Wild woods vs walled gardens

One the way back from the discussion of the Future of the Internet earlier today, I walked through campus and saw yet again two contrasting ways of growing plants:
Wild wood (OU campus)

The wild wood is a fairly heavily-wooded area in the middle of campus, with trees above an underzone of largely annual shrubs, herbs and grasses.
Walled garden (OU campus)

This walled garden is just outside Walton Hall, the Georgian manor house from which the campus takes its name.

I’m not the first one to note that these two contrasting approaches seem to have strong parallels with different ways of organising technical or educational projects (Martin Weller is fond of an ecosystem/succession model of HE institutional use of technologies, for instance). Or indeed with any human endeavour.

But it’s on my mind, and a good excuse to be outside in the Spring sunshine, and to (implicitly) sing the praises of lightly-managed endeavours like woods, meadows and hugely successful applications of generative technologies.

With a walled garden, you plan the layout of the garden, and spend intensive effort making sure that everything is neat and tidy and fits with the plan you’ve made.  It’s largely about control.  With a wild wood, things are a lot less tidy and constrained.

However, it’s a serious mistake to think that the two approaches are entirely at odds.

A walled garden needs regular attention, but with good planning that can be minimised, and if you imagine that you can control precisely what the plants do there … you obviously haven’t been gardening very long. You have to work with what actually grows, not your ideal of what should grow and how.

A wild wood also needs management and attention – it may not look it but it does – even when mature.  Certainly, over-management is generally a bigger danger than prolonged periods of what the medics call ‘expectant management’.  But if you don’t promptly remove invasive species you can easily end up with a stand of nothing but Japanese knotweed.  And you probably need some mechanism to keep the woody plants from over-dominating:  either natural grazing – a problem if you’ve fenced out all grazing animals in an effort to protect your wood – or something else. (This particular wood gets an annual cut of the meadow-like herbaceous lower layer – meadows are another of my favourite habitats.)

Creating a wood is a hard and long job.  You can do some things to speed up the process – though not much if you’re going for semi-natural woodland, but you’re looking at many decades, not a few years, before it looks anything like mature.  It can require as intensive processes of planting and weeding – at least in the early years – as any walled garden.  It certainly won’t look precisely like you imagined it at the start, if you’re even around to see it.

And all that is assuming that the local environment supports the sort of woodland you’re wanting. If it doesn’t, you’re looking at tremendously intensive inputs, if you can achieve it at all.

Walled gardens can be spectacularly beautiful and peaceful places.  But I much prefer woods.


Learning and working environments

Sitting in Tony Hirst’s mashup talk today, I was thinking about the tension between two fundamental approaches to creating a (computery) working – and learning – environment for yourself.

The first approach – the on-the-edge option, in Danny O’Brien’s terms – is to customise an individual machine: cosmetic things like changing the wallpaper, usability tweaks like arranging icons and elements to suit the way your mind works; and installing the software tools you use a lot.

The second approach – the in-the-cloud option – is to use online services you can get to from any machine that has a browser.  This way, you make do with the not-quite-rightness of any individual machine but can get to your stuff ‘in the cloud’, from anywhere.

These are in tension, and what you can do with both options changes both ways.  Web Bookmarks/Favourites is a good example – originally, you could only get at yours from your particular machine.  Then along came Delicious and you could get at them from any machine, with a bit of fuss.  And now browsers like Firefox understand such services and you can get the best of both worlds: bookmarks you can get at from anywhere but are neatly integrated in to your particular browser.  Or the worst of both worlds: bookmarks that live on someone else’s server (they have control, you might not always be able to get at them), and you still have to fiddle to make each machine you use work properly.

Richard Stallman is very suspicious of the cloud and would counsel you to keep your data where you control it – meaning a machine of yours (running a free – not just open source – operating system).  But the in-the-cloud option seems to save so much time and fuss: I don’t have to worry about all that setting up and customisation.  Perhaps that’s just the price of freedom and I’m not paying it.