Chilling_Silence's blogThe current / future state of broadband in NZ
There’s 3 types of people in NZ:
1) Those who are actually IN the broadband business and know what’s going on
2) Those who are spectating from the sideline and *think* they know what’s going on (But usually don’t)
3) Those who just don’t care
After a recent article in the New Zealand Herald, I had a good rant on some forums about how misguided a lot of what’s being written in the media about the whole Govt broadband rollout is, specifically I’ve noticed NZ Herald is a pretty bad repeat offender, and it’s not just one journalist (It’s a few).
Some of the articles are just outright stupid: http://www.nzherald.co.nz/hamish-fletcher/news/article.cfm?a_id=702&objectid=10705236
Adding a little perspective and anybody can see that Korea is a lot more highly dense with its population. New Zealand has only 4 million people spread across 270,000 sq/km. Korea has 70 million spread across 220,000. That’s a *massive* difference, and that’s precisely why Korea is able to run Fibre to an apartment / building complex, because the economies of scale (20x the number of people and prospective customers) means that where an ISP Â may run a fibre cable to a street in Korea and provide connectivity for several thousand, here in NZ that same street might only provide internet for a couple of hundred.
We’re a great example of “the little country that could”! The fact we’re going to be #10 is absolutely awesome if you ask me, and we would be foolish to be unhappy with it, or to think we should be higher up the chain.
So anyway, the state of broadband:
Right now, you can probably get Fibre to your home. I could.
It’d cost me $1,500 to get setup and $1,100 a month, but it’s possible.
There’s already a fair bit of competition in the market, so adding one more competitor (The Govt + Partner company) would just spread the already thin customers even thinner.
$1,100 per-month would get me 1 megabit. Throwing another $200 at it a month, taking the spend to around $1,300 a month and get 10 megabits per second. Considering I’ve done work for companies that have fed 200-odd staff from a 2 megabit pipe, that’s some serious internet that I could get for $1,300 a month! That’s with no bandwidth cap either, so I can send / receive as much as I like.
However, if I look at getting something like HSNS, which is a symmetrical (Upload is as fast as your download), over copper wire (Your existing phone line), I can save myself between 50% and 75% of the monthly cost vs Fibre.
Why so expensive?
Well it costs money to lay these new Fibre cables in the ground. They’re not cheap, not to mention the new equipment needed to service them (Switches, routers, other fancy things I won’t bore you with). Add in to the fact that everybody sees it as a cash-camel because it really is a “premium” service, and you have a recipe for disaster.
When it comes to phones, the cables are already in the ground. They’ve been in use for ages and paid for themselves many times over.
So, your connection to the Internet is mostly paying a “rental” of the line, and for the data that you’ll be consuming.
Data? Don’t they get it flat-rate?
Well yes, it’s flat-rate, but here’s how data works for the bigger players:
They buy based on the amount of concurrent data that they can transfer. For example, they might buy a 1 gigabit connection nationally, and a further 500 megabit (m/bit) connection worth of international data. International is always more expensive, whereas National (Within New Zealand – With the exception of TelstraClear and Telecom – They usually cost additional due to other politics) is predominantly very very cheap. Unfortunately most of the data NZ consumes is international.
Now, lets say that this fictional ISP that we’re talking about has 500 m/bits of International bandwidth. Lets say an average customer can connect at home at 5m/bit. That means they could have 100 customers utilizing their whole connection. Thankfully it doesn’t work like that.
It averages out so that your connection is using less data than you would on dial-up 95% of the time. When you load a web page, all that a faster connection will do is help bring the information to your computer faster. While you’re reading whats on the page (Such as this blog), your connection will be pretty much idle which means that others can utilize the data that you’re not.
So naturally providers over-subscribe their connection, mostly do-so quite safely, and why shouldn’t they! Some do it better than others, and that’s when you get issues with the likes of Xnet and Slingshot giving people constantly poor speeds. It’s because they don’t have enough bandwidth to cater for their customer-base.
So where are we being screwed?
Well service providers have a lot of overheads, like a Helpdesk to pay for, Billing / Accounts departments, higher level tech support teams, and they also have to pay a fee to Telecom for the use of their lines for the internet. It doesn’t help that New Zealand has an “expectancy” to receive a free modem when you join a new internet service provider which also bumps up the cost a bit of your monthly bill too. No wonder we have so many terrible cheap and nasty routers floating around, which funnily enough those same routers only end up causing people and internet providers more headaches.
The price of data for most ISPs has gone up to $2 per-GB over and above your base plan. Some of the nicer (Better?) providers only charge around $1 per-GB over your plan.
Do they need to charge you $2 per-GB? Probably not, if they’re managing things right, and if they’re big enough (Or reselling from somebody who is big enough).
See, something I’ve learned over the last 18 months is that the quality of an Internet Service Provider can change very drastically in the space of 3 months. Slingshot ran their unlimited plan and within 3 months it turned things to custard for most of their customers. They were under-resourced both in terms of raw staff, staff skill / training, and internet capacity.
There’s a very good reason that some relatively unheard of ISPs such as Actrix win ISP of the year over and over and over again. Because they know how to do things right, and they manage things well. They grow at a good and stable pace, they look after their network, and have enough bandwidth to take care of their customers.
What does that mean? Whats going to happen then with the Govt Fibre rollout?
Well this “Ultra Fast Broadband” (UFB for short) is going ahead. Most of NZ already can get Fibre at 100m/bit if they really wanted, but it’s just not affordable. Up-front as well as ongoing.
Here’s my prediction:
Not much will change!
Or, better still, where the changes happen, they’ll happen before the rollout is mostly completed, but just as part of natural progression.
VDSL2 is still a very much work-in-progress with Telecom (For reasons they’re not letting on, though speculation is due to the UFB rollout). Then theres HSNS (Which uses the SHDSL standard) which can also use your existing phone cable. This makes it a hell of a lot more affordable than Fibre and it’s more reliable than your current ADSL broadband.
ADSL2+ which I predict most suburban people have, should improve quite significantly to keep the “masses” happy, however it will always be a “best effort” service that’s semi-flaky, and some ISPs will still abuse.
HSNS will become more popular with businesses, as it’s offered by more ISPs, and word of mouth gets around about it. It’s ideal for small businesses and my prediction is that the price may drop enough for some high-end homes or the likes of gaming enthusiasts to also warrant having it.
VDSL2 will probably end up coming out also, but as a “premium” product. ISPs will be selective about who they allow onto it, they don’t want its name tarnished. It’s also going to be a bit more expensive than ADSL2+. At the end of the day customers will be able to choose between VDSL2 and HSNS for their connections, but I think if VDSL2 is priced right, then people will be more likely to jump on to that.
There will still be the higher-end Fibre. That’s going to drop a bit in price, but not by much. Fibre will always be the “Mac Daddy” of them all, that you go for if you want 100% reliability. More than likely a splattering of homes around the country will get it, but mostly it’ll be the medium businesses upwards that’ll end up getting it. Fibre will provide the backbone that everything else uses.
Data caps will start to increase slowly. There are some awesome ISPs out there that are going against the grain and seem to be doing things the “right” way. Actrix have the best customer service and highest customer satisfaction rating ever. They do that year in and year out. HD (Hosting Direct) have an amazing network setup, lots of bandwidth, and they manage their service very well so that you always have a fast and good quality connection. They’re not skimping on bandwidth at all, and talking with them they want to get more people onboard and using their National connection, to give them more pushing power to buy even more international bandwidth. Not that they need to at this point in time, but it’s great to see an ISP that takes things seriously and providing a top-notch quality and high-speed connection to their customers well.
That’s really going to be the separating factor between ISPs in the future, the ones who manage things well, and those who don’t. Currently people just put up with it (They shouldn’t, but people don’t know that better is readily available), but in the future people won’t.
QoS will also play a part. over 50% of the people I meet / know of have superb connections. The down-side is the QoS (Quality of Service), people need to have their traffic prioritized properly, and I’ve said before that I think that some of this will happen at the ISP-side. Your ISP will give you a portal that will let you as a customer manage if you want VoIP prioritized over say YouTube, or gaming prioritized over say email.
Overall, will the UFB rollout be a waste?
No, I don’t think it will be a total waste. It’s going to put in the backbone and infrastructure needed to provide these other cheaper connections.
It certainly won’t be the white stallion that the populace think it’s going to be, that the Govt will ride in on to save the day.
I’m beginning to think that the Pacific Fibre cable will be a non-event. The Southern Cross Cable is nowhere near capacity (Last I heard it wasn’t even at 50%), but the price of international data will still drop a little more. Not by much I’d think. It’s the ISPs that are milking it a lot there, from what I can tell.
Telecom is going to have an important part to play, with their copper telephone network across the country. As a business they’ve really changed quite significantly in the last 18 months or so, since around the time of their new CEO. It’s probably going to take splitting up the company though for New Zealand as a whole to see that, and stop with the “tall poppy syndrome”. For some reason people and journalists just seem to like to bitch about the big-guy coz they know that  they both can’t fight back against their badmouthing, and they’re an easy target.
The rollout is going to go ahead, it’s going to cop a lot of flak. It’s not going to be what people think, and people don’t actually need what they think they do. 100m/bit to your home? I’ve seen businesses happily run hundreds of staff off 1 to 2m/bit.
Don’t get me wrong, I’ve been on very high speed connections. It opens a world of possibilities around VoIP, off-site data and “cloud-based services”, it’s certainly an increase in productivity, but we’re not suddenly about to get pulled out of a recession by it… 95% of things that people do like Email, Facebook, browsing, chatting, video conferencing, things like that can all be done on their existing connections, but they’re just not currently managed properly both by the end-user and their ISP.
We’re in for one hell of a ride over the next few years to say the least! Lets hope the media keeps some perspective on things, rather than just over-hyping stories.
As always, I’d love comments, and I’m happy to discuss things constructively 🙂 I’m also more than open to being proven wrong
Cheers
Chill.
Leave a Reply