The other part of this report - the focus on universal access to 2 Mbps broadband instead of higher speeds in cities - I think is wrong. It assumes that you can't have both, and it also assumes that 2 Mbps is good enough for everyone, whereas in fact it's ridiculously poor for anyone in a modern society (we should be aiming for 1 Gbps minimum in cities).
Edit: One of the comments on the Reg article[1] points out that there wasn't a single engineer on the committee.
Many people would prefer a guaranteed 2Mbps that they can use all the time, rather than the advertising lies of "upto 30 Mb" - which is usually less than that, and which is then throttled if you use it for more than twenty minutes a day.
While 1 Gbps is a nice aim to have (and if you're putting fibre in you may as well put in extra, you don't have to light it right now) most people would be fine, now, with 5 Mbps.
2.5 Mbps is fine for YouTube (etc) and acceptable for BBC iPlayer tv.
Your comment is interesting because it shows the lack of vision that the committee showed too.
Broadband internet isn't just about watching iPlayer. _If_ we had decent, reliable, two way 1 Gbps internet that would open up so many extra possibilities, including:
- many more people being able to work at home effectively (fast downloads, clear multi-way video conferencing etc)
- ubiquitous cloud computing
- processors in many more devices, all connected all the time (with rich video and audio feeds as well as mere telemetry)
- gaming -- playing, and enabling people to develop much richer interactive games
- home medical diagnostics
- all those things that we haven't invented yet
[Edit: It's a shame the lack of vision isn't just for the Lords. Judging by the replies it extends to HN too ...]
Well, you're asking people to dig up the roads and install a bunch of dark fibre in the hope that someone else will invent a use for it.
That's a tricky position in the UK: We're in a severe recession; no-one wants roads dug up; the last time we had a fibre glut it didn't end well for pretty much anyone involved.
Speed of connection isn't what's preventing most people from working from home - stupid PHBs do that.
Synchronous 1Gbps is unrealistic for the next 5 years. Maybe asynchronous 100 Mbps is possible. Not forgetting that the aim is for everyone - isolated villages in Wales and Scotland - not just the 4 big cities who can afford it and need it.
I'd agree that if people are putting fibre in (or building new housing (as someone else mentions)) or even just digging up the road that there should be a requirement to install extra fibre. the stuff itself isn't that expensive and putting lots in gives you economies of scale.
I'm just more realistic about finding the money to pay people to light it.
Investing in infrastructure during a recession is what you're supposed to do. It creates jobs temporarily which increase consumer demand which drives the economy and increases revenue which reduces government debt AND you get the infrastructure you built. It's a clear win-win-win.
What 4 big cities are you on about? I'm sure there's fibre in a lot more than 4 cities in the UK.
I'm 25 miles outside Glasgow, in a little village called Renton and I have fibre from Virgin Media - 100/5
But in the cities these things are likely to come regardless - government mandates aren't needed. BT is covering about 10 million households with its fibre to the cabinet Infinity option, delivering 38 to 76Mbps, and is making rapid progress on fibre to the premises with its Infinity2 (or 3?) rollout, which brings us to 100Mbps. Virgin is also rolling out 100Mbps over its cable network.
BT is indicating it'll cover 2/3's of the country by 2014, but meantime there's evey reason to assume that speeds in their core network will continue to be increased - they've already doubled the speeds for Infinity once, have indicated 300Mbps for Infinity2 / 3 (fibre to the premises) "soon" and have been testing 1Gbps.
So if the government is going to get involved, where it makes sense to get it involved is in making sure people who don't happen to live near the exchanges that are most profitable to upgrade won't fall behind.
While I agree that it's possible to show a lack of vision, the possibilities you've outlined here baffle me. I'm already able to work from home effectively on my ~8Mbps connection and can very effectively video conference. There are very, very few technical reasons why people don't work from home, and far more cultural and political ones.
If you imagine we are dealing with a limited amount of money (which, let's face it, we are) there is far more benefit derived from providing everyone with a 2Mbps connection than providing people in inner cities with a 1Gbps connection.
I'm on 2 Mbps, work at home, and spend a sizable part of each day waiting for git repos, yum, apt etc downloads. In fact, I'm typing this comment while waiting for linux.git to update.
Why not work on a machine in a datacentre that you have terminal(s) onto? Or remote desktop/Citrix if that's your bag. You can control a machine with a quad-10gigE connection from a terminal with a 9600 8N1 connection... OK so I'm exaggerating to make my point but this scenario is exactly what so-called "cloud computing" solves.
- People worked from home perfectly effectively on ISDN. Dual-bonded if you were fancy. Lack of bandwidth has never been the problem, not for decades, literally.
- What does that even mean? It means services off in a datacentre somewhere and no/minimal state maintained on the client. You can, and people did, do that with dial-up modems.
- For what?
- Bandwidth isn't the problem there either. It's the budget to pay huge teams to create the volume of detailed content required. Games are converging with movies in this respect.
- No-one will touch this with a bargepole, not until it is economic to deliver 5-9s reliability.
- If you build it they will come? I am reminded of Wayne's World II.
Without wishing to make a "nobody will ever need more than 640KB ram" comment. There has to come a point where bandwidth speed starts to matter less. I have 100Mbps down at the moment, and my LAN (802.11n or powerlines), my hard-disk, or the ability of my CPU to decode video are gating factors in most cases and not my Internet connection.
Getting every home to a _good_ (e.g. achieved) 10Mbps is in my mind far more important than aspiring to Gbps connectivity.
100Mbps is reasonably fast right now, but when building infrastructure you want to focus non on your current needs but your needs when you finish building that infrastructure. Vary high speed internet has enabled me to got a lot more use from a small SSD by downloading games I want to play again vs. trying to keep local copy's of everything. Sure, by it's self that's not a big deal, but the internet is more than just cat video's on youtube.
PS: Ever download a show because you where trying to remember a quote?
I believe that you should account for future growth when planning infrastructure - i.e., 100Mbps might be sufficient for even the most demanding users and overkill for most but 5 or 10 years down the line it might be viewed as barely adequate – the same as we're discussing 2Mbps right now.
>focus on universal access to 2 Mbps broadband instead of higher speeds in cities... assumes that you can't have both
I don't think it assumes that at all, just that the free market will provide for the cities (which it is.. >100Mbps is available in most urban areas, 1Gbps trials are being held).
This title is incorrect on several levels. Yet again the Reg mangles details.
1) In the UK "broadcast" can mean over the Internet. It's a term defined in a couple of tv laws.
2) There's a difference between "terrestrial digital broadcast" (what the article is talking about) and other forms of broadcast; eg satellite, cable, etc.
3) No-one is asking for "broadcast" tv to end. The Reg article quotes the committee:
> "We recommend that the government, Ofcom and the industry begin to consider the desirability of the transfer of terrestrial broadcast content from spectrum to the internet and the consequent switching off of broadcast transmission over spectrum."
Weirdly that quote comes just after the Reg manages to give an incorrect summary:
> Rather than take up vital electromagnetic spectrum, TV should be delivered exclusively over the internet, the House of Lords' Communications Committee concluded.
Well, no, they're only talking about switching digital terrestrial tv to the Internet. They're not talking about cable or satellite tv.
This is pretty much just suggesting what industry started doing a few years ago already.
There's already several IP based "cable" services in the Uk, with BT Vision being the main one; all the main TV channels offers streaming, and BSkyB - the main satellite provider - offers their "Sky Anytime" service which downloads over a broadband connection, as well as Android and iOS apps to stream to up to 6 mobile devices.
With big performance upgrades being rolled out by both Virgin (cable) a BT, the number of subscribers that can trivially stream the TV shows is growing rapidly.
I see that as a "cost cutting exercise": it's there, so it saves them having to negotiate deals and stream the content, as users would expect the Freeview channels and might not want two boxes.
But the main appeal of BT Vision is the content it provides over IP - customers that don't care bout that might just as well stick to a Freeview box.
The use of RF frequencies to broadcast TV content is by far the best way to deliver that content for a certain subset of programs that cover live events for huge audiences[1]. Serving that stuff over IP would be a real challenge. I'd prefer it if we could keep a few RF broadcast channels around but move all the Family Guy and Top Gear re-runs onto on-demand IP based systems.
[1] Things like major sporting events, Saturday night talent show finals, large scale news events e.g. 9/11, Christmas specials and key episodes of certain soap operas.
I agree, it is challenging but there are already techniques which solve most of the problem. Namely IP multicast. Maybe this would finally push ISPs to properly support it?
Would there be huge audiences if they were not broadcast? Does mass media primarily serve mass interests, or create them? If the promoters of sporting events didn't have automatic channels into everyone's home, would people seek out those channels if they had to directly bear the cost and effort to do so?
Nice idea, but I can't see this happening for a long time to come, for the fundamental reason that we simply don't get the required bandwidth to the premises.
I live in a reasonable sized city in Scotland, and can't watch iPlayer streaming because we don't get the bandwidth (5km from the exchange...). FTTC will help, but its never going to have the reach of the wireless transmissions of broadcast TV (Highlands and Islands for example). Even worse is that with a massive number of new houses going up in this area, they aren't putting any "next generation" communications infrastructure in - meaning to get FTTP or even FTTC they will need to dig the pavements up in future, which is horribly expensive, and thus happens very slowly (if at all).
Until we have a method of delivering high speed broadband over the current twisted pair, or power lines (or any other existing infrastructure), this simply won't happen.
It's a great idea that's totally unworkable right now. Maybe 3, maybe 4 years time, when BT finally pulls it together and gets more fibre runs in, when there's a bit more of a sane understanding of download limits and so on then we'll see it.
I'd very much like to see this in place, personally I have not watched TV in a long time (I am in the minority of course), but there really is no need to clutter up the RF spectrum with what is obsolete technology.
Measures and facts like these only come in to place long after people stop using them however, so this isn't going to be implemented any time soon.
Is broadcast technology really that obsolete? In cases of natural disaster you can bet home internet connections will go down given the large points of failure (including the miles upon miles of underground cable). Broadcast radio and TV can reach an unlimited number of people (i.e. are not bandwidth constrained in any way) within a certain range of a tower.
That is a good point, it wouldn't really, not if a broadcaster tries to adopt the same model as TV - that is to broadcast messages. Perhaps by using some peer to peer technology? Or by caching copies of the video through out the network? There would definitely be problems that would need to be overcome.
I think this should work in the general case however, there would still be problems to overcome, and I can only see it helping internet connection speeds.
I wonder why the environmentalists never weigh in. Probably too technical for them (as a group) to understand, or maybe there's political implications WRT who to support and who to trash talk. I've been in RF for quite a while and at 10 to 500 KW output for each station, broadcast TV and radio electrical demand adds up pretty quickly, for frankly fewer users every year.
The killer is you usually see stats like 500 KW for a UHF transmitter (which is not entirely unusual, I know there's a 800 KW in Ireland where birds flying nearby must fall out of the sky cooked well done). So 500 KW for a coverage area of 4 million residents where I live is only an eighth of a watt per person per each transmitter 24x7. However ... only 1% of the population might watch that individual station at any given time... and only 10% or so still get the station OTA instead of cable or satellite or uverse or internet or whatever ... so suddenly thats 125 watts of transmitted RF per viewer, or perhaps 250 watts input power to the TX per viewer. Yikes.
Its interesting that the power budget for watching TV OTA is now roughly, insanely, 50:50 where roughly one watt of transmitter power is required for each watt of TV power. This ratio is going to get even worse over time. "Soon" there will come a time when you need to transmit a KW per viewer and that's not really economically sustainable, aside from being a really bad environmental idea.
Digital TV switchover has killed a lot of the mega-power transmitters, though.
In the North West, for instance, Winter Hill used to run around 1,000kW for each of the analogue TV channels. Post-DSO, this is down to 100kW per multiplex, as the DTT signal is far more robust.
I'd very much like to see this happen. A lot of posters seem to be missing the point.. the idea is use the TV spectrum to allow mobile broadband to become a viable main-use broadband, which would set a decent minimum standard across the whole country without the need for re-cabling etc. For those of us with already fast internet, it would allow decent speeds (and theoretically prices) when travelling.. awesome (and IMO more useful than 1Gbps home internet). (note to ROW: we don't have LTE etc yet).
Today, giff gaff offer unlimited mobile broadband for £10/mo (+texts and a few calls), although TOS states use on mobile only.
I can definitely see this happening but not for a while. We are only just completing the switchover from analog to digital this year (and if I remember correctly that process started around 4 years ago).
I would love to see the end of broadcast TV. I know too many people that waste their lives away sitting in front of a TV every night and letting it stream content. Yes, someone will come along and duplicate that functionality on the web, but let's just ignore that for now. Content should move to and stay on-demand only.
While we are messing around could we grab some of that TV bandwidth and run a looping broadcast of the most heavily trafficked media data for the last 24 hours to be cached locally ? Ceefax on steroids...
This seems fairly likely since OTA can do 1080p and its only 19.39 megabits/sec complete with all the overhead. After all the overhead is stripped away the raw mpeg2 transport stream is around 18.3 megabits/sec.
As a point of comparison, no one encodes DVDs and bluerays at full rate because then you run out of disks, and can't fit in all the garbage that purchasers supposedly desire, like mandatory unskipable previews and piracy warnings and so forth. But a good rule of thumb is your typical DVD is a round 6ish megabits/sec and a blueray encode stream should be around 30ish megabits/sec.
http://www.parliament.uk/business/committees/committees-a-z/...
Also, it apppears to be talking about terrestrial broadcast TV - not the satellite broadcasting used by Sky and others.
Edit: I love the comment from the report: "The Government's strategy lacks just that – strategy."