I retired from personal blogging in July 2008 but you can find me over at blog.xero.com
Follow @roddrury
Posted by rod@drury.net.nz in Communications at 4:19 pm on Thursday, 12 April 2007
Video from ASB Business this morning where I discuss the ‘People’s Fibre’.
http://tvnz.co.nz/view/page/414443?video_id=105955
blog
Brilliant and well put, would be good to see a debate on this.
How come no Xero.com shirt? :)
Afternoon Comrade Drury, is the peoples fibre like the peoples hospitals, schools, roads, power generation etc?
Why don’t you tell us the details of why the Aftermail VOIP problem became a “Tipping Point” in this for you. Why could you not get enough bandwidth?
How fast a line speed were you using and what type of line - wireless, frame relay etc.
Maybe if I understood this I could understand why you are beating such a loud drum about this issue.
Personally we don’t have an issue with Internet line speed at work - get 10 megabits fibre speed and can consistently download from places like MSDN at 1.25 megabytes per second.
Home on ADSL is another kettle of fish but then I would not want to use ADSL for commercial software development; there are better alternatives.
If you can convince me I will even wear the “Rod for President” button.
Dermott,
Harsh mate. I agree it is valid to ask what was up with the voip solution.
But i think the point about government owned core infrastructure is valid. Maybe this way our tax money might actually do so good. Untill something radical happens a the levels of investment in the core infrastructure aren’t going to allow the growth of a services company.
I think this is a good play personally. Wonder who in governement thinks the same
Paul, my point has always been -
I think our Internet and comms infrastructure could be better, way better.
But governments in NZ have no recent history of providing good infrastructure in other areas; we have problems in health (current Wellington hospital debacle), the roads are terrible, the power is marginal at best, we have no defence services anymore, how many other areas do you want me to mention.
So given the above what makes anyone including Rod think the government could run anything infrastructure wise? Regardless of how good it starts, gravity will take hold and in 5 or 10 years no continuing investment will have been put into the network and then we will be behind again.
There are other ways of doing this that involve the government other than ownership and I am surprised that Rod and most others have not mention these.
Other than the above issues, regardless what some government ministers think about whether this is a good idea or not, Michael Cullen controls the money and there is no way he will spend billions on this. Look how he prevaracated over the Air NZ debacle and could not make a decision and it virtually fell out of the sky before they made a decision. And Air NZ is not now successful because of government ownership as they have no say in how it is run.
None of the above denies there are serious issues but government ownership will neither solve these issues nor solve them quickly.
What will solve the issue quickly is the government saying to Telecom the following - either fix the network infrastructure quickly (like get NZ right up to date within 9 months) or you lose the stewardship of the network and it is taken off you by legislation and given to someone else i.e. Telstra, Singtel etc.
An example of this happening is in the UK where the railways was split into franchises with different operators responsible for meeting certain levels of service. Within the last 18 months one of them had it taken off them and it was given to someone else due to a lack of performance.
This is what the government should do now to Telecom. I agree with Rod that this issue is of huge importance to NZ and I think rather than the government waiting for Telecom to stop dragging the chain, which they continue to do, they should just say, the law has changed next week it will be put up for tender and you will be paid $x compensation for this part of the system.
I would still like to hear from Rod what the issues were at AfterMail, or anyone else’s specific issues. Time for specifics people rather than generalities.
Dermott, appreciate your comments but I still don’t think you get what I’m saying.
On your question. We subscribed to a commercial VOIP solution (packet8.net) that gave us a US phone numbers. We were on CityLink so I assume a pretty good. connection. It did not work. The quality was not good enough so we had to change back to a normal POTS +64 … call. That hurts business.
Skype for me is variable at best, but not a real commercial solution. My point is if we cannot even have basic VOIP what chance for really useful stuff like hi-res video.
I have tried to make it clear that the kite I am flying does not require significant govt funding. The Internet is a user pays model. The SOE acts as a coordinator. The money required may be more than a single company, even Telecom, could afford. So the SOE could issue a large infrastructure bond paying a modest interest rate. this would flow into the (say) cost + 10 model.
This would be good for Telecom too.
We all hear you govt ownership. But perhaps with a good goverance structure and clear charter it could work. It is a valid concern.
But the ‘lean on Telecom’ is just more of the same. If we want step change for the good of the country I think we have to be more radical.
Rod, interesting that on Citylink the quality of VOIP was not good enough. So was the problem with City Link or Telecom, who knows. I have never tried VOIP on our connection which would be interesting.
I agree with you on Skype; I don’t use this for security reasons, the super node using your resources issue.
It maybe from what you are saying then that we may be closer than I thought; or at least have some crossover areas.
I like the British Rail franchise model because if the provider does not perform you replace them.
My lean on Telecom was not the historical impotent leaning on Telecom but you have 30 days for a plan otherwise the rules change. I just don’t understand what Telecom’s problem’s are - lack of will or lack of money.
On related but different note I asked my IPS, when would they have ADSL 2 thinking possibly it might replace my ADSL at home (I get 2 megabits at best, old copper in ground and too far from exchange) and they said -
“We will wholesale ADSL from Telecom in the same way we get UBS/BitStream, however, we will be investing in a new VDSL netowrk. We will basically leapfrog ADSL2 and roll out 100 Mbit Broadband. This will happen over the next 8 -12 months”.
Interesting
Rod, looks like Telecom have a similar view to you -
http://www.nzherald.co.nz/section/3/story.cfm?c_id=3&objectid=10434051
The questions is can they not make it pay or see it as a problem.
If I was you I would start talks with someone like Singtel.
I have been trying to download the video.mp4 podcast from ASB Business and it keeps reporting error. The other video podcasts come down fine.
Nice visions Rod.
>On your question. We subscribed to a commercial VOIP solution (packet8.net) that >gave us a US phone numbers. We were on CityLink so I assume a pretty good. >connection. It did not work. The quality was not good enough so we had to change >back to a normal POTS +64 … call. That hurts business.
This is an argument against buying purely local capacity.
If you buy pure local connectivity to an exchange, you are at the mercy of the upstream carriers who peer with that provider. You won’t get a guaranteed QoS. Sure, your local connectivity might go very fast; your national might be less fast and your international equates to junk.
If you buy _access_ (national or international) on top of the local, then you can get decent bandwidth. But that costs more. Doing QoS across multiple providers is very hard.
I think the answer is a service-based model. You buy voice as a service, you buy data as a service. Anything else in between is irrelevant - it’s up to the service provider to work out the details of the provision. And in the same way as normal logistics, keeping that supply chain as short as possible and having agreed parameters at the handoffs between the providers is a Good Idea.
Sure if you’re a telecommunications guru. I just want to run a business that can export effectively. Most people are not going to roll their own, they just want a useable service, from one provider with a reasonable rate card.
why should the entire industry change if you make poor decisions about your ISP?
Can you suggest an ISP that provides from central Wellington QoS for VOIP to the US and UK so I can have reliable US and UK Phone numbers, desktop video and high resolution video conferencing. Oh yes and please post the rate card for the above.
I want to believe.
The company I work for just purchased 2megabits of guaranteed international bandwidth. That, according to the spec from lifesize is enough for one HD video conference with room to spare (supposedly two, but I like headroom). I am, however, told that it isn’t cheap. This is available through Telecom.
A friend of mine is reporting that italk’s service works fine from Peru, and he’s using it to make cheap calls to Peruvian mobiles while over there. But then, he’s a VoIP nerd [wave!] and has been using VoIP to call his Peruvian in-laws for the better part of a decade.
I suppose it all comes down to how much you are willing to spend. It will be interesting to see what the charges on the Southern Cross do when Telstra gets their own cable online next year.
Oh, and the hardware cost for HD video conferencing? http://blogs.zdnet.com/Ou/?p=207 US$12000 per endpoint.
Telecom’s moves this week won’t have any effect on the Packet8 problem, which may be due to lack of bandwidth, or possibly even a fundamental problem with the protocol used by Packet8. Nor will it change the fact that carriers are more than likely breaking regulations by selling phone service to people without valid local addresses. For example, here in NZ a VoIP carrier isn’t supposed to sell you an Auckland number if your address is in Wellington.
Netco is an interesting development. I’ll crawl back into my hole and wait to see what happens!
Jason
Likeminded, I think you are making some assumptions here.
Firstly, choosing a good ISP is easier said than done. Once businesses commit to an ISP they can be reluctant to change and as Rod said earlier, all he wants to do is run his business. Even those in the IT industry are not checking every month what changes are happening. And changing is not straight forward for a business; its different for someone sitting on the end of an ADSL connection playing online games or frequenting chat rooms.
ADSL as example is saturated and suffers from high latency. This is true if no money is going in on the backend for more resources. Lots of people bought wireless services which also has high latency. Companies buy a service for Internet access, email, FTP etc, then other services spring up later like VOIP and video conferencing and the service you bought for email and Internet access turns out to be not so good. The obvious answer is fibre but it is expensive that I know from personal experience.
Secondly, this blog is all about giving and taking, information is shared in both directions. Sometimes strong opinions are stated. I don’t know anyone in the IT industry who shares real life business lessons for IT businesses more than Rod Drury. No one, period. He could easily do what a lot of blogs do and turn comments off and turn it into a personal soap box but he hasn’t.
While I have disagreed with some of Rod’s proposals I would 110% agree that the Internet infrastructure is at the moment running on fumes; it needs a gas injection quick.
Sigh.
The issue here for us likeminded export focussed folk is the international pipe.
International costs DOMINATE an average ISP’s model and therefore the price to the consumer.
Whilst your proposal is lovely in it’s intent, it *does not solve the problem*. Nationalising the national infrastructure simply makes the lanes to where the bottleneck is wider, therefore exacerbating the bottleneck/congestion problem.
Yes, in theory funds released from national costs can be redeployed to assist international transit costs, but in my view, given the spend associated with either, is just not enough, and there’s too much eggs broken to make the omelette (which doesn’t satisfy the appetite anyway).
Dermott, choosing the right ISP, or changing to the right one is far less damaging for national productivity than radically changing how the industry works in NZ.
“We were on CityLink so I assume a pretty good. connection. It did not work.”
“I assume”. Good God.
Do you understand that Citylink had nothing to do with this? Do you understand it was your ISP? Do you even remember who the ISP was? Do you remember the service you signed up for? How much were you paying for the international aspect required to deliver this “essential service”? You were prepared to pay a foreign company for a seemingly critical service (enough to turn the industry upside down for), but how much were you prepared to pay to get access to it?
I’m sorry I’m getting somewhat heated over this, but sheesh, you should know what the problem is before you come up with the answer.
Frankly I think the Xero strategy is a far better plan. Provide applications in NZ that other countries want to pay for. Therefore importing dollars into the local economy. The more dollars the Xero’s of this world bring in, the more prepared they are to pay for the transport that brings those dollars in. That makes far more sense to me. Applications Rod. You know about that stuff.