New-wave computing

The music industry is finally coming to terms with the web as a way of countering the file swapping that it blames for a massive drop in CD sales. Record giant EMI is to make 140,000 tracks from 3,000 artists available for download, and Apple has the go-ahead put 200,000 songs from various artists and labels online for sale in the US at 99 cents each
    The moves will be closely watched by the movie industry, which also faces a major shift in the way it sells its products. While courts tussle with the issues surrounding DVD copying, the technology is already moving on: the online world is turning into a sort of multimedia jukebox, with all types of content on tap.
    Three developments are accelerating the changes, the effects of which will spread far beyond the entertainment industry. Always-on fast web links, cheap high-capacity storage and ubiquitous wired and wireless networking will also change profoundly how we work, communicate and obtain information or entertainment.
    This special report examines three aspects of this new wave of IT: the marriage of the handheld and Digital Audio Broadcasting (Dab); the larger untethered computer, as exemplified by the smart display; and, on this page, the changing face of broadband.
    Today's entry-level broadband in the UK hardly lives up to its name by delivering just 500Kbits per second - 10 times faster than dial-up, yet barely fast enough for a video stream. I have been trying Telewest's pilot 2Mbits cable service, which is more like the web should be, summoning pages for the most part as if it were flicking through a book.
    This should be the entry level, and it surely will be one day when the scale and economics of access allows. Danes typically use links twice as fast, and Swedes commonly enjoy 10Mbits access, but their compact populations are said to be easier to mesh than the in Britain.
    Of course, a slow server, or a sluggish link en-route, stays slow however fast your own line, so you are never immune from the world-wide wait. And today's content is designed for slow lines, which will not always be the case: pages are likely to pack more kilobytes as delivery speeds rise, so we will be clawing for faster links for a long time.
    Currently, the problem is rather the reverse. There is little content available that needs 2Mbits, and so little incentive for people to pay a premium for the extra speed. This is a reprise of the Catch-22 that has long plagued the web: you won't get the content until you get the users, and you won't get the users until you get the content. 'This is why we are offering this service for the first time to consumers. Someone has to make a start,' said Telewest web consultant, Fergal Butler.
    It has to be said that 2Mbits is also available from some ADSL operators, at a price. A second reason for the Telewest pilot (which was offered free on a first-come basis to 1,500 subscribers to its 1Mbit service) is to establish how much people will be willing pay. Not that these early adopters - keen online gamers, teleworkers, or people sharing a connection - are a typical slice of web users.
    In the long run, broadband operators are likely to move away from a flat-rate charge. The very mention of the possibility sends ripples of fury through web discussion groups, yet it seems a perfectly sensible move that could be good for all involved if properly implemented.
    You would probably get a flat-fee basic service much like you do today (better, one might hope), though perhaps with a cap like the 1GB-a-day limit imposed two months ago by NTL. If you wanted a faster link for a while, perhaps to watch a film, you would pay a little extra; and if you visit an online shopping mall, say, the faster access might be paid for you.
    But you'd pay for quality of service as much as speed. Indeed you do already when you make a dial-up phone call rather than messing with telephony over the Internet, which is not good at the timely delivery required. Operators, for a price, could set up an IP link of the required quality - and it could be used for video-phoning as well.
    This touches a sore point, because videophones require bandwith upstream, which is restricted to 250Kbits/sec even on the 2Mbits/sec pilot. It is likely to remain so. Butler blames peer-to-peer (P2P) users swapping audio and video files: 'If we increased the upstream speed they would simply swallow up the bandwidth.'
    P2P users tend to blame greedy operators, record companies, movie companies - everyone, that is, except greedy P2P users. The problem lies not with their own downloads so much as the fact that their machines act as file servers for others, crowding out local traffic.
    The cap on upstream speeds is one reason for the relatively slow take-up of videophones (or video messaging) and remote surveillance, which are sure to become major web applications. They could be seen as intrusive and oppressive but on the plus side they could transform the lives of housebound people by easing their isolation. Working parents could check up on their children, or see whether it was a thief or the cat that tripped the burglar alarm.
    Fast downstream speeds also give the operator a chance to establish a potentially profitable portal that gives users much better transfer rates than the wider web. Dial-up users won't know the difference because their local link is usually the slowest in a connection.
    With 2Mbit links and fast local servers, we have the start of a system in which you can have any amount of programmes, films, music and other material on tap. The BBC, which has one of the world's best content archives, already has this in embryo on its site where you can listen to recent radio programmes. Clearly, people will want scheduled programming too, but even this is changing with the advent of high-capacity storage.
    Low-cost hard disks can easily store the equivalent of 25 movies, and allow you to
    time-shift TV to the extent that some in the industry believe that within 10 years only 10 per cent of programmes will be seen at the time they are put out.
    Storage at the user end also gives operators the option of providing programmes as a single file on a low-quality IP stream, rather than broadcasting or streaming them in real time.
    There are many possibilities here for a flexible pricing: you might pay more for an instant video-streamed movie, slightly less to have it sent as a file within a few minutes, or less still to have it delivered in the slack night hours.
    Telewest's Butler reckons a content-on-tap system like this will be functioning within five years. Over at NTL Broadcasting, head of product development Simon Mason stresses that for a mass market such systems will have to be 'easy enough for my grandmother to use.'
    It wouldn't be hard to design a graphical interface friendlier than that of the average video recorder. But there is a wider point here: some of the trickiest problems of this new wave of IT could turn out to be ergonomic rather than technical.Now turn to
TV over the radio? It's in the picture
The mobile is developing into a medium in it own right, love child of DAB broadcasting and the web.
The tabla rasa
The smart display is a blank slate on which we could write the future of computing

Like this story? Spread the news by clicking below:

 del.icio.us     Digg this     reddit!