Getting resilient

Sitting here in the sun, painting with the late Carl Plansky’s finest Williamsburg oil colours to the reassuring background of birdsong, on the first really warm day of the year, it all seems so distant from the worries of the last few winters.

Just over two years ago we were still in the grip of an Arctic cold that climaxed in snow before Easter, and froze us well into June. Then a mere 16 months ago we worried watching rising rivers, soaked day after day by incessant rain. Whatever the weather throws at us, we are told to expect more of the same and worse, as a result of climate change. So we must just man up, and get resilient.

Strangely I see no sign of any efforts towards resilience on the Internet. Day by day we become increasingly dependent on it, not just simple stuff over basic broadband connections, but rich media consuming the sort of bandwidth that might have satisfied a whole university little more than a decade ago. Furthermore the tangled web gets ever more confused, ever more dependent on labile and fragile mechanisms, and our work lodged in ‘cloud’ servers which (like their meteorolgical counterparts) can vanish without warning. Seldom has the computer industry used such an appropriately premonitory term for any of their services.

In times of trouble, among my primary references are the excellent live weather charts provided by the Met Office, particularly its rainfall radar. Now commendably high resolution and whizzy, when server load is high and bandwidth under pressure, maps can fail to load. Information about local travel disruption, including roads known to be impassable due to floods or roadworks, is burped out as and when via Twitter feeds. There is a really lovely interactive map courtesy of Elgin, but it usually lags reality by several days.

Websites claiming to provide information about disruption to particular travel services are often of only historical interest: it was galling to read online that there was a “good service” running whilst I had to wait over 90 minutes for a ferry. I dread the problems which might convince them to admit to a ‘bad service’. Generally speaking tweets contain the best snippets of information, but are often erroneous, and if your employer blocks access to feeds, you are left in the dark.

Central information from councils and Government agencies is no better. If you manage to discover the right page on the right site, the information is often too vague to be of much help, or no longer accurate. The Environment Agency advised us that our stream was at risk of flooding for a period of well over a month, but failed to warn us of the increased risk when it actually did burst its banks.

When your need is greatest, and you can only connect via a mobile device, tracking all these disparate sources of information is a nightmare. Officials may make great play about disaster planning, business continuity and other buzz phrases, but when the chips are down we have to rely on the rumours that spread in tweets.

Instead of hiring disaster planning experts and business continuity managers, the first requirement is a planned national network of local information sites, pooling information feeds from all the different agencies and organisations involved, and disseminating it effectively even over low bandwidth connections.

Whilst these could link to less immediate detail such as how to register a birth over Christmas, they should concentrate on key matters of life, limb, safety, and communications, not sexy website design or the generation of revenue from advertising (another popular way of squandering bandwidth). Being simple, basic, and functional they do not require burgeoning management trees that turn so many clean and cheap concepts into expensive vapourware – NHS systems developers need not apply.

I am a great believer in the Internet remaining relatively unregulated, free to all, and free for all. But there are times when a little direction is needed, to bring order and priority to this free for all.

Updated from the original, which was published in MacUser vol 30 issue 07, 2014.