[kictanet] [Skunkworks] Even Safaricom these days thinks KIXP is non-existent!

Michuki Mwangi michuki at swiftkenya.com
Tue Apr 10 20:58:38 EAT 2012


Hi Brian, et al,

See my comments inline.

On 4/10/12 2:32 PM, Brian Munyao Longwe wrote:
>     1.  Tom Omariba (MTN Business)- Chairman
> 
>     2.  Chris Senanu (Access Kenya)– Vice Chairman
> 
>     3.  Kenneth Munyi  (Iway Africa)- Treasurer
> 
>     4.  Michuki Mwangi (Internet Society) – Chief Technical Officer
> 
>     5.  Tejpal Bedi (Chair,Kenya IT and Outsourcing Society)- Director
> 
>     6.  Michael Terik (Kenyaweb.Com)- Director
> 
>     7.  Beatrice Mudhune (Internet Solutions)- Director
> 
>     8.  Abduaziz Osman (Sahannet Ltd.)- Director
> 
>     9.  Laurnet Giraud (Orange Kenya)– Director
> 
> 
> 
> This is the crux of the matter. Anyone familiar with company law knows
> that the directors and shareholders of the company can elect at any time
> to amend the articles and memorandum (within the confines of the Company
> Act). If TESPOK really wanted KIXP Ltd to be independent they would have
> taken this action ages ago. Which brings me back to my original point -
> KIXP Ltd is captive to TESPOK whims. It might be difficult to unentangle
> the two now that their fortunes have been so closely mixed up. All of
> KIXPs income is paid into TESPOK accounts, and thereby subject to
> discretionary use by TESPOK. Is there a governance problem here? I think
> so... for years the dust has been swept under the carpet, can we grow up
> and make sure that our institutions, especially those as critical as
> KIXP have the right structures in place?
>  
> 

Brian, considering that you were involved in the process of setting up
TESPOK and KIXP, and in understanding of the list of members (that
connect to or get services from KIXP) help me understand the following.

1. What was the objective of TESPOK at setup?.

2. Why was TESPOK involved in the setup of KIXP - what was the objective
then. Does it mean that if CCK did not pose a license requirement then
KIXP Ltd would not have been setup?

3. What would be the objective of changing this TESPOK/KIXP model today
- what has changed?.

4. What would be different about the board of the KIXP from the current
TESPOK board. (see 2 above).

5. In your earlier post you referred to KIXP as "weak KIXP that cannot
seem to consistently engage newcomers to the industry with the benefits
of local traffic exchange." Could you please substantiate your comments.

In your earlier post "how many of the TEAMs/SEACOM/EASSY
bandwidth-holders are peering at KIXP?". I think a more appropriate
question would be what percentage of Kenyan Networks (ASN's according to
AfriNIC) are visible at KIXP. To this i can answer. We have over 80% of
Kenya's ASNs. 52 of the 62 allocated by AfriNIC are visible at KIXP (as
at Jan 2011). Though its 90% since some of the ASNs are of companies
that have been acquired, closed shop or using satellite only, etc.

In addition, there were 55 ASN's from outside Kenya visible from the
KIXP of which >30 were from East African countries and 13 from US. In
summary there are 107 ASNs from 17 countries reachable via KIXP.

Therefore and selectively quoting your earlier email "KIXP would become
the de-facto facility for providing industry actors with data
interconnection and interchange." and i will add "not just for Kenya but
the entire African Region."

quoting your earlier post "KIXP be given full autonomy ....be run as a
business, similar to LINX in the UK, and other successful IXPs around
the world." You will be surprised to know the extent at which the TESPOK
Board has gone to learn from LINX, AMSIX and other IXPs in the world.
For your information, KIXP is a member of the European-IXP Association -
https://www.euro-ix.net/news-and-events and the objective is to learn.

In this regard, we are well aware of the models used by LINX and others.
The fundamental components being that;

1. They all remain non-for-profit entities like TESPOK/KIXP
2. They opened their IXPs to non-ISPs for services. Same for KIXP.
3. They have a business model which bills on port speed. Same for KIXP
4. They are virtual IXPs located in more than one site - KIXP has
started this as well (few carrier-neutral data centers in Nairobi).
5. Routers not required at IXP locations (remote peering) - Started
since Jan 2012 to reduce entry and operating overheads.
6. LINX has a policy engagement model with Govt. A role played by TESPOK
7. etc.

I believe this gives a good overview of the Governance of TESPOK/KIXP in
comparison with others. Remember the IXPs and ccTLDs alike are very
contentious on governance model since one size doesnt fit all. It is
however clear that a successful governance model is what works best for
the local Internet community. Our current model is inherent of legacy
issues that are still present to date i.e a license from CCK amongst
others. However, it has not restrained us from growth, ability to
collaborate with stakeholders or deliver on the objective compared to
other IXPs setup at the same time in the region. We are the 2nd largest
IXP in Sub-sahara Africa after JINX.

So my question and its of particular interest to me since am on the board.

1. What would you like to see different with respect to the Governance
of TESPOK/KIXP?.

2. What would be the objective and goals of the proposed model?


my 2 cents.

Regards,

Michuki.


















More information about the KICTANet mailing list