Ecppm 2004 wednesday afternoon notes part 1¶

The afternoon notes, in two parts. See the overview page for the links to the rest of my notes. Sarah Bowden (Arup and comit project) - applying site processes for the introduction of mobile IT

They researched the handling of information on construction sites and produced an as-is map (what is paper-based, what is computer-based, what is printed out with some notes scribbled on it, etc.)

The introduction of ICT is a circular proces. You do something, you get feedback from practice, you do something, you get feedback, etc.

The research is grounded in practice, as they started using it on a building site a few months ago. This allows above feedback loop to continually improve the results.

Irina Kondratova (Institute for information echnology, NRC canada) - Mobile field data entry for concrete quality control information

Entering data while on the building site... Your hands are busy with other things and using a pen on a palmtop computer is inconvenient. Best is still to just a mobile phone. So: voice access technology, for instance voiceXML Using a standard like this allows you to choose from a few separate solutions.

Nice term: V-commerce! Services that use voice technologies.

They did a test with a simple system. Problems: background noise, rather complex navigation structure, constrained speech. One possible solution is multimodal interaction, the integration of visual and voice interfaces through the delivery of combined graphics and speech on handheld devices. You can combine XHTML and VoiceXML, for instance. A prototype for multimoday field data entry is under construction (based on a pocketPC).

Conclusion: using voiceXML you can allow hands-free information retrieval, so you can overcome the limitation of the user interface of a field handheld computer.

(Note: Akasig is looking into open source voice stuff, might be interesting to check out.)

Karsten Menzel (Technical university Dresden) - Issues of context-sensitivity in mobile computing: restrictions and challenges in the construction sector

His talk was in two parts: CoWoMo: concept of the mobile worker and Main-E.

(Hm. I didn't immediately get what this talk was about, so not many notes here).

There is a lot of information. Individual actors in the process belong to a certain company, but are also placed somewhere in a project organisation. How to handle this? So they defined context aspects. Actor, device, location, time, environment, etc.

With this you can partly answer the question of who, why, where, how.

Mobile computing requires in-depth understanding of construction processes.

Danijel Rebolj (university of Malibor, Slovenia) - a context-based communication system for construction

Problem: information flow on`and related to the construction site is unreliable.

Hypothesis: mobile computing can significantly improve efficiency of the construction industry. The "missing link" hypothesis.

They tested a lot of stuff on construction sites (which looked like great fun to me!). Setting up wireless networks, wearable computers, PDAs.

The missing link hypothesis was generally confirmed, but:

• using a process model is essential to automate communication.
• communication networks should be dynamic and very user-oriented.

So: the dynamic communication environment. It should be context sensitive, context adaptable (=> filter information) and it should support indicating the presence of users.

You can extract some information from company and project databases, like users and their relations (boss, supervisor, architect, etc.). This part of the context sensitivity is doable at least. Personal networks!

Functionality they wanted:

• instant messaging
• data exchange/accessibility (file based and model based)
• audio/video communication
• session/meeting management

Such an integrated dynamic communication environment can represent one of the key applications for the effective use of it in construction since it efficiently connects the participants to eachother and to the information.

Conclusion: mobile computing is already effectively applicable in construction, but a quantum leap can only be assured by re-conceptualising information systems:

• from application oriented towards user, location and context oriented
• from hierarchical towards network oriented

(I like those conclusions. Now, conclusions aren't something you normally like or dislike - they are true or not true... But still. More emphasis on what the user wants is good. More emphasis on when and where the user wants to do something is good. Network oriented: fits in well with the internet! My comment: make your data also network-oriented. Not one big central database (unless absolutely needed). But interlink several information sources. That means those sources need to be accessible... To me, "making information accessible using the internet and according to the philosophy of the internet" is a central theme).