Board with Shared Media already?

I used the fact that you can put web-based Shared Media on attachments to add a board to my avatar (others have been streaming their webcam). When he rezzes one of the gene markers on the Giant TB genome, it chats its details, these are picked up by the board and the appropriate genecard with further info from BacMap is displayed. The LSL function used is llSetPrimMediaParams.

The Shared Media zoom function means that you can enlarge and centre the display quite easily provided the avatar stays fairly still. I guess it would not be hard to rez an infoscreen on the marker but this is an interesting alternative, given that space between markers can be an issue. As yet there's no filtering of cross-talk from markers activated by other avatars.

The Webkit plugin seems a little unstable but otherwise is proving interesting. I'm sure others will find more attractive modes of display.

Then everything changed

Linden Lab launched its version 2 viewer today. It will take some time to get used to and for me it's a tricky decision as to whether to use it in this week's class or not. Ultimately it will depend on whether it runs OK on the student hardware and whether the install is as straightforward as the old client, i.e. can be managed by yours truly.

Given that it is supposed to be easier for n00bs to use, there is every incentive to adopt it although I've had a few crashes and freezes (it is beta after all) and, of course, I don't have any documentation prepared. On the other hand this is a one-off 2-hour orientation, the perfect use case for the viewer.

For me the primary selling point thus far is the Shared Media system that allows you to play HTML (including forms, scrolling and JavaScript) on a prim as well as Flash. The big change is the loss of dependence on parcel media. The prim you see above is running STRING (with Flash), the PDB (although the Java molecular viewers don't work, naturally), YouTube and Google Wave. Not only do these all work (despite the odd error message), once configured you can take the prim into inventory, rerez it elsewhere and it still works though you may have to login to Wave again and pages can take a while to load.

Doubtless a list of functional apps will fast emerge. Naturally, single-user apps (as opposed to static pages) don't suddenly become multi-user/immersive (scrolling is also non-immersive apparently) but Etherpad, Solvr, Google Docs, and Google Wave should be OK. As the system is scriptable, I hope to link it up to the giant genome in due course. Many other uses will doubtless emerge.

Now we are three...

I failed to notice on Tuesday that it was my third SL rezday. This isn't overly remarkable except that many avatars apparently quit after two years, presumably when either the novelty wears off or their self-set objectives are (under)achieved. Sometimes there are also new shiny toys to distract them.

I'd like to claim that educators are made of sterner stuff but I'm not sure that's at all true. I've been having "fun" this week playing with Google Wave (and yes, I know everyone else is playing with Buzz). I should add that the "fun" was in aid of a paper for the forthcoming VWBPE conference in March (which isn't, come to think of it, very far off...).

On Tuesday I also had the opportunity to demo StoryMachine which, despite or because of the fact it was almost the first thing I wrote, looks like it will never be finished because (a) the code is truly horrible, (b) I keep adding stuff to it. This week's addition was the ability to rez nodes by touching the outer ring. Once rezzed, the nodes can be renamed on channel 21 and then have the same functionality as nodes rezzed in the normal way from notecards. Not amazingly flexible as far as brainstorming is concerned but marginally better than it was before.

As the University pilot island project ends in June, it's a pleasure to have a few prims to play with again on the Nature sims, the place where my avatar and StoryMachine started out almost three years ago.

Possible inworld vendor for proteins

I've been exploring the use of a Rez-Faux-compatible vendor for some of the proteins based on Troy McLuhan's approach, both as a means of display/distribution and as inworld backup. There's not much evidence of a market for them and their prim-intensive nature makes them very much a luxury/vanity item in SL, perhaps less so in OpenSim. Anyway, the Gizmo VendBob vendor (L$1000 on xstreet or inworld) does a pretty good job, albeit that it only sells at a minimum as dollarbies so I shall have to offer refunds until I get around to some "magic box" approach. The vendor comes in two editions, the smaller (not shown) displaying just a single item/image at a time in contrast to the 6-image display. To rez a protein, you display it and press the Rez button.

It's easy enough to setup although you need a box for the customer to open as well as the usual Rez-Faux "box of prims", the latter being used with a positioner script to ensure that the molecule is in the right place and orientation when rezzed. You can incorporate an image and explanatory notecard for each object/molecule as well as hovertext over the enlarged image. The order of display is determined by a config notecard so, with a little editing, I can at least deliver a list of the contents, though a PDB search and display would be preferable. The config can also include a timeout so you can go some way to conserving prims. Indeed, I may have to fake the display of some (e.g. as an image on a prim) as they would exceed the present sim capacity.

What I haven't investigated is the extent to which this can be used as the front-end to an educational exhibit. Could I include an animated pointer, for example, or have atoms and bonds light up in response to chat emanating from some accessory widget? Given that this isn't coming onstream for a while yet, there's plenty of time to investigate options.

Taking a Peek360°

Via Dedric Mauriac's blog, Peek360° is a new free service that generates a panorama from a sim name and a set of coordinates. As you can see, a bot teleports in (you may have to wait as there can be a queue) and takes a series of photos that are then automagically stitched together for you. You are notified by email when the panorama is ready and the Flash-based result is then made public, together with a URL, embed code and options to share on a number of social networking sites (though not, as yet, the new LL-owned Avatars United). The result is good enough for many purposes though not perfect and it does tend to scroll rather quickly though you can pause it. There is a warning that lag can be an issue (a second attempt may be in order) and I guess there are some issues in that the bot can be sent anywhere with public access.

While not exactly SL-in-a-browser, I think the service has a number of uses, not least in reducing the technical barriers to producing panoramas (previous pano-related blogs). It could give new avatars a more immersive feel for a location, more of a context, before they teleport in (teachers might use it to point out features to explore in particular directions, for example). You could also combine a series of embeds and slurls to setup a trail for students to follow. It might be used to record student work and students themselves could use it to compile a record of places visited. It would also be a nice complement to static images in a RL presentation and, if the images were downloadable, might be used for historical archiving purposes.