Before you post another message about fast broadband, read this...

Forums › New Zealand Broadband › Before you post another message about fast broadband, read this...

1 | 2 | 3

graemeh

2078 posts

Uber Geek

#400839 5-Nov-2010 16:12

Beccara: There is alot of fibre out there but not enough to say "Almost anyone in cities can get it", The minute that fibre doesn't ran on your side of the street your talking four figures to hook it up, Fibre might be around but its not competitive or wide spread

You continue to miss the point. You CAN get fibre. It's just that for you personally the cost outweighs the benefit.

Talkiet

4793 posts

Uber Geek

Trusted

#400841 5-Nov-2010 16:15

Beccara:

When I lived in Oriental Bay in Wellington I didn't have fibre, Some of my clients had fibre run across the street from them and would have cost $2-6k to have it run across the road, Only a handful of people I know in Wellington could get it and that was living in central city apartments. There is alot of fibre out there but not enough to say "Almost anyone in cities can get it", The minute that fibre doesn't ran on your side of the street your talking four figures to hook it up, Fibre might be around but its not competitive or wide spread

I never said it was free, or even cheap! For a business that _requires_ the benefits of fibre over other more pervasive access types, $2-6k is probably a reasonable amount.

I would agree COMPLETELY that for someone wanting a cheap $2-300/month internet connection, $6k will seem very expensive.

If what people are pining for is virtually free fibre hookups and then want to pay $50-$100/month for the service, then we do have to wait for someone to subsidise the rollout (UFB).

No-one's ever been able to give me a good reason why any company should invest a huge amount of money for a small and unguaranteed uptake by extremely cost conscious customers with cheaper alternatives available that are virtually as good for most needs.

Cheers - N

Please note all comments are from my own brain and don't necessarily represent the position or opinions of my employer, previous employers, colleagues, friends or pets.

wjw

162 posts

Master Geek

#400846 5-Nov-2010 16:19

Talkiet: No-one's ever been able to give me a good reason why any company should invest a huge amount of money for a small and unguaranteed uptake by extremely cost conscious customers with cheaper alternatives available that are virtually as good for most needs.

And this won't change until there is a killer app or someone only wants a 30 year return on their investment, which doesn't happen often.

Talkiet

4793 posts

Uber Geek

Trusted

#400852 5-Nov-2010 16:26

wjw:
Talkiet: No-one's ever been able to give me a good reason why any company should invest a huge amount of money for a small and unguaranteed uptake by extremely cost conscious customers with cheaper alternatives available that are virtually as good for most needs.

And this won't change until there is a killer app or someone only wants a 30 year return on their investment, which doesn't happen often.

Or (to be fair), until someone with a large Pot-O-Gold (tm) decides it's a public good and donates enough money to make the business case stack up for private partners.

Personally I strongly support the idea of government funding of a step change upgrade in the infrastructure to get around the chicken-egg issue of fibre uptake. I have some reservations about the specific process and $$$ involved.

It's exciting to think that NZ may get a huge jump start in a pervasive fibre network, but in my personal opinion, it's a very risky investment right now, and it's not been as fully evaluated as I would have preferred given my taxes are helping pay for it.

(specific note - the above all represents my personal opinion as a taxpayer)

Cheers - N

Please note all comments are from my own brain and don't necessarily represent the position or opinions of my employer, previous employers, colleagues, friends or pets.

Beccara

1469 posts

Uber Geek

ID Verified

#400853 5-Nov-2010 16:27

Talkiet:
Beccara:

When I lived in Oriental Bay in Wellington I didn't have fibre, Some of my clients had fibre run across the street from them and would have cost $2-6k to have it run across the road, Only a handful of people I know in Wellington could get it and that was living in central city apartments. There is alot of fibre out there but not enough to say "Almost anyone in cities can get it", The minute that fibre doesn't ran on your side of the street your talking four figures to hook it up, Fibre might be around but its not competitive or wide spread

I never said it was free, or even cheap! For a business that _requires_ the benefits of fibre over other more pervasive access types, $2-6k is probably a reasonable amount.

I would agree COMPLETELY that for someone wanting a cheap $2-300/month internet connection, $6k will seem very expensive.

If what people are pining for is virtually free fibre hookups and then want to pay $50-$100/month for the service, then we do have to wait for someone to subsidise the rollout (UFB).

No-one's ever been able to give me a good reason why any company should invest a huge amount of money for a small and unguaranteed uptake by extremely cost conscious customers with cheaper alternatives available that are virtually as good for most needs.

Cheers - N

Apologies, I misunderstood you're point. You right that if you want to pay for it it's there, As for why they should invest? Look at where we are heading, It used to be dialup was fine for what users did, Then they wanted to watch some guy fall off a bike on youtube and e-mail around their 10MP camera photos quickly, Soon they will want to be able to watch full movie on a whim from their TV, Not just Sky's hand picked ones but any movie they can see on Amazon. Full HD video calling to the kids on their O.E.

wjw

162 posts

Master Geek

#400859 5-Nov-2010 16:33

Talkiet:
wjw:
Talkiet: No-one's ever been able to give me a good reason why any company should invest a huge amount of money for a small and unguaranteed uptake by extremely cost conscious customers with cheaper alternatives available that are virtually as good for most needs.

And this won't change until there is a killer app or someone only wants a 30 year return on their investment, which doesn't happen often.

Or (to be fair), until someone with a large Pot-O-Gold (tm) decides it's a public good and donates enough money to make the business case stack up for private partners.

Personally I strongly support the idea of government funding of a step change upgrade in the infrastructure to get around the chicken-egg issue of fibre uptake. I have some reservations about the specific process and $$$ involved.

It's exciting to think that NZ may get a huge jump start in a pervasive fibre network, but in my personal opinion, it's a very risky investment right now, and it's not been as fully evaluated as I would have preferred given my taxes are helping pay for it.

(specific note - the above all represents my personal opinion as a taxpayer)

Cheers - N

I agree with all of that, the FTTH trials I have been involved certainly don't inspire confidence regarding future proofing and scaling.

stuzzo

534 posts

Ultimate Geek

#400884 5-Nov-2010 17:22

Billions of dollars are being spent on roading improvements in Auckland. I'm sure someone has had to produce a business case for them but really it would be highly subjective. The money is being spent because Aucklanders have demanded it through the polls.

As a rural dweller I won't see the benefits of this or FTTH unless it flows through to the rest of the economy.

As far as getting a jump start on the rest of the world I suspect that chance is long gone. We are really jumping up and down to reach some sort of parity with what will inevitable be the standard around the globe.

Sharesies

Trade NZ and US shares and funds with Sharesies (affiliate link).

webwat

2036 posts

Uber Geek

Trusted

#400910 5-Nov-2010 18:45

Talkiet:
wjw:
Talkiet: No-one's ever been able to give me a good reason why any company should invest a huge amount of money for a small and unguaranteed uptake by extremely cost conscious customers with cheaper alternatives available that are virtually as good for most needs.

And this won't change until there is a killer app or someone only wants a 30 year return on their investment, which doesn't happen often.

Or (to be fair), until someone with a large Pot-O-Gold (tm) decides it's a public good and donates enough money to make the business case stack up for private partners.

Personally I strongly support the idea of government funding of a step change upgrade in the infrastructure to get around the chicken-egg issue of fibre uptake. I have some reservations about the specific process and $$$ involved.

It's exciting to think that NZ may get a huge jump start in a pervasive fibre network, but in my personal opinion, it's a very risky investment right now, and it's not been as fully evaluated as I would have preferred given my taxes are helping pay for it.

(specific note - the above all represents my personal opinion as a taxpayer)

Cheers - N

Looks like you bought the Telecom philosophy as well as just working for them. This stubborn aversion to open access next-gen infrastructure has put Telecom at a great disadvantage instead of ahead of the industry. Telecom should have been ideally placed to lead the move to open access, but has stayed with the idea that its focus should be on managing regulated services such as UCLL to prevent offeing its customers more than necessary.

That article was quite biased, essentially sets up a straw-man instead developing realistic assumptions. Who would possibly need 100Mbps internet to support a single HTTP user as implied? FTTP is more than just fibre to houses, its connecting the premises and could be Multiple Dwelling Units or businesses. Even a single house these days can be expected to include several computers plus various connected devices, and lots of multitasking. So a better example of how FTTP service would be used would include an HTTP session, a video on demand download or online game, an IPTV service, and a VoIP call all happening concurrently. ADSL does not perform well in this multiuser environment, partly due to international bandwidth and also due to the lack of speed and QoS between DSLAMs and handover points.

Ask your Telecom people how much bandwidth is provisioned to the average ISAM per user, and how much allocated per user to the handover points. FTTP will perform quite well at this without need for interleaving and with capacity in an open access business model being more responsive to demand.

There is also the question that should be asked of the government: what is meant by "uncontended 100Mbps" on a Layer 1 physical network, which by definition has no data or speed at all (so there is only a "split ratio"). The contention happens on backhaul speed eg 10Gbps/20Gbps/40Gbps uplinks, compared to the speed of connected services. A connected service could be lots of 100Mbps or Gbps point-to-point Ethernet but most will be G-PON, upgradable to 10G-PON. Shared 2.5Gbps G-PON should easily average 100Mbps per user most of the time because its burstable. The optical split ratio (Layer 1) is less important than the backhaul contention (Layer 2), but MED specified contention didn't include the backhaul. An access node might service 48 PONs or double that, so low split ratios demanded by CFH will just add costs and force providers to fit more PON cards onto bigger chassis and probably average 20:1 contention for advertised 100Mbps userland services, one way or another. Of course there will hopefully be some providers able to differentiate with better speeds vs cheaper service so users can choose between cost and speed.

It would be really disappointing if the new Pacific fibre link comes online and nobody has a fast enough connection to use it much.

Time to find a new industry!

webwat

2036 posts

Uber Geek

Trusted

#400973 5-Nov-2010 21:26

Another mistake in that article:

"Every time the signal is boosted, it is converted from light to electric signals, put through electronics and converted back into light for the next leg of the journey."

I believe Southern Cross has a few repeaters, but this will not be the case for any new international cables such as the one being planned. New trans-pacific cables, and perhaps any upgrades to SC, have EFDA amplifiers that boost the optical signal without doing any conversions. Some of the complicated stuff they do to compensate for the various kinds of issues are fascinating but I just know future links will be much lower latency.

Time to find a new industry!

freitasm

BDFL - Memuneh
79290 posts

Uber Geek

Administrator

ID Verified

Trusted

Geekzone

Lifetime subscriber

#400983 5-Nov-2010 21:58

stuzzo: If he has Telecom in his signature I would have thought it would imply that he in some way the sanction or instruction of his employer to speak on their behalf.

If not why have it there. Most people work for someone.

Hmmm... Not exactly. It only means that if someone around here needs help and he approaches this person asking for an account number to be sent via PM then we trust him to be someone that can receive this information.

I will make it clear: having a company number under a user name here on Geekzone doesn't make that person a spokesperson for that company.

The only exception here was PaulBrislen (who now left Vodafone New Zealand and joined TUANZ as its new CEO), and accounts with company name, such as Slingshot, Orcon, TelstraClear.

cbrpilot

955 posts

Ultimate Geek

Trusted

Spark NZ

#401133 6-Nov-2010 13:37

I found the article very narrow sited. Yes international latency is an issue, and a difficult one to solve (bit hard upearthing NZ and moving it closer to the rest of the world).

The article only looked at the effects of latency on HTTP of popular websites. Yes it certainly does decrease the average throughput of loading that web page - mainly due to the fact that each webpage is comprised of hundreds of small files which must individually be transported. The latency of requesting each and every file (together with that added by an DNS look-ups) creates this issue.

That's fine. We've learned that if you live in Invercargill that your average Facebook page may take longer to load. Big deal I say.

Web pages are not the killer that one needs fat pipes for, and not what I would imagine that the Govt had in mind when deciding to launch UFB. The killer apps for UFB will no doubt be different to simple web surfing, and need not suffer in the same way. E.g. let's imagine for a second that streaming media was the killer app for UFB (be that video conferencing of some other VOD-like service). Latency for these types of applications is not really too critical - you're transferring a single file, and once the TCP connection has stablised (window sizes more or less stable), then the latency makes very little difference. So long as the window size within TCP is big enough (and the files are big enough or reuse the same TCP stream), then there need not be this big tension between bandwidth and latency.

I'll say it again, the article view was very narrow, and should not be extrapolated to mean that all types of application will not hugely benefit from increased bandwidth.

My views are my own, and may not necessarily represent those of my employer.

Screeb

698 posts

Ultimate Geek

#401220 6-Nov-2010 20:04

cbrpilot:
The article only looked at the effects of latency on HTTP of popular websites.

Web pages are not the killer that one needs fat pipes for, and not what I would imagine that the Govt had in mind when deciding to launch UFB.

I'll say it again, the article view was very narrow, and should not be extrapolated to mean that all types of application will not hugely benefit from increased bandwidth.

This basically sums it all up. Who has ever talked about regular web browsing speed when talking about FTTH? This paper is just a ridiculous red herring. No one who knows anything about how TCP works ever suggested that web browsing would improve sufficiently to care about - besides which, who has ever complained about regular sites loading slowly on properly-performing ADSL2+ (or even decent speed ADSL for that matter)? I'm on 15Mbps cable and most pages load as fast as I would ever want or need. The only ones that don't are because of a problem on the site. I don't want faster broadband so that web pages load faster, I want it so that I can download and upload files faster.

Oh and it should also be noted that, as well as Telecom being a member of ISCR (as has been mentioned), ISCR itself is apparently dedicated to writing about how regulation is terrible, and in particular, one of the authors of the paper in question, Bronwyn Howell, has a nice track record of defending Telecom and being incredibly anti-telecommunication-regulation (despite apparently not having any formal education in IT). The other author, Mark Obren, is even more interesting. He has written publications arguing for the commercialisation of the national road system. Also, the paper "acknowledges" "the helpful comments" of Anton Nannestad, who works for Telecom as a regulatory economist. I don't think listening to any conclusions any of these people make in regards to national infrastructure and regulation is particularly... useful.

1 | 2 | 3