Liking Puppeteer version 7

Puppeteer is the classic prim animation tool in SL and version 7.0 sees the introduction of a very useful anchor system. If you observe the video (on how not to spread a bacterial suspension on agar), you will see that the pipettor and the glass spreader move as integral items despite being part of the linkset comprising all the objects on the mat (remember you can watch full-screen to see the detail). This would be tedious and error-prone if done entirely manually but the latest Puppeteer has an anchor system whereby one prim is designated as anchor (the large black prim, pBody, in the case of the pipettor) and the others are linked back to it via any intermediate prim. Move the anchor prim and all others linked to it also move to maintain their relative position. For this to happen, all relevant prims must be uniquely named and the linking system then simply requires that a statement be placed in the Description field, e.g. anchor pPlunger to pBody to anchor the plunger to the body. Very nice.

The video itself was shot using SL Viewer 2 with shadows enabled. On reflection, shooting aginst the ocean backdrop wasn't too clever but I think the shadows add a lot.

Anyone who has a broken shared media dashboard, you are not alone: there is a bug. I'm busy on other stuff at the moment but will get back to it very soon. Apologies.

Shared media dashboard beta version

A couple of people expressed interest in this. You can now get a copy of version 0.35 at the following slurl. Just rightclick and buy for L$0. All the usual caveats regarding beta unsupported software, etc. You can now add a brief description to the cards and landmarks, CD1-9 and LM1-9 respectively comprising a space and a maximum of 7 characters, no spaces. For example, CD1 Demo. A major bugbear is the following script which doesn't change height with the avatar and can get confused as a consequence. Following is easily switched on/off, however, using the touch menu in the follower. I've left it off by default. As well as fixing that, I want to see if I can use a data URI to generate a menu on the follower. Or do I...

Shared media dashboard concept demo

I extended the shared media dashboard a little as illustrated in the (very ad hoc) video above. It now loads collections of 7 web pages from up to 9 notecards and also allows text substitution, i.e. you can add a line in the format sub,YYY,health to the notecard and all subsequent lines with YYY will substitute health in its place. This may be useful where you you want to display a collection of URLs with a common parameter.

As per usual, if you have the right permissions, you can also set screen URLs using drag-and-drop from the web browser favicon. Moreover, the screens will report back in chat on their current URL in a format suitable for copy/pasting into a notecard.

The screens are now contained in a follower for use on open spaces such as the giant TB genome. The follower, in addition to the Display and Report options already described, also has an Open&Shut toggle that shows/hides the screens. Additionally, it stores up to 9 landmarks and the follower acts as a teleporter when one is selected, bringing up the teleport map and then deleting itself when the avatar has moved on (at present it only auto-deletes after landmark-based teleporting).

I'm sure people will come up with more imaginative and elegant uses for shared media but for me this is an interesting starting point that hightlights both the potential and the limitations of this new feature. In particular, I'm interested in seeing how the concept scales as I do see major differences in page-loading at times. There are, of course, prim count and permissions issues as well.

VWBPE 2010

TB in SL: Thinking Big about the Very Small

The Virtual Worlds Best Practice in Education 2010 conference has finished. Getting my presentation together was a last minute job as I had had a busy week and a few things went awry. In particular, I forgot to lock the slide viewer and someone decided to be playful, changing my slides. No excuses, I do know better and I could have stopped and locked the viewer.

Why didn't I? Well, I'd actually become immersed in the presentation and the backchannel. I was supposed to take questions at the end but found it hard not to comment as observations and queries were made by the relatively small audience (it was 2am SLT). I thought it worked pretty well and it didn't stop avatars answering one another's points.

Using voice and addressing the backchannel meant that I did fall behind a little on the accompanying SpeakEasy chat though the presentation itself ran pretty much to time. This was despite my rezzing various objects to illustrate the slides, such as prim and sculpted proteins, a ball-and-stick drug molecule and a small section of the giant genome. Having said that, I did plan to rez the Puppeteer player (as an example of how students can use the molecules) and the shared media dashboard/follower (ditto for the genome). Another time, hopefully...

Given that our pilot sim is closing in June (I hope to continue elsewhere), it was great having the chance to explain what has been achieved and what remains to be done. The VWBPE is such a good advertisement for the virtual world concept, bringing together as it does educators from so many different countries, a significant number of whom would doubtless not be able to attend a largescale international congress in real life. Long may it continue.

Thanks to Shailey for the marvellous pictures of the presentation
to the organisers for their hard work (greatly appreciated),
to the sponsors for their support and, last but not least,
to the audience for a truly memorable experience

Shared media automation

Two very quick observations on shared media on a prim:

1. It works with Project Sikuli, a screenshot-driven programming environment. Opensource Obscure previously noted that you can access SL client menus but you can, of course, also automate zoomed-in web pages -- the code above creates a new Etherpad, waits for the page to change and then types "Hello world" in the main text field. I guess ultimately that's no different from using Sikuli with an ordinary web page except that you could also set visual cues inworld for it to sense, at least if they are on or adjacent to the zoomed-in prim. I'm not sure how Sikuli works alongside other software either; I've only used it on the viewer in fullscreen mode.
   
2. It works with the web version of Evernote. This means that you can capture inworld images and share them on a prim with students via a public notebook. Moreover, any text in the image becomes searchable (notes can also be tagged).
   

As I said, I'm not sure whether these have any major advantages over using a standard web page except, of course, for shared presence. Having said that, I guess you could also use Sikuli to carry out simple SL operations like teleporting.

As both require local installs, their use may be problematic anyway on university networks.

Shared media dashboard

This is my first attempt at compiling a multi-screen analytical dashboard for the giant TB genome. It's based on the new shared media and draws heavily on the TBDB database at the moment, not least because it includes some Flash applets alongside the usual Java ones that don't play with shared media at present.

There were several issues, notably that TBDB is based (sensibly) on the standard H37Vr genome whereas I (stupidly) used the CDC1551. The plywood prim at the front does its best to translate between the two. However, many of the pages in TBDB also use an internal id for the gene in question, the Correlation Catalog, Gene Expression Scatterplot and Operon browser being useful exceptions. I would, of course, like to be able to open a collection of pages automagically but I am some way from doing that.

At the moment I'm using CiteExplore for literature searching. KEGG Pathways was working at one time but then stopped for some reason. CoryneRegNet also gave mystifying problems but it could just be me.

There's an Etherpad console at the front; I was pleased to note that you can copy text via the clipboard.

There's no grand design behind this, mostly I'm just playing. Flipping between the three screens at one level is aesthetically pleasing if you have the screens zoom-enabled. They overlap at the edge so touching the adjacent one makes it centre automatically. If you zoom out, you can move to the other row of screens equally easily.

Shadow on a prim

I'm trying to get a new maze setup and thought I would try shadows (best yet) and the new projected textures courtesy of the new viewer and a nice tutorial by Opensource Obscure). I guess the textures (e.g. the map emanating from the default prim above) could well add a certain something though hardware constraints will probably mean that it has to be optional to start with.

Very pleased to see TiddlyWiki running nicely on a prim too.

Africa: Texas Style

I ran a session with the maze for UTD last week (NB. Africa was edited slightly so it would fit their sandbox). The theme from my perspective was as much rezzers as mazes so I did a demo of a few rezzers before the students textured the mini-modules for the maze and then the final maze itself got done. Well, almost nothing went to plan though this lovely set of screenshots suggests a different story. An hour just flew by...