :-) We are betting that we can find enough people
interested in scratching this particular itch to make a success of
this. For me this project isn't an interesting experiment, its
success is personal for me. I am betting my time, my reputation, as
well as funds from my patron on this, as are others involved in the
project. We can make this happen.

> When communicating with the goverment and local telcos, one would have
> to be able to separate meritless, harmless threats from serious,
> business threatening ones, and then know how to deal with the serious
> ones! Under most goverments, competition is simply outlawed before it
> becomes a problem, and local telcos can simply refuse service.

Yes, that is an issue, more or less so depending on what country you
live in. In South Africa, the regulator just granted 544 telco
licenses. It's a bit of a free-for-all at the moment.

> Can we find any example of a similarly complex business model,
> utilizing what was experimental, peer-produced technology, which
> became a widespread, highly successful business model?

Lots of examples to learn from. FON, Meraki, Open-Mesh and a host of
other services that have been mentioned on the list. For peer-
produced hardware, have a look at Chris Andersen's open hardware UAVs

> Cellular is the right technology for the third world.  It has near
> 100% penetration and none of the limitations of village telco.  The
> per-minute rates are low enough to allow business to get done today.
> The monopolies recognize this, and will provide data services via
> cellular handsets when businesses can afford client equipment and
> internet accesss.  The technology is proven and robust and the
> infrastructure and distribution network pervasive.  This argues
> against the need for and likely interest in village telco, which will
> have a fraction of the reliability and reach that makes cellular so
> valuable.

This is a very valid point. In a perfect world, we would be using an
Open Source GSM base station technology and use regular GSM handsets
to roll out the network. BUT, you need a Spectrum License as well as
an Operator's License to do that and they are very hard to get. For
grass-roots wireless networks to spread, you need unlicensed spectrum
that allows anyone with the wherewithal to get started rather than
getting wrapped up in regulatory red tape.

The recent decision in the US to open up the Television White Spaces
offers potential for regulators in the rest of the world to do the
same. This would mean having unlicensed wireless devices in the
600-700Mhz range. That could be very exciting.

Even if that doesn't happen, there is still an important role for the
Village Telco. The problem with mobile phone networks is that they
are walled gardens that can charge what they want and interoperate or
not when they want. This is what's wrong in developing countries.
Access is not just be able to get to a phone, it is being able to
afford to make calls whenever you need to. I have pontificated more
about this here:

http://manypossibilities.net/2009/01/why-wifi-in-africa/

http://manypossibilities.net/2009/01/mobiles-versus-laptops/

That said, I think the mobile issue is a very important one to push
back on in discussing the Village Telco. The VT has to deliver value
where mobiles already exist but are too expensive.

Thanks for the very thoughtful first post!

Cheers... Steve


>
> -Rich

>
> On Thu, Jan 8, 2009 at 2:09 AM, Steve Song <stephen.s...@gmail.com> wrote:
> > Happy New Year all.
>
> > It is possible that the lengthiness of my previous message may have
> > dissuaded anyone from opening the attached spreadsheet which is our
> > current best guess for the revenue model for the Village Telco.  I'd
> > be grateful if anyone were willing to take a look, comment, ask
> > questions, suggest flaws, additions, deletions, reframings,
> > suggestions for making it more readable,intuitive.... you get the
> > idea.
>
> > Many thanks in advance.... Steve
>
> --
> -Rich
>

> http://rbodo.blogspot.comhttp://www.linkedin.com/in/complete

[Charles Wyble]

<snip a bunch>

> This is a very valid point. In a perfect world, we would be using an
> Open Source GSM base station technology and use regular GSM handsets
> to roll out the network.

Say http://openbts.sourceforge.net/ and Android? :)

BUT, you need a Spectrum License as well as
> an Operator's License to do that and they are very hard to get.

See http://www.kestrelsp.com/FieldTest/index.html for one adventure in
that area.

For
> grass-roots wireless networks to spread, you need unlicensed spectrum
> that allows anyone with the wherewithal to get started rather than
> getting wrapped up in regulatory red tape.

Very true.

>
> The recent decision in the US to open up the Television White Spaces
> offers potential for regulators in the rest of the world to do the
> same. This would mean having unlicensed wireless devices in the
> 600-700Mhz range. That could be very exciting.

I'm much more excited about 3650 to 3700 mhz in the USA:
http://www.socalwifi.net/index.php/3650_to_3700_mhz_spectrum_information

20 watts. Lightly licensed.

>
> Even if that doesn't happen, there is still an important role for the
> Village Telco. The problem with mobile phone networks is that they
> are walled gardens that can charge what they want and interoperate or
> not when they want.

Right.

This is what's wrong in developing countries.

Among many other things :)

--
Charles N Wyble cha...@thewybles.com
(818)280-7059 http://charlesnw.blogspot.com
CTO SocalWiFI.net

[David A. Burgess]

Valid points on both sides, but I'd like to toss out a few more and
ask a few questions.

First, GSM handsets will always be much cheaper than 802.1x gear.
The radios are much simpler, cellular handset markets are already
driven by huge downward price pressures and the supply of refurbished
GSM handsets is already good. Given that, GSM deserves serious
consideration as a last-mile solution for VoIP because its advantages
are probably great enough to outweigh the associated difficulties.
The barriers are strictly regulatory, not technical, not even
economic. (This is a "people problem", which is part of what makes
it so frustrating for engineers.)

Second, in any rural area there are vast spans of fallow spectrum.
This is true even in rural areas that actually have cellular
coverage. A carrier's typical allocation is usually 5-7 MHz wide and
in a rural area they are only using a few percent of it, if any.

Third, the $1-$2 APRU rural market is not addressable by incumbent
carriers, so they have little economic advantage in preventing
someone else from trying a limited-feature second-tier service in
places where the spectrum is fallow anyway. The only threat to the
incumbent is that they might loose first-tier customers to the
cheaper service. The main way to counter that threat is to make sure
the second-tier service cannot be a good substitute for normal
cellular. For example, the second tier service may not support
handovers of active calls between access points, or second-tier
subscribers may not be allowed to roam onto first-tier networks.
That's all acceptable for a last-mile link and a real show-stopper
for cellular. Basically, GSM becomes a DECT substitute, but with
much longer ranges and cheaper handsets. You now have two different
products for two different markets and very little direct
competition. The advantages to the incumbent? They get a little
extra money on spectrum leases and can get the regulators off their
backs about universal service. People may be offended by the
suggestion of limited-feature networks for the poor, but by my
thinking it makes a lot more sense to provide basic service that they
CAN afford than to insist on a feature-rich network that will never
actually be built. And, if it matters, the phones in the village
will look and sound the same as in city because they'll be the same
phones, just using the networks in different ways.

So here are some questions:

What countries out there are deregulating spectrum to the point of
allowing leases?

Does anyone think this line of reasoning might actually work with any
carriers out there?

Does anyone think this line of reasoning could be used to convince
regulators to take actions to free of fallow spectrum for new operators?

I don't mean to hijack this list, so if you feel it's appropriate to
respond privately, that's fine.

-- David

OpenBTS on the web:
http://openbts.sourceforge.net
http://openbts.blogspot.com
http://en.wikipedia.org/wiki/OpenBTS
http://www.gnuradio.org/trac/wiki/OpenBTS

[Vickram Crishna]

Also a newbie on the list, and I don't want to repeat stuff about regulators that we all know and live with, so let me just respond to a couple of issues that irk me:

GSM (or CDMA, both of which are now commonplace in India, hence discussions of device pricing etc are much the same) has a problem that doesn't seem to attract much attention. I refer to the fact that handsets are designed to boost signal levels when the tower is further away. This means that the typical rural subscriber to a first tier service is being regularly exposed to far higher levels of microwave radiation than her urban counterpart, and is well known, the jury is still out on the health issues relating to such radiation.  

Secondly, and I know this might not be as fantastically critical, operating the device at higher EIRP levels means it draws far more juice from the battery, which then needs recharging more frequently. Electricity in rural areas is much more expensive than in the cities for the simple reason that it is much more unreliable and unavailable, hence the real cost of energy is higher than the metered cost.

Both these issues are far better dealt with by having many more local rural towers, ie microcells, and as long as this is the way to go, then why venture into regulatedly thorny spectrum areas, when there are open alternatives? I forget who said it, and said it better, but to win battles, one must choose one's battlefield carefully.

As far as the radios are concerned, it is hard to believe that medium priced handsets (which are all smartphones these days) are less complex (hw+sw) than 802.11x gear. I think the pricing is merely a market issue. If anyone from the manufacturing side is on this list, especially from Taiwan or Korea, I'd like to hear a response to this simple query: What street price would you set for a 802.11x smartphone (with arguably no GSM or CDMA, but maybe Bluetooth) if 100 million could be sold in 2009? 

--
Vickram
http://communicall.wordpress.com

[rael lissoos]

Hi David

I agree that the current GSM etc technology is cheaper and in many cases
better suited to voice.

There are however 3 main issues that do allow for a village telco niche

1. In many developing countries the spectrum is highly regulated and the
GSM operators are exploiting their licensed oligopolistic position to
extract super profits through very high charges. (this you mentioned,
issues are not technical). This is the main reason for the viability of
the VT. W
2. The village telco set-up costs are low enough to allow a small
operator to cover a niche neglected market. Wifi and FOSS allow a low
cost entry. On a smaller scale this is more cost effective than GSM or
other network set-up. Hopefully with volume demands from many village
telcos scale can be created for alternative products especially handsets.
3. Product differentiation, if the VT is just viewed as a competitive
network then the feature list is an issue, handoff roaming etc.. if it
is seen as an alternative then it may stand on its own.

On anther note the business plan is not hypothetical any longer, we have
currently been running a VT project in South Africa. It is just
starting to reach a mature state with a few thousand users, and we will
be able to analyse the data and systems soon to establish the economic
viability of the exercise.

Rael
dabba