news Communications Minister Malcolm Turnbull has taken a pick axe to an article published by electrical and computer engineering academic Mark Gregory, claiming that the RMIT senior lecturer’s criticism of the Government’s new MyBroadband broadband tracker site constitutes a “confused and illogical” attack, and that Gregory had misunderstood the site’s rating scale.
Turnbull launched the MyBroadband website last month as the culmination of a significant study conducted by the Department of Communications into the availability and quality of broadband in Australia. The study was an election promise by the Liberal MP contained as part of the Coalition’s broadband policy unveiled in April 2013 ahead of last year’s Federal Election.
According to Turnbull, the report is the first of its kind to be undertaken by an Australian Government, with data drawn from all major Australian telecommunication carriers. It describes the broadband technologies available as well as the speed that can typically be achieved over each available technology platform.
However, in an article published last week on Technology Spectator, Gregory wrote that he gave the website an “F for fail”. In his article, Gregory cited a range of issues with the website, starting with the fact that it is based only on theoretical data and not primarily on real-world speeds measured by broadband speed testing applications; that it does not include dial-up or ISDN technologies, and that the rating system used to measure broadband availability on the site is misleading.
Some of Gregory’s criticisms appear legitimate. For example, the site gives areas an “A” rating for availability of HFC cable from Telstra and Optus, no matter whether the individual premises in that area are actually able to connect to the HFC cable infrastructure.
It is extremely common for both Telstra and Optus to refuse to connect residents and businesses in the HFC cable footprint if they reside in so-called multi-dwelling units such as apartment blocks, business parks or shopping facilities, because the two telcos generally require the entire facility to be connected in one go, rather than running the cable to individual premises.
“It appears that if a listed technology can be accessed from the property then it is listed as an ‘A’, otherwise listed as ‘not available’,” wrote Gregory. “If HFC goes past my street or stops 50 meters from my home do I get a ‘B’ or a ‘C’? There is no mention of the broadband availability scale on the website.” The academic pointed out similar issues with the rating system for the performance of the various technologies.
In addition, the final analysis of a crowdsourced comparison of real-world broadband speeds has shown that the MyBroadband site is significantly inaccurate, with most Australians receiving speeds more than 25 percent slower than those listed.
However, in a rebuttal published on his site yesterday, Turnbull strongly disagreed with Gregory’s article. “Mark Gregory has made a confused and illogical attack on the MyBroadband website. Gregory misunderstands the approach taken to identify areas with poor access to broadband services,” wrote Turnbull.
The Minister said the idea of local area ratings for broadband was important as the MyBroadband analysis did not attempt to break down individual data points for the millions of houses in Australia that have a fixed-line connection. “That would be possible – but it would take many years to undertake and wouldn’t be all that useful anyway,” he wrote.
“The whole point about broadband upgrades is that they occur on a neighbourhood-by-neighbourhood basis, not house-by-house.”
Turnbull stated that “Gregory’s assault on logic reached new heights” when the academicclaimed that an area with any level of access to Fibre-to-the-Premises should receive an ‘A’ rating for quality.
“The problem with that approach is that as soon as one house in a local area or Distribution Area (which typically contains 200-1000 households but sometimes as many as 5000 households) receives an FTTP connection, then the whole area’s quality would be assumed to have improved,” said Turnbull. “In other words, hundreds of households would receive a higher quality ranking (and hence would be seen as being in no need of prioritisation) just because one house was getting a great service.”
“So the broadband quality rating is dependent on the availability of each technology within a local area – and this was made quite clear in the broadband report (available online here in PDF format) which explains how the rating system works.
Turnbull stated that it was possible, for example, for an area to receive a rating of “C” for Fibre to the Premises availability (the best available category of broadband) because FTTP could only be available for 40-60 percent of the geographical footprint in a certain area, for example, with ADSL (available to 80 to 100 percent of premises, for example) delivering an estimated median speed for the area of 16.46Mbps, for example.
“The overall quality rating is based on the combination of the highest quality service that is available to each premises in the area,” wrote Turnbull. “I understand this rating system is complex. But that is not a reason why it should be dropped altogether, as Gregory suggests. An alternative course of action would be for academics explaining complex policy issues to actually read detailed reports produced by the government before bagging them.”
opinion/analysis
So who’s right here? Both are right (or alternatively, both are wrong), depending on which way you look at it.
If you consider the overall purpose of the broadband availability and quality report promised by the Coalition prior to last year’s Federal Election, it was never to provide comprehensive information to Australians about the availability of broadband at their specific premises. The point of the exercise was actually to give policymakers such as Turnbull a tool to show, on a region by region basis, where Australia’s broadband was most in need of urgent upgrades, so that these areas could be targeted by the Coalition’s Broadband Network rollout.
Under Turnbull, however, as is often the Minister’s wont, the report has morphed into a populist tool which aims to give individual Australians an idea of what they can expect from their premises’ broadband availability.
Gregory’s right — the MyBroadband site is extremely misleading when it comes to ranking the broadband of individual premises. Especially when it comes to HFC, it’s a joke. It’s cold comfort indeed to know that your premise has an “A” rating for HFC cable, if neither Telstra nor Optus will connect the damn thing. And the MyBroadband v Reality folks are also right — the system mainly uses theoretical estimated data, so it doesn’t match up with real-world speeds.
However, Turnbull’s also right: The tool does provide a rough, region by region guide to the quality and availability of broadband in Australia. It is a very useful tool for policymakers.
The fault here is in the marketing of MyBroadband. This was not intended as a consumer-focused tool, and it should not be. It should remain what it was designed for — as a useful tool for policymakers to measure broadband availability and quality at a gross level across Australia. The Minister’s populist streak led it into being something more — and thus to the inevitable criticism as it has failed to successfully fulfil a task for which it was never originally intended. Australians would not be trying to measure their broadband speeds through MyBroadband if the Minister had not told them in a nationally televised press conference that they could.
Image credit: Parliamentary Broadcasting, RMIT
LOL @ Mal.
There is only one fool here Mal and it’s not Mark, can you guess who it is??
Someone should get Mal a bigger shovel make the hole he’s creating for himself easier to dig.
GENIII
I think there might be a pretty predictable trend here:
1. CBN does some wholesale analysis
2. Some wonker in the Minister’s office thinks of a public spin angle to get some ‘look how good things are with the MTM model’
3. Big Mal goes public.
4. Anyone of 25,000+ people who actually know how this “stuff” works shoots Mal down
5. Mal plays the man and not the ball and tries to cite personality or mental deficiencies in the person from step 4 in a way to create an “Us vs Them credibility stoush”.
6. Rinse and repeat.
“The problem with that approach is that as soon as one house in a local area or Distribution Area (which typically contains 200-1000 households but sometimes as many as 5000 households) receives an FTTP connection, then the whole area’s quality would be assumed to have improved,”
Yet mybroadband uses that very same measure for HFC availability. There are no details in the report on how the figures are actually reached only a list of factors taken into account. When we compare this to actually measurements and mybroadband is consistently coming up short it make the whole thing like a load of marketing*.
*A load of marketing is like a load of feces but isn’t even useful as fertilizer also I guess you could print the marketing out and then compost it.
“. It should remain what it was designed for — as a useful tool for policymakers to measure broadband availability and quality at a gross level across Australia.”
It’s not even useful for that, given the insistence on relying on poor quality existing infrastructure. The lack of real world speeds mean that they can assume they’ll get something decent if they drop new infrastructure into a given area but then look all confused when the copper refuses to operate at anywhere near the minimum speeds.
Yet more ad hominem from a federal minister. That’ll earn our respect – not.
The site might be less misleading if it didn’t ask for a residential address. Some vague general locality might be more appropriate for the data returned.
Exactly. But it does ask for specific location.
““That would be possible – but it would take many years to undertake and wouldn’t be all that useful anyway,”
It would be possible and wouldn’t take very much effort to collate precise data for active internet users…
http://www.myfraudband.com.au/ SPEEDTEST and it would be very useful indeed.
The user has entered their address, so gather REAL DATA. This would be in line with Malcolm’s previous statements/non-core facts/(whatever his utterances really are), Real world data … not theory … talk to the men on the ground…. How about collect real data?
Of course a request along the lines of “To be accurate, please retest several times during the day and week.” Should be incorporated, with number of tests denoting the accuracy of the results {“test when its raining and click the checkbox Rain}. If Malcolms pet team is incapable of implementing such a test, then I am sure many of the ISP’s would be happy to contribute their knowledge.
Of course sites like speedtest.net already have this data on an IP basis, and have probably avoided address information due to privacy. Malcolm’s site has no issue with asking for address, and is probably collating metadata on what locations are looking at the site. Why would he not take the opportunity to gather thats precious Real World Data?
Now we just need a technical celebrity to ask why he isn’t taking the opportunity to test real speeds via MyBroadband, and to be subject to a personal attack from Malcolm.
MT, as politicians do, is playing both sides of the fence.
On one side he says:
“The whole point about broadband upgrades is that they occur on a neighbourhood-by-neighbourhood basis, not house-by-house.”
On the other side he says:
“The problem with that approach is that as soon as one house in a local area or Distribution Area (which typically contains 200-1000 households but sometimes as many as 5000 households) receives an FTTP connection, then the whole area’s quality would be assumed to have improved,”
If it was the case that one house, or very few houses, was able to be served by FttP but the others in the same “neighbourhood” weren’t able to be served by FttP then shouldn’t the “neighbourhood” boundaries, for the sake of the MyBroadband site, be changed so that service availability is more likely to be consistent throughout. Maybe match it up to which exchanges these neighbourhoods are connected to. That way, if ADSL2 is available to one house, then it’s highly likely to be available to a majority of houses in that neighbourhood, thus getting rid of the “one house connection skewing the average” problem.
In regards to MT’s second statement that I put above, I’d argue that if one household had access to FttP then it’s likely that others in the neighbourhood also have access to FttP and therefore, yes, the whole neighbourhoods quality has improved.
However, I also think that Mark was quite harsh and pedantic in his analysis, but I’m sure he wasn’t misstating facts at any point, so it’s all good.
If it a policy making tool, why the hell is it marketed towards the general public? I mean even the name “MyBroadband” is misleading if it isn’t for looking up my broadband speed.
I think you’ll find that being a “policy making tool” is a prerequisite for a seat on the front benches these days.
If my area is any indication then myBroadband is rather useless as a “Policy Making Tool” considering he originally stated that his policy was to prioritize the poorest areas of service.
My area’s stated 7.3Mbps average & good mobile rating is a joke.
No ports, so only one house on 1.2Mbps ADSL2, & 1 bar mobile at best in our street.
Real life example of how it gets it wrong. My area in Wollongong shows my overall fixed broadband availability to be an A. Quality lists as an E. In the quality section, it lists my ADSL quality as a B, and the only option available. The estimated median speed is 18.69 Mbps.
No problems there, I can understand where those details come from. We only have ADSL, and that speed is roughly the highest available in the area. But the little chunk of Wollongong I live in has one basic flaw – its the other side of the street that gets the 18 MBps. On my side, the copper winds around the long way, ultimately meaning I have around 6 Mbps.
So, the chunk has 18 MBps, my house has 6 Mbps. And because of how they have divided the area up, my side of the road is ALWAYS going to be misrepresented. My worry is that in futureworld, where the LNPMTMNBN is in play, their determination of line quality will be based on this median measure, rather than the real world situation.
Made all the more annoying that the northern end of my chunk is where the FttH rollout currently ends. 100 meters up the road, they get 100 Mbps/1Gbps/whatever, but where I am right now I fear that if we get FttN I’ll get 6 Mbps. Permanently.
I don’t even.. What?
Website only for policy makers? Why was it advertised and targeted at normal Australians then?
No the website had one purpose: Propaganda.
For those that have no clue about technology will look at it and go “Look here, the government says our internet is great! Why do we need this fandangled fibre for anyway, I get plenty in my weetbix!”
Gamertech posted this comment on your other article http://delimiter.com.au/2014/03/03/real-speeds-25-percent-slower-turnbull-mybroadband-tracker/
I think the comment is awesome, and something that Turnbull should listen to:
Gamertech basically said:
[quote]
From Turnbull’s response in the Business Spectator (BS for short)
“An alternative course of action would be for academics explaining complex policy issues to actually read detailed reports produced by the government before bagging them.”
An alternative course of action would be for politicians explaining complex technology issues to actually read detailed reports produced by the industry before bagging them.
See how that works Turnbull?
[/quote]
It’s a perfect comment to go with what Turnbull said about academics i think.
Of course it doesn’t help that detailed reports don’t get released the report for mybroadband is vague and none specific.
I have thought of politics as generally harmless…
But the current NBN watering down will damage Australia’s
future more than anything else I can think of.
We need faster internet for many reason, most of which we don’t even know yet.
But I thought of one current reasonable example:
I am lucky enough to have a stable 100Mbps cable connection, but
it is not fast enough to watch 360deg HD streamed…
eg. http://visualise.com/videos/mercedes-f1-360-video
Plus 4HD is coming soon.
Plus my wife and kids will want to watch something different.
Rough calculation 2000Mbps minimum requirement.
Google are rolling out 10000Mbps this year to the US.
Australia is being let down by a non-upgradeable, last century NBN,
which by the time they finish lying and watering down will be lucky
if 80% of people get even 10Mbps.
India and south east asia will take all the jobs, their Internet infrastructure
is going to eclipse ours. Our economy will head towards 3rd world proportions.
http://www.minister.communications.gov.au/malcolm_turnbull/news/mybroadband_site_identifies_australiaas_broadband_blackspots#.Uw03UPR99Yg
“Error 404: Page cannot be found”
Looks like Malcolm’s been waiving his “transparency” wand around again…
The LNP sure like reinventing/hiding history.
Cached copy of the missing page is here: http://webcache.googleusercontent.com/search?q=cache:http://www.minister.communications.gov.au/malcolm_turnbull/news/mybroadband_site_identifies_australiaas_broadband_blackspots#.UxadbvS1a6g
The MyBroadband site states that Adsl is available out of DA 31 Tweed Heads ,when in actual fact it is not available as it is too far from the telephone Exchange.
The fact that Adsl is not available was confirmed by Telstra using an address next door to the Pillar(DA31).
The concern here is that this area shall not be included in the 700.000 people that Turnbull has said he will help as the site say’s they can get Adsl.?
I live next door to you Norton in Tweed Heads West (Gray St). The MyBroadband site says that in our green area we have ADSL but we have NO ADSL at all, only mobile. How in hell did we get an “A” for availability when all we have is mobile is beyond me.
OK, Mr Turnbull we get it, everyone else is wrong and your right.
Perhaps being the odd one out makes you the foolish one.
This is my comment on Malcolm Turnbull’s website (awaiting moderation):
Your last comment is not helpful, & I agree with others could easily be rephrased as:
“An alternative course of action would be for politicians explaining
complex technology issues to actually read detailed reports produced by
the industry before bagging them.” (Yes, I have read the full NBN Strategic Review)
On Quality Ratings, you are far too generous, my suggested ratings:
A Plus: NBNs 50/20, or 100/40 currently on offer over FTTP (Upload is critical)
A Minus: Telstra HFC (100/2Mbps) where generally available.
B (minus): Optus based HFC.
C: FTTN / VDSL2+ (assuming copper lengths under 800m)
D: ADSL2+ is too variable (My ADSL2+ used to sync at 7/1Mbps).
Also on ADSL figures your estimated speed numbers should be reduced by 25% to account for real world conditions not lab estimates based on heavier gauge (thicker), newer, copper in good condition.
I currently pay $110/ month for Telstra Bigpond 100/2Mbps
Exetel sells 100/40Mbps NBN for $70/month, 20x faster upload@$40/month less.
I would happily pay >$100 for even 50/20 Mbps, my current upload is too slow.
I realise I’m just another single voice, and I absolutely understand your desire to reduce capital cost, speed up rollout and thus revenue flow, but the long term revenue of the MTM (Option 6) will be significantly lower than the previous governments plans (Option 1 or 2).
http://www.nbnco.com.au/content/dam/nbnco/documents/NBN-Co-Strategic-Review-Report.pdf
Would the government please seriously dumping FTTN from the MTM and:
– Stick with the previous satellite and fixed wireless coverage areas
– upgrade and infill HFC where available (to speed up deployment and revenue)
– replace everywhere else with FTTB or FTTH
(TPG & Telstra have offered to build FTTB, how about a deal whereby NBN can resell their services at a pre agreed rate?).
FTTN is a waste of money as it will be obsolete before it is build and/or overbuilt by other telecommunications companies with FTTB and not used.
Thanks in advance.
FTTN is not a waste. I am currently in NZ with VDSL connection and it works right now. And I am not the only one. How much of a waste is that? Wish people would stop the constant bickering in Australia and actually do something. NZ bit the bullet many years ago and has achieved what they set out to do with FTTN. And that has not excluded FTTH either.
My Australian ADSL is severely crippled with a miserable 1Mbps upload speed. That is 5x slower than what I currently have in NZ right now, not tomorrow, not some indefinite time in the future.
Interestingly, NZ has also replaced all of its wooden power poles with concrete ones. Wake up Australia!
If it was available TODAY it wouldnt be a waste. But its not, and for most people, wont be until some time around 2018-2019. At which point we’re staring in the eyeballs of needing more than what FttN can practically deliver.
A constant problem with starting FttN now, rather than 5 or 10 years ago like it should have. If Telstra and friends had rolled it out in 2005 like other countries, investing more in it today would make much more sense, but because they didnt (outside a few areas like TransACT), we’ll be investing $10’s of billions in technology and be stuck with that technology way beyond its use by date just to justify the expense.
You dont invest $30b in something then replace it 5 years later.
Agree completely that FTTN made a lot of sense 10 years ago when Telstra proposed doing it.
Unfortunately it makes absolutely no sense to start rolling it out now. It is just throwing away money.
If we continue rolling out FTTH as per the previous government plan we will be finished in under 10 years by the coalitions own stats and therefore only 20 years behind Japan and most of Asia.
I think it’s sad that policy makers have to rely on faulty data.
MyBroadband is pretty pointless for my area. Estimated median ADSL is almost 16mbps. That’s not possible since our exchange is 100% Telstra DSLAM and about 80% of the ports provisioned are stuck at ADSL1.
I discovered this after churning from Internode to Bigpond. They moved me from an ADSL 2 to ADSL 1 port. No one I spoke to at Bigpond made any sense and clearly made stuff up on the run (including the awesome “your exchange doesn’t support ADSL2” When I pointed out that they moved me off ADSL2 on the same exchange, I was then told that I’d been moved to another exchange….. there isn’t another exchange).
It’s also overly pessimistic on claims for mobile broadband. It says 3G when we actually have LTE.
The site just seems to generate random crap with little relation to the real world.
Qualifications versus mandate?
http://www.rmit.edu.au/staff/mark-gregory
Malcolm Turnbull is just applying his masters in ad hominem
However, Turnbull’s also right: The tool does provide a rough, region by region guide to the quality and availability of broadband in Australia. It is a very useful tool for policymakers.
If it’s consistently off by 50%, how is it a useful tool?
But what about the web page? Ohhhh, by tool you meant the web page, not Turnbull.
Wait… so the HFC gets an ‘A’ even if individual premises can’t get it, because we’re dealing with regions…
Whereas the FTTP in a region can get a lower score on the grounds that some individual premises can’t get it?
lol, what sort or arrogance is required whereby someone with no (zero, zilch) engineering experience thinks it OK to question the professional opinion of not only a qualified engineer, but someone who holds a PhD in the field!
It’s time politicians were put in their place. It’s ludicrous to think these people can make any sort of informed decision in today’s modern (read: complex) world.
So I guess that we don’t need an NBN now because Australia has fibre availability somewhere.
(Extrapolating to the max.)
How is it useful for policy makers when they see an A for HFC yet most in the area can’t get access?
‘theoretical’ speeds advantage carriers by making them look good. overestimated ‘speeds’ disadvantage end users by overstating what they can expect.
maximum data transfer rates are great and all, but end users are more interested in connection reliability, connection quality, and the maximum average transfer rate. peak rates are great and all, but they don’t reflect actual use – nor the capacity of installed infrastructure to deliver quality service.
focussing on maximum data transfer rate is like evaluating roadways by the fastest possible speed one can drive along them. actual use conditions generally involve multiple shared traffic scenarios, mandated speed limits (accounting for various user, environmental, and technological factors), and the quality of the road surface and related furniture and equipment (among other criteria). each of these have their equivalents in evaluating network performance.
public policy ought to be informed by actual ‘evidence’, actually compiled for the purpose at hand, not by public relations. (public relations may well be needed to ‘sell’ policy, but it’s a poor performer in generating workable ‘interventions’ to ‘real’ ‘problems’.)
oh gods, what am i thinking?
“If you consider the overall purpose of the broadband availability and quality report promised by the Coalition prior to last year’s Federal Election, it was never to provide comprehensive information to Australians about the availability of broadband at their specific premises.”
Then why call it “MyBroadband” then?
The name itself implies it is about your own experience.
Gregory has hit back at Turnbull over MyBroadband
Turnbull’s birthday cake moment:
http://www.businessspectator.com.au/article/2014/3/6/technology/turnbulls-birthday-cake-moment
Kind Regards
Mark bites back
http://www.businessspectator.com.au/article/2014/3/6/technology/turnbulls-birthday-cake-moment
All’s under control everybody! Telstra & the coalition must have put their plan in place prior to the election. Telstra has been training new CT’s/contractors for the remake of their distribution network, the last leg connecting Malcolms nodes to your premises.
Because it is being implemented by bean counters and administration type people (not engineers unfortunately) you can now rest assured that your new CBN network is going to work so well.
They have even sourced a brand new joint enclosure that’s going to keep all that nasty water out – apparently, they just find that bad nasty wet joint, put the new joint on and she’s sweet!
Well maybe for a little while, until once again they discover that if you put a new cable joint enclosure on wet CABLE that nasty old water is just going to work it’s way back under the insulation, into the connectors and hey presto, high resistance joints again, keeps the Techs in a job though.
And yes they ARE putting on new enclosures onto WET cable! (cost too much to replace the cable, that’s why your telephone/ADSL service has been deteriorating for the last 15 years – “Going to be replaced by fibre one day, lets just keep putting band aids on it”.)
So if you are lucky enough (unlucky enough) to get FTTN you may get an upload/download speed that Malcolm indicates depending still on how close you are to the node (going to have to install a lot of these nodes Malcolm), but just keep in mind – wait a few months, a bit of water in your pit, joint not sealed properly, water working it’s way back into the connectors of that new you beaut joint, back to square one!
FTTP is the only long term future proof solution (depending how well it is constructed – at the moment too many shonks in the network). I don’t want any part of a copper network, copper pairs or HFC which is now up to 19 years old, I don’t want any part of the CBN, like many of you give me back my NBN!
Comments are closed.