The Great Fibre Debate of New Zealand commences
There’s much resistance to change when it comes to building networks. What we have is a known evil, but one that could be managed and improved upon through good network engineering. That’s all very well in theory, but in reality, commercial imperatives and other factors dictate that good network engineering comes about rather too rarely. It’s too expensive and requires dedication at all levels, basically. This isn’t by itself sufficient reason to spend billions of dollars on a new broadband network, but for those who argue that there’s no need to upgrade what we have, it’s worth realising that there’s a big gap between theoretical perfection and what’s attainable in reality. For instance, VDSL2 looks like it will take copper into fibre-optic performance territory, on paper at least. And it is fast, but only over short distances and it is far more power-hungry than ADSL and ADSL2+. Telecom’s roadside cabinets shorten the copper local loop to customers making VDSL2 possible, but it doesn’t change the power requirements of the technology. What’s more, if you ask Orcon and Vodafone, cabinets either put severe limits on broadband competition in the New Zealand market, or even kill it altogether. Orcon’s starting to publicly sweat over the proposed pricing it has seen in the Commerce Commission’s draft determination for sub-local loop unbundling and the changes to the backhaul from being dark fibre — meaning the access seeker gets to put its fibre-optic equipment at either end of the cable, and control the capacity at the cost of higher capital expenditure. If the price of S-LLU isn’t right, Orcon says it won’t fight and won’t go into the roadside cabinets. Vodafone has been making similar noises about what it is expected to pay to Telecom’s network company Chorus for cabinet access. This is entirely appropriate: access seekers are rightly worried about the lack of infrastructure competition that could leave them locked out from retail business. Meanwhile, Telecom’s pointing to the $1.8 billion or investment it has committed to with the fibre-to-the-node programme, stating it will improve broadband performance for New Zealanders. And, provided it provisions cabinet capacity adequately, this is correct. Telecom’s even putting in passive optic-fibre interfaces in some cabinets, meaning FTTP is closer than you think as the FTTN rollout races on. You can’t fault either party — Telecom and the access seekers — for trying to get the best deal out of the regulator, but it does leave customers in the lurch and for a long time as well, thanks to determination processes being agonisingly slow and complex. The last few years of watching the regulatory process show this clearly. Things aren’t going to get any easier from now on though, with the government’s proposed fibre to the premises network plans being so short on detail. When pressed, everybody agrees that FTTP is necessary and the only technology that’ll scale well in the future, as far as we know currently. That doesn’t mean there’s any agreement on the implementation though, or how quickly it should be done. I’m starting to see why the Australian government ditched the RFPs for its NBN, because that process was going nowhere. The lesson for National is to avoid getting bogged down in minutiae and squabbling and remember that there is no perfect solution to anything, broadband networks or otherwise. — Ralph Chivers, TCF: Fibre benefits are complex and multifaceted — Rosalie Nelson, IDC: Building the Great Wall of Fibre – can we deliver? — Louis van Wyk, TUANZ: Regional players form united fibre front — Techsploder blog: Broadband needs greening too
Below are some quick speed tests done on Telecom’s new XT WCDMA network, that’s launching on the 13th of May, using speedtest.net.Ignore the ping times, which Speedtest records as being much higher than they actually are, for some reason. As for the speed tests, sponsorship and PR manager Rebecca Earl told me earlier that customers can expect average speeds on the XT network of 3Mbit/s down, and 1Mbit/s up, so the results fall within that. Outside business hours and with the network not so busy, the speed revs up considerably however.The theoretical download speed is 14.4Mbit/s, with the upload running at 5.7Mbit/s – however, the tests were done with Telecom’s launch modem, the Sierra Wireless USB 885:This nifty little device supports 7.2Mbit/s downloads, and 2Mbit/s for the uplink, upgradeable to 5.76Mbit/s. Interestingly enough, the 885 runs in the 800MHz band as well as the 850MHz one for WCDMA, on top of 1900/2100MHz. No 900MHz WCDMA support though, so if it works on Vodafone’s network, it’ll be only be in towns with 2100MHz signal. Apparently, it’s the same device that Telstra in Australia sells for its NextG network, and which retails for A$299. Telecom will undoubtedly get it delivered in its own livery. How much will it all cost then? No idea, or what kind of data, voice and text plans Telecom will have on offer when XT launches.Don’t read too much into this first test. More to come, I hope.Telecom's XT network passes first tests
Like a bunch of others, I’ll be firing up File Transfer Manager today, to load down a copy of the Windows 7 release candidate. And that’s all I’ll say for now. — Review: Windows 7 RC1 adds speed, UI improvements -- and promises more to come — Hands On: Windows 7 Release Candidate (Build 7100)
XKCD Swine flu
Robert X Cringely One swine flu over the cuckoo's nest
What's worse, a global pandemic or a Twitter-induced panic about one? Cringely says problems with social media are just symptoms of a larger disease. First, I'd like to allay the fears of all those in Cringeville. You cannot catch swine flu by using Twitter — even if you're one of Porky's followers. But you wouldn't know it by looking at the swirl of misinformation and panic that has flooded everybody's favourite microblog. Net.effect's Evgeny Morozov says the swine flu has turned the Twitterati into the jitterati: There are quite a few reasons to be concerned about Twitter's role in facilitating an unnecessary global panic about swine flu... having millions of people wrap up all their fears into 140 characters and blurt them out in the public might have some dangerous consequences, networked panic being one of them. What could happen? People shun bacon and ham, sending the pork bellies market into freefall. People cancel their travel plans, especially to Latin America, and walk around wearing surgical masks. People with head colds decide they're really dying from a porcine-borne bug and flood emergency rooms. That in turn could cause shortages of the Tamiflu vaccine for those who actually need it. And did you know that swine flu isn't really a flu at all but an attack of Advanced Biological Warfare aimed at reconfiguring our DNA? Hey I read it on the interwebs, so it must be true. In fact, the Centers for Disease Control and Prevention has its own Twitter feed, which as I write this boasts 14,139 followers, or only 1,439,680 less than Ashton Kutcher. Its latest tweet: 20 confirmed cases of swine flu in US. 1 hospitalised. All have fully recovered. http://bit.ly/uycgL #swineflu See, not so much to worry about. The problem is that, unlike most Twitter users, the CDC only tweets when it actually has something to say. That little info-snippet went out 14 hours ago. Meanwhile, a glance at Twitter search shows tweets with the tag #swineflu coming in at the rate of around 20 per minute. That's a rumor/panic/speculation-to-information ratio of more than 16,000 to 1. Google Flu Trends, which tracks the incidence of people searching for terms related to "flu," puts the current swine flu threat as almost minimal. Google says its search data tracks very closely to actual flu treatments recorded by the CDC. But what happens to Google Flu when panic causes a surge in searches, bumping the search trend into the red? Flu Trends then becomes part of the endless cycle of hysteria. There are plenty of credible sources for information about swine flu on the web; Mashable's Ben Parr offers a guide to some here; Time magazine offers up five things you need to know; and of course, the CDC Web site. The bigger problem, as Morozov points out, is how unrestrained social media could be manipulated toward far worse ends: I think it's only a matter of time before the next generation of cyber-terrorists — those who are smart about social media, are familiar with modern information flows, and are knowledgeable about human networks — take advantage of the escalating fears over the next epidemic and pollute the networked public sphere with scares that would essentially paralyze the global economy. Often, such tactics would bring much more destruction than the much-feared cyberwar and attacks on physical — rather than human — networks. The real problem isn't Twitter (or Facebook, et al.), the problem is a) stupid people, and b) people who prey on stupid people. Included in that last group are the 24/7 cable news networks, who will do anything to whip viewers into a frenzy to boost their ratings, and the blogs who'll do the same to pump up their page views. So I make this vow: This blog will sell no swine flu before its time. Also: Don't believe everything you read on Twitter. Even if Ashton is your BFF.