Life, Photography, and Security

Random thoughts that have crossed my mind

Thoughts on...

computing/internet

February
Mon Tue Wed Thu Fri Sat Sun
   
7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29        

Subscribe
Subscribe to the RSS feed.

2006-04-03

camillo.särs.net → www.ged.fi

Now that .fi domains are also available to private individuals, I registered “ged.fi”. This means that my site will primarily use “www.ged.fi” in the future. Or at least until I change my mind.

[/computing/internet] permanent link

2005-12-01

“Broadband Strategy”

I have followed the Finnish discussion on national broadband strategy pretty closely. In short, it’s a mess. The discussion seems to focus a lot on how consumers will buy DSL or cable broadband Internet connections, and how to provide services for those consumers. That’s of course nice, but I don’t think it has a lot to do with strategy. A strategy is a long term plan of action designed to achieve a particular goal, as differentiated from tactics or immediate actions with resources at hand.

But, you may say, the Finnish national broadband strategy has 50 action points. Isn’t that a “long term plan of action”. Well yes, it is. The problem is not with the actions, but with the goal. What is the vision of the Finnish strategy? There is a hint in the document, though. The Government’s broadband strategy is ‘technology neutral’, in that it does not favour one particular technology over another, but instead promotes competition among them and their complementary use.

The strategy fails to look into the future. It is fairly easy to make some estimates on the growth of capacity demand. The strategy would then describe what the biggest challenges to meeting this demand are, and how to tackle those challenges. Let’s try it.

The most bandwidth intensive consumer application currently is video. Current “broadband” connections are already struggling with the demands of video-on-demand. How will this develop during the next few years?

HD-DVD, regardless of technology, will become reality in 2006. This pushes sales of HD television sets. Current DVB broadcasts are optimized at a quality level slightly above VHS, at 2-4 Mb/s. This level of quality is not acceptable on HD sets, which require a bandwidth of up to 8 Mb/s for decent MPEG2 quality. DVB-T networks are resource constrained, so they squeeze 5-6 channels into a 21 Mb/s multiplex. They cannot meet the quality demand, so quality content for HD sets has to come through other channels.

MPEG2 compressed HD can take up to 18 Mb/s. For the sake of this argument, let’s say an average family has four members. The exact number doesn’t really matter, I am aiming for a rough estimate. It is reasonable to assume that at some time during an evening, at least three family members are watching TV. It is also fairly safe to assume that at some point in time, two recordings may be running simultaneously. This gives us a peak estimate of five concurrent video streams. For MPEG2 HDTV, that means 90 Mb/s of video.

The conclusion is evident. In a few years, a typical family can be expected to have a peak broadband demand of close to 100 Mb/s. This number is two orders of magnitude bigger than the 256 kb/s currently considered “broadband”.

The Finnish broadband strategy is technology neutral. That is why it cannot succeed. The real challenge in broadband today is how any country can help the actors on the market embrace new technologies which enable true high-speed connections for the “last mile”.

Fiber optics have a capacity of 10 Gb/s, which easily can be extended using WDM technology into the range of >1 Tb/s. In anything but the short term, the only reasonable technology available is fiber optics. Installing fiber optics is very costly, so there is a very real need for a national strategy for deployment. The national broadband strategy should be to support rapid deployment of a national-wide fiber optics network.

[/computing/internet] permanent link

2005-03-13

Dan's Mail Format Site

I ran into a fairly extensive site about email formats and formatting, by Dan Tobias. It will tell you everything you never wanted to know about email, but also how to avoid sending emails that some recipients may not be able to read. Higly recommended for those who have too much time and an urge to find out a bit more about email.

[/computing/internet] permanent link

2005-02-25

Webmasters should master the web

I am fed up with non-standard web sites, but even more fed up with webmasters that claim they just have to tweak the simplest html to render properly in different browser versions. Wrong. Since the late 90's, all browsers have rendered standard html in virtually the same way. It's just incompentent webmasters who do not know or do not care about this.

I found a good overview of how different browsers react to different doctypes. The summary for dummies (and webmasters) is: Use a standard doctype, and you are fine.

[/computing/internet] permanent link