sneaking a peep

at film playback for the website... work in progress.


Tricky stuff, as there's a lot of possible video clips that the film could comprise, in any sequence; plus long audio tracks that form the soundtrack - but don't start from the beginning.

... and that's after getting all the information about the walks out of the mediascape!

I'm glad I'm not the one making it...... ...thanks Ben!

Tracing the Park

'Tests and traces' was yesterday's theme in the park... look out for some video snippets of the first e-merge walkers at http://www.e-merge-walks.com/ when the festival starts. It was great to see the images playing in the location they're made for, and some useful feedback from the Zeph & Alex, thanks.




Pretty quickly the walks were up on a test-site on the web (with wifi & hot chocolate at the ICA cafe), so thankyou to Ben who is busy on the web side of the film-making, right now!

The Royal Parks have kindly agreed to use of their illustrated map in e-merge, so the first task was to walk around the park marking where I was on the map at various points, so mscape could turn the image into a 'real' map.. clever stuff.. 50 points later, and there's a bit of give, but pretty accurate for the central area of the park, so walkers will be able to view their journey on the map if they want to.

Then checking precise location of 'no-go' areas (like the new dredging site, and the park police station) so that the content regions will be in just the right places.

Interesting observation: mscape logs a trace for every map in the mediascape, so as I had Bristol still in there from Friday, as well as the illustrated map I was testing out, it recorded 3 traces...

creating a 5.5mb file for a 95 minute walk, and a trace overlay that made me gasp for a moment!!

Fortunately it was all salvageable with a few tweaks in excel (thanks for the tips, Zeph!) - so my legs didn't ache in vain!

growing

Here's the best test yet - now with real video and slideshow content from the park. The ipaq performed well, and the screen was very visible in the dusk! The battery seems to have held up fine, and even with a lot of media playing there were no real problems. That's a very happy feeling after the last couple of days of systematically figuring out some niggling playlist errors - which ones were bugs, new features in 2.5, quirks of my method, errors with the device, or inconsistencies within the programme.

The walk starts at the top (near the lane) and snakes around to end about a third of the way down the left side. It lasted 21 minutes, and then the film lasted for 15 minutes. The park is a lot bigger, so the scale of regions will be different as well as the layout, so the relative length of films will be much shorter.

With this walk I was watching my position on this map showing all the content regions and trying to walk through as many as possible. That's harder than you may think, in a bumpy muddy field on a slope, with no obvious physical mark of where the regions are!

It's interesting how different the route of my walk was because of this. I haven't decided yet whether to show people in the park where the regions are, or just show them the map of the park. They don't have to look at the map - just walk and enjoy the sight without needing to watch the screen.

lots of strands...

... these 2 shown below are from 10 second and 30 second traces of the same walk made thismorning and used to test the website bits that will go live next week... thanks Ben!

... there's also some background to the project on the BEV blog , and a page up on the e-merge walks website (where the films/walks will be shown during and after the festival).

... content-wise, it's coming together well, informed by the mscape region layout as well as the music, park layout and the footage captured.

... organisation also involves things like booking forms for equipment, checking logos, making sure the correct software's installed on ipaqs, user-testing and another visit to the Park etc... so quite a lot of strands to weave together into one!

step by step

Once again, trying to fix a couple of unlikely scenarios ended up making things that worked go a little peculiar... So enough fiddling around.. here's another walk made thismorning.

I do like the images traces make!
(The scribble at the top is when the device lost fix for a bit, and tried to work out where it was.)

Making and posting them here is a welcome little break - bit of a visual treat and tiny celebration or waymarker - whether it's successful or not, it's still a step closer to the destination.

Good news from 2.5 - it remembers the logger configuration!

x 30 x 3 = 2.5

... or something more even! 30 content region pairs, each with video, slideshow, state and scripts. Moved into mscape experimental 2.5, and everything's going well except that the playlist seems to behave differently to 2.1 (probably doesn't need the workarounds I'd built in to it before any more!) so there's still a few bugs to iron out, but this is pretty much what will transfer to the Park. So it's over to Ben to add in the flash parts and extract data to recreate the films on the web. Exciting stuff!

Why is there no red trace (showing my walk)? Every time the mscape gets saved as a new version it needs the logger re-configuring, and in my excitement to run round the field again I forgot! Maybe tomorrow...

wild geese

Friday was a bit of a wild goose chase - I got carried away with trying to

pre-empt any kind of user action that would cause the mediascape to work differently from planned. Eventually I got to the point where building in lots of precautions was starting to 'break' it for ordinary use... A few steps forward, a couple back again.. in both the muddy field and on the computer!

The things that happen in the tester or on a micro-walk are less likely to happen in the Park-scale version.

Setting expectations and introducing the context of the experience is really important for walkers, and that will happen through the intro. narrative, instructions on screen, as well as giving any advice needed when they book out their ipaq at the ICA. Those bits are missing from the sketch at the moment, so it's easy to get over-vigilant about what people may do! There's also a fine balance between wanting it to be robust for public use, but also being a piece of innovative development, part of ongoing r&d - without the commercial end product that (should) necessitate a further level of usability. - and there's also the reality of one person's time!

So the latest stage has been adapting the local test to a sensible scale and making sure it works for the kind of scenario I want to create in the Park. That's the first trace below.

I will be extending the regions over the next few days so I can try out how the content runs together, and check performance of the ipaq once it's got a lot more imagery loaded. So the second trace shows the edge of an easily walkable area of the field - without falling into hedges, boggy land, or the neighbouring building site!

Looking SSE across the field-test...

sketches in time

Building the mediascape structure...
Following on the theme of change and archive, it was time to move on from using Something More as the structure to develop e-merge. I suppose one of the things with moving image of any kind is that you have to sketch it out in time. With locative media the dimension of space needs sketching out too (topography or place at least - height/volume doesn't feature much yet).
What happens when someone moves outside the 'active' space? What happens when their device doesn't recognise the place? To what extent do they need to see their journey as part of the mapped landscape, or does that detract from the experience of landscape, or experience of the piece?

My use of mscape explores ways to use pervasive media in harmony with an unencumbered direct experience of place. When you go to a beautiful place, is it really enhanced by looking at a screen or listening to a commentary? Is there a way to use the technology to enhance experience that both adds to the occasion and evokes a memorable artistic experience? And of course there are all the concerns of interactive media design about meaningful user input - to what extent can, or would we want to, control our choices and the freedom we offer to others?

Those are the indulgent philosophical thoughts of the morning, reflecting after the midnight hours that birthed a working sketch (of time and space) unique to e-merge.
It has a beginning, middle and end! It knows where it is and what to do if it doesn't; it can choose music; doesn't drift (too much); knows when to play the film and is even polite enough to ask if you'd like different music or another walk!

Now back to the finer details - timers and finding the most reliable way of logging are the tasks for thismorning. [Note: sketch including timers & log here]

coffee and cake

make great companions to creative thinking! Lots of meetings today in Bristol about e-merge and other mediascape projects . The Watershed does great wheat-free sustenance: orange polenta cake, chocolate truffle cake. Beats flapjacks!

Thanks to the Pervasive Media Studio for loan of an ipaq kit to use in developing e-merge; Constance Fleuriot for mentoring support; Ben blindfish for help with the clever stuff (more to follow about mscape/web and log/playlist technicalities).