Step 1. Design it.
Step 2. ?????? (Area 51 it until it's built)
Step 3. Profit.

oops better make that four times.

As an AV designer I can tell you all for free that for audio to be truley intelligible (with concern to human ergonomics), it should be a minimum of 24dB above any ambient noise at the listeners ear. By consequence, those who answered WHAT? will have to yell approximately two times louder in order to be heard. Trivia - I live for it.

Believe it or not, the function is the most interesting part of design something new.

I've designed both AV systems and content management systems but to this point have found no pliable way to fit the "dream system" into my work.. None of the above recommendations are going to be a solution on their own, you need to design a 'system' made up of complimentary components to create something that is greater than the sum of it's parts. Be it turn-key or bespoke. the 'use flash/silverlight' question and others like it are completely irrelevant until you know how it's going to work overall in a practical sense. Form follows function.

Some quick conceptual ideas..

- A system that handles multiple media formats and can transcode, most likely via a seperate encoding server. Sorenson Squeeze as a concept is a good start, using preconfigured profiles and drop points etc.
- A system that is open to input by a variety of people, both internal and external parties (e.g. staff, external academics, students, and so on). This sort of functionality would provide an open 'wiki' digital content atmosphere to building application specific knowledge resources. This also, if done successfully would provide unlimited interest from third parties.
- A system that is available via a variety of platforms that are commonly accessible and woud act as a natural pathways form existing media reserves. E.g. public web page for uploading content to a specific museum attractions, library ISBN recommendations and so on.
- A system that allows web based media editing, primarily for formatting 'quick-dump' content to make it either suitable for delivery or contextually palatable. See Kaltura for inspiration.
- A system that delivers content to a physical identifiable location in the museum. It should be as simple as 'tricerotops' content arriving in a 'tricerotops' folder and being picked up from there by a mini pc or slim client.
- A system that allows for scheduling and composing of digital presentations. NEC panel director shows a good example of practical back end functionality.
- A system that allows dynamic touch based content call up by museum participants at the exhibit itself. A simple split screen web page with sub-topic content menus and a running presentation would work.
- Or even better, wireless touch tablets with RF/bluetooth proximity link to each exhibit, dynamically loading content as the person moves around the museum. Choose a rugged model, make people sign for it and throw an RF tag with beepers at the exit, done.
- Humanize the entire process.

I can not stress enough the importance of designing the function and feature set before you choose a technology path and allow yourself to be swayed by personal or community bias (it's easy to stick with what we know and swap to function follows form). It's about the user first, the content second, everything else comes third.

Either way, enjoy. :)