Well… well well well.
What we wanted to do was get a bunch of students to come into our computer lab, sit at a computer in pairs, wander around opensim, take their photo from their blog post and post it into a picture frame. That was the bottom line. Simple quest based goal – find the house, find the picture frame with your number on it, put your picture in the frame. If this is easily accomplished, then we can go back and get the blog posts into the ‘notecards’ (which aren’t really notecards yet) so that when people walk around the house, they can see a picture, click on it, and see the blog post that the student wrote.
<--Background - if you've no idea what i'm talking about, read this-->
The Living archives is a project that takes 3 junior high school classes and helps them research 19th century history. Their journey started with some themes found in the Anne of Green Gables novel and chosen by an assessment of artifacts from the period available for digitization at local heritage sites… The students then went to the locations, chose images, documents and objects to be digitized. They then did enough research to contextualize that content in a blog post to be associated with the picture. They are also going to take those images and text and bring them into Opensim, an ALPHA open source MUVE (secondlife but on your server with more control) where they will put their images into context by putting them into picture frames in a period house.
<--End of Background-->
We preinstalled the clients, got them connected to the right server, and logged in 13 users. The students arrived. I gave my 5-10 minute song and dance intro… and then the students grabbed hold and started flying around. 5 minutes of pure success… and then, our first server crash. You could see all 13 computer monitors and see their avatars diving towards the ground… with the first one i just assumed they were playing around, with the second a small part of my brain was hoping on a coincidence, with the third, that little voice silenced, I looked to my buddy (designer and opensim dude) nodded and started a new song and dance. I explained to the students that the work that they’d done was very valuable, showed them that their blogs were now ‘authorities’ on google on the obscure arcane subjects of ‘herb juice’ and ‘ice boats’ while chris desperately worked to get the server up.
We never really got everyone up again. One of the other inconveniences we discovered with the current release of opensim is that it doesn’t particularly like concurrent logons. So you’d really like to give several seconds between people logging in. This, as you might imagine, makes getting 13 students to log back in at the same time a little difficult. We tried, and got a few more in, and then lost them, and then got them back in again. Eventually we ended up cycling them to one computer connected to the overhead projector one at a time and got them to get their pictures in.
Not exactly what we had in mind, but many lessons learned.
1. We (and by extension the development community) has realized that it’s probably time to swing the development back from making the software ‘more awesome’ and swing back to making it ‘more stable’. It’s an inevitable balance, and one that we sped past in our own development… we probably didn’t actually need the ‘smoke script’ for instance. Cool to see the chimney working, but extra work for the system.
2. Include another concurrent project that you can divert more time into in the case of a technical failure. Now, we all know this, but i had about a half hour of material, not the 2 hours that was required for yesterday.
3. Start slow and build on simple successes. Tomorrow, when we try this again, we are going to start out with a simple tutorial and are going to run 4 student groups through at a time. Images first, blog posts second. If that works out, we’ll start adding more people. But just moving around inworld ‘is a success’ and accomplishing that needs to be focused on and regarded by both instructor and students as success.
4. Don’t panic. We easily could have had more students go back in… but when, after a second restart, we didn’t see images loading, we thought we’d lost the server. Turns out, it was just the jp2 converter taking its time. Had we given it more time, it would have worked fine. no need to panic. breaking new ground is never going to be painless
5. Never forget that forging new roads is going to be difficult and that selling that is key to success. The students need to buy into the fact that they are breaking ground.
6. It may be worthwhile, as Ian Truelove says, to start the students in a standalone server. We’re not doing that monday, but might in the future.
wish us luck tomorrow.