Quoting the article by Rebecca Wanjiku… For a regulator like the Communications Commission of Kenya (CCK), you would expect them to be promoting services if the whole talk of how infrastructure has improved and how it is getting better. Read more…http://www.wanjiku.co.ke/2012/04/cck-hosting-its-website-abroad/ RgdsGG
_______________________________________________ kictanet mailing list
Grace/Wanjiku,
While I see the argument you are raising for local support, I think there is a bigger question as to whether the Internet is local? Also how many of the local hosting companies are locally owned? I think you should evaluate the total economic impact to Kenya and not base it on face value. It would be great if as an industry we answer some key questions and then decide whether it is beneficial to use “local” or imported. My phone is Korean, Laptop Apple (Chinese or US – you decide), Office furniture from a South African company, imported from Italy etc.. back to the subject of local hosting ….. How many Kenyans are in the diaspora and using the Internet. I think you are asking the wrong question. You should be asking about the quality of service to the visitor of the site. Various considerations go into where to host a service; 1) Cost 2) Security 3) Quality of Service – where is your primary audience, what devices (mobile (feature or smart phone?, desktop, tablet ) 4) Search Engine Ranking and optimisation – Discoverability 5) others …
Who is best positioned to provide this kind of comparison data? Can you investigate and give us feedback?
Thanks
Joe Mucheru
Ochieng Maxwell Technology Officer Messaging Solutions Tel:+254-20-6000295/6 +254-20-3569132 P.o Box 44926-00100, Nairobi
Sent from my iPad 2
thought CCK should think of the bigger picture when it comes to cost, how much does it cost Kenyans to pay for international bandwidth to access CCK website and what is the purpose of the IXP and how do we promote it? most Kenyan sites hosted outside the country mainly run on vps or dedicated hosting and are managed by web developers or “brokers” whose understanding of security is very minimal and are mainly control panel experts so I doubt if security is ever a major concern.
with EA undersea fiber I thought audience location should not matter unless the website is hosted on dial up in someone’s bedroom. Devices thought should be a developer thing. My assumption would be hosting locally would help Kenyans improve on their security knowledge , create a few jobs, give KPLC an extra income and another excuse for blackouts and not sure but would assume even CCK data is some sort of local content which needs to be consumed locally.
I think the most concern for hosting websites in Kenya is cost and KPLC, cost for CCK I doubt is an issue considering their responsibilities. I also doubt they can’t find a provider with some reliable power backup system locally.
My 1 cent
_______________________________________________ kictanet mailing list
Mr. Mucheru, You have succeded thinking like a businessman.
But don’t forget CCK is the government. The government banned importing furniture for its offices. The spirit of that gazette notice was, “the government should utilize local resources first, before looking abroad for unavailable resources.”
India surpassed Japan as the World third largest economy http://www.forbes.com/sites/kenrapoza/2011/09/21/india-likely-to-replace-japan-as-worlds-third-largest-economy/
How is India doing it? How is China doing it? We all know the long-long-long term effect of supporting local products and services (not minding the nationality of who owns them)
1. Buy Local — Support yourself 2. Create more good jobs: Small local businesses are the largest employer nationally, and provide the most jobs to residents. 3. Invest in community: Local businesses are owned by people who live in this community, are less likely to leave, and are more invested in the community’s future. 4. Encourage local prosperity: A growing body of economic research shows that in an increasingly homogenized world, entrepreneurs and skilled workers are more likely to invest and settle in communities that preserve their one-of-a-kind businesses and distinctive character. 5. Keeping Local Ksh in the Economy: We are always crying how the dollar is going up, e.t.c. while we are importing useless stuff.
Any fact facing us is not as important as our attitude toward it. – Norman Peale
10rdmwesh
______________________ Mwendwa Kivuva For Business Development Transworld Computer Channels twitter.com/lordmwesh kenya.or.ke | The Kenya we know
_______________________________________________ kictanet mailing list
One of the biggest costs that relate to webhosting is bandwidth. Then you can think about power and manpower (IT skills), not necessarily in that order. To run a really big data centre, you need these resources in large numbers and eficiently. Out there the cost of bandwith is like 10 times cheaper than what we have, if not more. Someone can correct me on this. Certain countries have highly subsidized bandwidth. Power is very expensive in Kenya, dont even talk about the urealibility. I was once at electricity house to pay my elec bill and I could not because there was a blackout there!
So, countries that have sorted themselves in all these variables are the ones running the big data centres. And because of the scale, they are able to offer the hosting services at a very low price. We will need a lot of time to atch up with them, unless of course we run faster than we are running right now.
Evans
I find it hypocritical, that we want to be at the forefront of lobbying government to outsource locally, but when that is done, we in turn outsource abroad using sleek phrases and acronyms like BPO/ITES, Lack of Capacity, Costs, Security… the list is endless!
While I see the argument you are raising for local support, I think there is a bigger question as to whether the Internet is local? Also how many of the local hosting companies are locally owned?
The Internet is as local as the Internet is abroad and vice-versa. The question of whether a firm is locally owned or not, is a “local-hosting” irrelevancy that needs to be addressed differently. It may be entertaining to watch Nigerian oga movies produced in Abuja, it would be better to see a few Kenyan actors star in those movies and a few scenes shot in Kajiado, the best experience is to have a hearty laugh watching hilarious Naswa/Pasua comedy clips.… whether some Nigerian owns the production company behind Naswa is equally an irrelevancy as far as “local-do-it-ourselves” goes.
The net effect of every shilling spent in Kenya as opposed to being converted to dollars and spent abroad, is pretty straight forward I should think. I find it hypocritical (if not defeatist – and bordering on an economic crime) that a company “saves” by hosting abroad and later claims to partake in CSR activities and benefits from Tax incentives for the phony CSR!
It would be great if as an industry we answer some key questions and then decide whether it is beneficial to use “local” or imported. My phone is Korean, Laptop Apple (Chinese or US – you decide), Office furniture from a South African company, imported from Italy etc..
I want to believe that the decision to import an Apple computer is more informed by the lack of a local alternative in functionality, aesthetics, prestige or some peculiar sentimental value. If only there was a local brand called Chungwa that could equally tickle your gadgetry taste buds….. The same “lacking” cannot be claimed for local hosting.
How many Kenyans are in the Diaspora and using the Internet.
What difference does it make? If anything, they ought to be on the frontline advancing our cause as a “hosting”/ICT destination, while we develop the requisite capacity to absorb the business they forward our way. I think it is immoral from a policy perspective to target the Diasporas while 40 million are languishing in traffic Jams and endless dropped calls, just to mention a few of the “easiest-to-solve” of our local ICT problems…
America did not just wake up July 4th and decide to outsource software development to India or hardware to China. It is the nationals of these countries residing/visiting America who built their respective cases and they did so with the confidence that back home there was sufficient capacity and capability. Unlike in our case, they managed to do so despite language and accent barriers and differences in political ideology and nuclear capability. We seem to have our cart in front of the horse, expecting the IBMs and Googles of this world to come develop capacity for us then swing us some!
I think you are asking the wrong question. You should be asking about the quality of service to the visitor of the site. Various considerations go into where to host a service;
With respect to CCK hosting, for as long as they are dealing with a middle man (aka Broker/Reseller) then topics like Security, Quality of Service ought not to arise, and if they do, then the contract ought to go to a firm with demonstrated local infrastructure.
Regards
Sent: Sunday, April 22, 2012 11:38 PM Cc: KICTAnet ICT Policy Discussions
Grace/Wanjiku,
While I see the argument you are raising for local support, I think there is a bigger question as to whether the Internet is local? Also how many of the local hosting companies are locally owned? I think you should evaluate the total economic impact to Kenya and not base it on face value. It would be great if as an industry we answer some key questions and then decide whether it is beneficial to use “local” or imported. My phone is Korean, Laptop Apple (Chinese or US – you decide), Office furniture from a South African company, imported from Italy etc.. back to the subject of local hosting ….. How many Kenyans are in the diaspora and using the Internet. I think you are asking the wrong question. You should be asking about the quality of service to the visitor of the site. Various considerations go into where to host a service;
1) Cost
2) Security
3) Quality of Service – where is your primary audience, what devices (mobile (feature or smart phone?, desktop, tablet )
4) Search Engine Ranking and optimisation – Discoverability
5) others …
Who is best positioned to provide this kind of comparison data? Can you investigate and give us feedback?
Thanks
Joe Mucheru
My comments on the issue:
The government does have a prerogative to develop local industries. A simple example is Embraer in Brazil, which grew to it’s current level (with Kenya Airways buying jets from it) primarily from the government purchasing. We don’t want the government to run the companies in the ICT space, just use the taxes it collects from us to develop the very same industries.
If the private sector has it’s act together (as is the case with many companies e.g. Cellulant, which has won a contract from the Government of Nigeria recently, Seven Seas etc) and we have the capacity, it’s wrong for the government to outsource the work… The government of Rwanda currently has a model that for large government contracts, you have to work in tandem with a local firm, which has two companies (Rock Global Consulting & Matrix Business Solutions) experience accelerated growth and are now capable of handling a lot of the business the government has without external partnerships. They have grown their capacity. The government loses less money and this in a small way, fixes their balance of trade and increases employment locally.
The US government is currently trying as much as possible to end outsourcing with companies like Apple/Cisco etc.
I am not saying that we should adopt a model that was used by India/China in the mid-late twentieth Century (extreme market protection) but I believe it should be tempered. Importing milk from Tanzania for example, will simply kill our dairy industry. Market protection has it’s ills as well, but if well done, grows your economy.
I agree, to a point, with Joe’s approach, however, with CCK, given the nature of their business, will necessarily have most of their traffic being local. We do have some good local hosting companies. If for instance, the tender was inclusive of all that and a preference for local hosting (if the website were to be local) or international (in this case the US), then this should clearly be specified at the tender stage.
The government wants to have 500 companies the size of Seven Seas technologies (according to what I read from the recently ended Connected Kenya Summit – correct me if I’m wrong http://softkenya.com/kenya-ict-board-plans-500-new-firms-by-2017-to-push-kenya-to-top-10-ict-hubs/) which basically means that in the next five years, we will have, 500 companies with over 1B KES in revenue (500B KES – 6.01B USD, with current rates). How are we ever going to achieve this if the government (currently the largest consumer of ICT Services) does not channel it’s resources into local firms?
The government has data centres which are barely utilised http://www.cio.co.ke/news/main-stories/Kenyan-Government-to-launch-3-data-centres
I once tried highlighting issues with Government in my *Last Word *column, though found that I was burning my fingers in the process, and I shall keep my distance from that discussion.
In my coming column in the same magazine, I have highlighted that we are pushing too much for “local sourcing” rather than an export oriented economy.
If local firms are already exporting, how can we help them export more ? How can we add more firms exporting?
While protecting the local firms may looks prudent, we may find ourselves in the receiving end when those who we were exporting to decide it is unfair that we are “keeping them from our market”. this has to be done very careful.
South Korea, unlike the rest, decided to be crazy and protect its exports – the country is the 8th largest importer and 7th largest exporter – by ensuring that key components are always in supply http://internationalbusiness.wikia.com/wiki/South_Korea‘s_export-oriented_import_protection_economy
Brazil on the other hand, did not have a smooth ride with its import substitution program. It had to devalue its currency (weren’t we all crying with a weak currency last year?) and the military decided it was best suited to fix things at some point http://countrystudies.us/brazil/62.htm
_______________________________________________ kictanet mailing list
In all these discussions, I think Kenya should avoid creating some form of affirmative action when it comes to contracts.
Companies in Rwanda may grow because foreigners have to work with them… But are they technically competent
There are conversations going on right now around 140 friday about how to strengthen the sector… For me, I think we should build good quality stuff or do good quality work and the contracts will flow
You win some, you loose some
Sent from my BlackBerry®
+1 on Agosta’s comments , plus him being one of our top software exporters, they should be valid .
Just to clarify
Am not a top anything (that’s for the big boys)… I strongly believe that our work should speak for itself
So in building KITOS or whatever we will call that org, we should focus on what we can do as a sector… Then govt may follow
Not start asking that all Kenyan projects should be done by Kenyan firms… Technically, all these “Kenyan Firms” have foreign shareholders…
Thanks Sent from my BlackBerry®
Dear all,
I think that one thing that has not been taken into consideration is the fact that CCK does not operate it’s own hosting infrastructure. If I recall from my ISP days and also assuming that CCK is still a public entity, they are subject to procurement procedures – and hosting their website was one of the hotly contested contracts that we used to compete for.
I therefore would like to suggest that, in this case at least, the guns are pointed at the wrong target. Which ISP has the hosting contract? Could it be that their hosting infrastructure (like many others) is based in the USA?
Food for thought….
Brian
Brian is right. A local ISP or hosting business would be awarded the tender and proceed to provision the hosting service on infrastructure in a different country.
That would be UUNet, me thinks?
Kind regards,
Muchiri Nyaggah | PRINCIPAL PARTNER @muchiri Cell: +254 722 506400
eGovernance, Healthcare, ICT and Financial Services Innovation for Africa
SEMACRAFT CONSULTING PARTNERS Nairobi, Kenya. http://www.semacraft.com | http://www.semacraft.com/blog twitter: @semacraft
Domain Information Query: cck.go.ke Status: Active Created: 13 Feb 2003 Modified: 27 Dec 2009 Expires: 01 Jan 2016 Name Servers: ns1.uunet.co.ke ns2.uunet.co.ke ns1.iafrica.com ns2.iafrica.com
Registrar Information Registrar Name: MTN Business
Phares
This statement comes across as marketing types seeking to ride off Seven Seas runaway success in Africa.
How do you increase maize production by increasing the number of farmers a hundred fold, yet existing farmers are wailing left right center and you are doing nothing to increase the amount of nor fertility of existing arable land?
Regards
_______________________________________________ kictanet mailing list
I strongly believe that local traffic and content should stay local. When you host your content in a foreign land you become more exposed to intellectual property (IP) theft and other forms of Cyber Crimes.
In this age on government sponsored Cyber Spying, one needs to be very careful.
Companies like Huawei have been barred from running sensitive National Security Projects in Australia, Europe and the US.
http://www.bbc.co.uk/news/business-17509201
The US Federal government undertakes rigorous tests on all imported hardware in search for firmware-based spyware.
When you contract a foreign company to host your content you should have stipulations in the contract to keep local content local.
If you need to avail your content to users in different geographical locations, you could make use of local and remote mirrors/cache replication, load balancers or Round-Robin DNS. Global companies like Google and Facebook make use of these technologies.
I think that we can all agree that International bandwidth is expensive. Trying to save on hosting but incurring higher costs on international bandwidth seems contradictory, not to mention loss of foreign currency. If CCK is main audience is the Diaspora, then the website should be hosted abroad with a local mirror in Kenya.
Hosting prices are dirt cheap on Amazon and RackSpace. I dispute the fact that there is lack of local talent to manage such portals. One needs to take a peek on LinkedIn to realize just how much Kenyan talent is out there. The Diaspora also holds tremendous pool of talent and with the creation of the Diaspora Ministry, this pool of talent can be harnessed.
What is lacking in my opinion is the will power and infrastructure to implement this.
http://www.cio.co.ke/news/main-stories/Kenyan-Government-to-launch-3-data-centres
The figures in the story are either wrong or the Datacenter is seriously in need of some major redesigning.
In this age of 3TB Hard Drives costing less than $200, and the price of SSD Drives coming down, it does not seem Vision 2030 friendly to setup a National Datacenter with a capacity of 24TB of Production Storage. I usually recommend to my clients a minimum of Thin-Provisioned 5TB Storage per user, even Hotmail Sky Drive offers 5GB with Google planning to do the same. There a various capacity planning storage calculators out there provided by different depending on the content you plan to host.
The article also claims that the 2 Datacenters will consume an astronomical amount of 200 Megawatts.
Someone must have got their math wrong. For Example Google runs some of the most power efficient Datacenters in the world.
According to this report, the global power consumption for all Google Datacenters currently stands at 200 Megawatts.
http://www.oregonlive.com/silicon-forest/index.ssf/2011/09/google_divulges_data_center_power_usage.html
Power efficiency is enhanced through the use of Virtualizing and commoditizing physical resources like CPU, Memory, Hard Drives and even Networking.
Best Practices in Datacenter design and the advent of new Datacenter Technologies like Cisco UCS, VCE vBlock and NetApp FlexPod has greatly improved power consumption.
The key to hosting local content will have to be the government leading the way by providing incentives to Private companies to build world class self-sufficient Datacenters largely unaffected by the constant power blackouts in the country.
A successful world class Datacenter should maintain a minimum of “four nines” SLA which equates to about 4.32 minutes of downtime per month. The Datacenter should have a reliable connection to a remote Disaster Recovery Site (DR) preferably though Dark Fibre, taking into consideration the Recovery Point Objective (RPO) and the Recovery Time Objective (RTO).
Gabe Shompole, Virtualization and Storage Consultant Atlanta, GA.
Lastly, it will be interesting to see where local companies do their hosting when “Big Data” finally hits Kenya.
Gabe Shompole, Virtualization and Storage Consultant Atlanta, GA.