Technology frustrating at first and in the end, but in between very clear what great potential the urban (even rural !) tapestries idea has. Esssentially to create a series of overlapping, self-authored maps of information and experience, as varied in range and contnet as the people who create them Essentially a collective memory of space.

Like the approach very much: the parallel running of technical trials with bodystormiing techniques to develop ideas without the limitations of the technology now or in the future.

Also the incremental approach to trialing: clearly, the future stages of week or month-long trials when individuals caqn build their threads over time will be more productive than the short demonstrations possible today.

Still - very valuable - look forward to following the development.

Posted by: Greg Hilty at December 8, 2003 01:15 PM

I found quite hard to pin the right place in the map. I was trying to get a specific cafe but it was giving me the market nearby. I found it easy to access the pockets I created.

I found it easy to follow a thread. I didn't do it while on the streets though. Some of the threads I would have liked to follow.

I didn't find the map very easy to navigate. Being used to a A-Z like maps where you can read the names of the streets didn't make it easier to find my way around with no street names. I had to hold in my mind the place I started so I had a reference. Not easy

It was very cold, so it wasn't really confortable to stand outside. The network wasn't very fast either and I found myself getting a bit impatient.

I created a thread called Brazilian London. If I were to continue I'd add places with Brazilin names, places where you can find Brazilian people, or places that are run by Brazilians...

I think I'd like to have created a thread that is a log of the places I've been in London and my comments about them. I get a lot of people visiting me over here and always find myself trying to remeber good places I've been to to take them or make a suggestion.

I enjoyed reading some of the threads. When they were a very general comment it didn't really appeal to me. I was more attracted to entries that would give me some kind of context as well as ersonal opinion, or experience.

All in all it was very nice experience. I really enjoyed playing around with it. I can see all sorts of interesting things emerging from this kind of public interaction,

Posted by: Luciana at December 8, 2003 02:00 PM

Overall, fascinating. A really worthwhile exercise, tho hampered by technology. But the principles are really interesting, and the underlying principles of this project are suitably exploratory and open-ended.

(scuse spelling mistakes - my fingers are almost frozen on this cold november day in london!)

It was very difficult to use the software. There are immense interaction design problems to solve within this area - the software is already struggling with the information density from 2.5 days worth of trials. How will we represent the combined responses of thousands of people over thousands of days? This gets close to mapping psychogeography of course, or beginning to actually draw Raymond Williams' 'structure of feeling' - which is an extraordinary project.

Many interaction design problems within the software thus far over and above that too i.e. some basic stuff about denoting buttons, responses to users while uploading, mandatory fields etc. But that can all be solved.

Don't know if Flash was the appropriate technology to use, but Java would have its own problems.

The map itself is on the one hand too complex (i.e. every annex, every detail of every building) and on the other, not complex enough (i.e. no street names). Perhaps a middle ground, of a more abstracted map (information design) with some key street names, building names would've helped (the principle wayfinding nodes in the area i.e. naming british museum, senate house, the various squares etc.)

The icons used to denote threads need a lot of work (i.e. there seems to be no logic underlying the choice of symbol (diamond, square, circle) and colour) - they *could* mean a lot obviously - beginning to denote similarities in terms of type of content, or user.

There's loads here about markup too, relating to this idea of detail of mapping. How people understand place and space only tangentially relates to the post office addressing system. Obviously this system is data rich, but only really meaningful when delivering post. Already out of date in terms of the business names in place. And extremely difficult to pin the place you want i.e. a particular branch of caffe nero on southampton row. In trying to denote the fact that they do a decent macchiato, you have to select from an unordered list of anonymous 'Southampton Row' addresses, and hope for the best.

Similarly, people will want to abstract and aggregate addresses into their own markup of space i.e. the post office doesn't represent 'Bedford Square', but the individual addresses around it. Yet people think of Bedford Square. Or Senate House, Or Victoria House ... or as Southampton Row as a street, rather than all the individual addresses within those places. So some kind of controlled vocabulary to enable people to input/mark up their own representation of space, which can then be aggregated post-hoc? Hmmm.

Much of the trial was conducted inside Victoria House, following the map without leaving the building. Which was ok, but cheating somewhat. When I actually left the building, I was intrigued as to how I was more inclined to click on symbols on the map I was walking past. I'd explored the area in terms of my own mental map, whilst inside the building, heading for the places I knew about. When outside, I was exploring the physical space I was moving through, clicking on things I was walking past. Hadn't foreseen that I would do that.

Using it on the street, one got the usual feeling of 'about to be mugged' even in Bloomsbury. However, each passing day we see more personal tech in the street and this gets to be less of a problem. A few funny looks, tho (but maybe that's just me). Wasn't really a problem.

I overheard an interesting point about how the cold (it was bitterly cold today) was affecting the wireless mesh a bit. I find this interesting. For some reason, the Peter Cook (Archigram one, not Pete and Dud one) quote about 'when it's raining on oxford street, the architecture is less important than the rain' comes to mind. The weather conditions warping the wireless net is an interesting idea. Weather varying the content in place might be interesting i.e. 'OK it's cold - your nearest warm coffee shop is only 50m away' ...

Returning to density, we generally created qualitative, individual responses ... however once there are thousands of threads in place, it will surely be a question of some quantitative aggregation too. i.e. "This comic store, Gosh!, is rated as 'Very good to Excellent' based on 40000 ratings in the last 6 months." And then have the option to 'zoom' in to some individual responses.

Would be nice to reference the Web, and then pin web addresses against location (not just to place businesses, but also provide further context). There's loads about the local history of this area online - it would be odd to create a ghettoised information space around this area which doesn't relate to the web.

In terms of threads, obviously it'd be nice to be able to input content around a particular category, such that you could orientate the map around a particular interest. i.e. view the map by 'architecture', or view the map by 'coffee' ... thereby enabling a way of traversing the area along a particular special interest. I've mentioned the idea of viewing a map or representation of the city in terms of music, history, or architecure on my blog.

I managed to make a few threads, one around Senate House (architecture of), another around the British Museum (politics of), one around Victoria House (architecture of) and one around Caffe Nero (central heating of). So mine were to do with buildings, spaces, history etc. rather than people.

Relating to density again, there's loads of work in terms of the overlap between reputation models and category etc. i.e. don't show me entries by this person, don't show me local community info, don't show me history entries by this person etc.

Again, the interaction design of this needs a lot of work in terms of indicating threads. I didn't 'follow' any threads - as I was unable to discern where they were and what they were.

Anyway, despite the PDAs being tortuously slow (or rather the software) - the experiment was fascinating, and great to see something actually in practice (as basic as this was) around an idea that I've been thinking about for ages. Has certainly influenced my thinking already ... Many thanks ...

Posted by: dan at December 8, 2003 02:11 PM

I just wanted to respond here about Dan's point regarding street addresses; We're aware about the Squares and Gardens appearing only as "holes" in the map. We used the Ordinance Survey's coordinate mapping of the Royal Mail's postcode data as our initial set of refrence locations because without it we would have none at all(well actually we had about fifty, we now have 97,000+), but this can and indeed will, be extended to include user defined points, and other physical and abstract "features" contained in the OS's data.

Posted by: Danny Angus at December 8, 2003 03:07 PM

Like the idea of the urban tapestry.Potential for games and quizzes, treasure hunts etc.

It failed to work on the street.

In principle the authoring needs to be quicker and more user friendly.

Basically the idea is a good one but the technology is not quite up to speed.

Posted by: simon smith at December 8, 2003 03:25 PM

easy to understand how it works. would like to be able to access thematic threads e.g. art deco. would like to be able ask a question in a particular location and come back to find the answer. need the map to point out to me where i am so that i can orient myself. i would have liked to have had a better experience of the technology outside. i never managed to access information once i left the room and hence couldn't comment on what the experience would be like. i can imagine that i might get run over by a car in my eagerness to read the threads. i would be very happy to author if it was a faster user experience. audio would greatly enhance the experience. it would be a lot of fun to add your own photos and voice content. can see where it is going but i feel it has a very long way to go before i would use it.

Posted by: algar at December 8, 2003 03:37 PM

in the hq, everything seemed to be going fine. apart from a frustratingly slow keyboard, and the fact that it crashed a couple times, the overall concept was obvious and exciting.

outside, nothing seemed to work. I managed to make a post but susequently couldn't find it. After that, it didn't really work at all... I couldn't get a signal, and couldn't read anyone else's posts. So following threads wasn't really possible.

My biggest problem was finding out where I was in the first place. Not blessed with a wonderful sense of direction, I couldn't identify where I was in relation to the screen in front of me. In an ideal world, I want the dot to follow me automatically, or at least tell me which way is north. Too much of the navigation appeared guesswork for my liking. That meant that clicking on a dot was guesswork, and so I was less inclined to go there. I want to know what's behind a dot, before i go there.

Lovely idea though, just a snatch ahead of the technology for now....

Posted by: richard at December 8, 2003 03:38 PM

point 1. the premise is fantastic - very much a state of the art premise and the design process thats gone in to this is superb - i really wanted it to be a great success....
point 2. that said, technology is never perfect and here in freezing cold of central london we suffered a great deal resulting in a very limited experience.
point 3. authoring didnt work on the street for me but did in the main base. it took a while to get threads rather than authors but once i had that handle then you could feel soo much potential around the chance to leave a question on someone's doorstep.
point 4. i had no problem with the map but you had real sense of desiring a popup / location bubble to reassure you.
point 5. technology on the street left me with a sense that the pda felt out of place compared with a phone... felt exposed compared to texting
point 6. my threads focused on a personal tale
point 7. i would like to have left far more questions about the inside of buildings, the peoplke who lived / live here and whaT THEY MAKE OF THEIR WORLD.
point 8. few more immediate thoughts on Ut part 2/3 would be to rteally focus on the interface - flash seemed to really slow it down and i was left feeling that sms would have beeen far quicker. that said i'd be interested in knowing how a younger audience adapt to it.
point 9. parting thought was that the technology will never be perfect so prebuilding in for the imperfections will create a more fulfilling world.
keep it going though as it is part of where i'd like our world to grow to

Posted by: adrian woolard at December 8, 2003 03:39 PM

Excellent idea, I really think the idea is worth developing and it presents an excellent use of wireless technologies... however... the current technical pilot lets the system down and it is difficult to assess the system without a robust and complete technical platform. I think perhaps developing the system using a tablet PC technology might have been better - from the point of view of rapid prototyping and a running on a system with a bit more omph. Certainly the Flash client on the iPaQ was too slow to make the 'on the fly', 'author here and now' aspect of the project actually viable, which is a shame. So I think that let it down somewhat, but I still think the concept is sound. A second technical point would be to actually use a location based service, where your location is either determined automatically. I know the project were thinking of using GPS to do this, but perhaps a better idea would be to use either access point tagging or BlueTooth tagging... as a simple way to 'roughly' locate the user then a map based 'final selection' could be implimented.

So great idea... I'm sure once the technical problems are ironed out it could potentially become a popular system.

Posted by: Eric Freeman at December 8, 2003 03:48 PM

Basically a very good idea. However it was quite difficult to use. On the street it was hard to access the pockets and to author material. Although were in sight of aerials it did not seem to connect properly so we re booted several times. I did add some comment about a location but was unabled to check whether it had gone to the server, I could nor retrieve it. The map on the device was hard to use without the printed map, maybe because of lack of street names, or larger view. I didn't really manage to create a thread, partly the slowness of the system, partly lack of time. I thinkl it would take quite a time before really good quality starts to emerge. Keep going.......

Posted by: George Auckland at December 8, 2003 03:52 PM

The first feeling - one of relief to be back in a keyboard with instant feedback!

The second feeling - really glad to finally see an implementation of something we've been talking about for so long.

The third feeling - and a really party poopering one here - that Proboscis have to be responsible about being the parents of this idea. Remember WAP? Or video telephoning? These are technologies that were all done terrible harm by making them available to the public at a stage where they were much too immature. As a consequence, people's first experiences are so bad that a backlash against the whole genre is inevitible. Be very very careful, for the sake of people interested in augmented space projects everywhere! If you show this to people who are not aware of the limitations of software and hardware development you may do some real long term harm. So just be careful about who you invite.

Posted by: Tom Steinberg at December 8, 2003 04:57 PM

The detail of the map is wonderful, but i'm pretty lost without street names. At leat tobe icons as visual markers would really help.

In these freezing cold days, a speedier UI would encourage me to leave longer trails of thought.

Once i was out on the street i had lots of ideas about the trails i wanted to leave - landmarks around, my mood my engagement with the environment stimulated lot's of thoughts. But i wanted to capture these with image and sound as well - a picture of now.

Posted by: anne at December 8, 2003 05:43 PM

I really enjoyed the experience of taking part in the trial. I found the software really difficult to use and I'm not interested in reading incidental comments but I would be really interested in using the pda as a reader to trace the input from the bodystorming sessions, to either follow the thread of someone's life through the streets of london or to uncover a story. So I think the pda would work better as a reader rather than a writer.

Posted by: Dane Watkins at December 8, 2003 07:39 PM

Authorship while mobile?

I have questions about the extent to which people want to author (create content for pockets) while on the move. Audio yes, but text? Sitting down in a cafe or on a bus, but while walking down the street?

It reminds me of the techno-boosterim of the 3G mobile networks - the idea that users would create revenue streams from video messaging - that you'd be walking down a street staring at your mobile device doing a video call. (Not that I think you are techo-boosters.)

On the trial I found the device's text input unengaging and frustrating. However I would love to talk to the system and listen to it. (I know you have tech problems to do this at the mo.) Even better, I would like to author pockets from the web - which reminds me of an artist project commissioned by Steve Dietz at the Walker and trailed in NYC.

I also question the prototyping methdology. While it was exciting to come here on a freezing cold day and see what you've done, are there ways you could have created a non-functional Flash prototype on the device which mocked up the experience, rather than building lots of functionality that you can't yet test? Would this have saved resources and enabled a different kind of feedback, brining in your insights from your analogue research before committing to building the server applications.

User interface issues:
In your next iteration I recommend that you consider
- how to relate the detail of the screen map to lived experience for users who don't know the locality (well) or have poor eyesight
- improving information architecture/labels/typography/emphasis of screen elements
- showing on screen that I've already visited a pocket
- building in audio cues
- having an evaluation from an expert UI designer

Other ideas
- Commission two or three live artists to make performances, site-specific events using the system, for audiences to engage with during specific time durations. One working with sound, one with sound+image, and so on.
- Think about the relationship of your device to a user's existing devices. If I am walking along, with the iPAQ in my pocket or bag, for example, what is it that triggers me to look at it? How can it intersect with my main device(s), eg my mobile and incoming SMS, for example?

But in general - congratulations for getting it together and for making it happen, and making the argument that ideas should lead tech innovation, not technology.

Posted by: Lucy Kimbell at December 8, 2003 07:46 PM