Educators – what would you most like to see next in OpenSim? February 19, 2010
Posted by justincc in opensim, opensim-applications, opensim-dev, secondlife, virtual-environments, virtual-worlds.trackback
Hi folks. I might not have talked about this much before but one of my chief interests is finding ways in which OpenSim can be better adapted for education and business uses, either as a standalone installation or as part of a larger grid.
I do think that it’s still very early days yet for virtual environments (VEs). Graphically, clients are primitive, especially compared to the latest video games. Outside of research labs and universities input devices are poor – trying to navigate a 3D scene using a 2D mouse is very unintuitive. And backend VE infrastructure systems are either proprietary or only just beginning to gather steam (OpenSim, for instance, still being alpha level code with its fair share of bugs).
Perhaps more importantly, I think that it’s also still extremely early culturally. To me, it’s a bit like the early days of the cinema where people where early films were little more than a fixed camera pointing at a stage. We’re only just groping beyond the naive ideas of ‘3D websites’ and exact in-world replicas of classrooms and offices. I think that it’s going to be quite a while before we really know how to use the medium.
Nonetheless, I’m always fascinated to hear of how people are experimenting with OpenSim and other virtual environments, with an eye to using them to achieve their goals. For instance, it was great yesterday to read Paul Sweeney’s report on a lessons-learned meeting for UK-funded virtual-world education projects, a meeting at which OpenSim was mentioned quite a few times. Among other things, this adds to the discussions of emerging educational VE uses in last month’s UK Virtual World Watch report and to the earlier activites in this space.
In fact, what I’d like to do today is concentrate on educators (in which I also include trainers, museum curators and anybody else providing a learning experience) and ask them a question. What would you most like to see next in OpenSim? Here are a few (!) candidates off the top of my head.
- Higher stability and reliability.
- Mesh-based objects instead of/as well as prim-based ones.
- Easier configuration.
- Better integration with existing Second Life based educational system such as SLoodle.
- A proper guest system where users can log in with guest accounts (e.g. Purple Guest, Blue Guest, White Guest) without needing to create an avatar account first.
- Implementation of missing Second Life features such as the ability to take groups of objects into inventory.
- Bundling within OpenSim of currently 3rd party-sourced components such as Groups and Search.
Don’t restrict yourself to this list – please feel free to nominate anything you like. To simulate the limited resources available, it would be really great if anybody commenting could restrict themselves to just one (primary) choice.
I should also make the disclaimer that I probably won’t get much free time to implement any of these soon – already I’ve had to put experiments with updating sculpties from scripts on the backburner. But I’d be very interested to hear from educational users what their priorities are and this may still influence others in the community both directly and indirectly (that’s one of the great things about open-source).
justincc's opensim blog 
None of the above:
– SLoodle integration is already working.
– Stability is quite ok for mid sized projects.
– Mesh based objects should be adressed later on, they are not needed now.
– Guest system can be done by just predefinig some accounts, not necessary to work on it now.
– Features are complete enough for current projects.
– Bundling is nice but has to be done by experts providing the software as a package for educators anyway.
What we really need is:
– A professional Open Source Web Portal System for
a) Community building and easy website setup
b) Frontend for the enduser to manage the avatars and sims
c) Backend for the administrator to manage the region- and gridservers
d) Focussed on ultra easy user signup and simple administration
– Standardized standalone and small network package including basic feature like voice, groups, profile, search, etc. as VM-image with professional update system
– Finalization of the 3Di-browser based viewer to lower the hurdle to get access to the vr-worlds
– Simplified Viewer that hides the complex userinterface options behind an expert mode that can be disabled
The above is for education and will be enough for bringing OpenSimulator into everyday usage of educational projects.
And for the rest of the virtual world to become better we have to add:
– Professional Open Sourced money system that works secure and stable
– Get the HyperGrid2 Protocol to work so that the virtual universe gets real (“Tear down this wall!”).
Those two topics will kick of OpenSimulator, as their absence is blocking most professional projects right now.
Wish List:
———–
– osSetDynamicSculpie – like dynamic texture (on the server side)
– Flash support for the in-world web client (on the client side – snowglobe?)
Those would change the world (virtual world anyway) 8) ~ !
– Rich
===
If the Opensim project is wishing to ask questions of educators then the Opensim project would need a paradigm shift.
With due respect, you are assuming educators would ask “merely” technical questions and technical wish-lists.
Educators are more likely to want answers on:
– How does Opensim allow interaction with disabled persons.
– Opensim required a broadband connection. This contravenes accessibility conventions.
And finally, educators like Post-Partisanship which Opensim does not do well on. One of the biggies is “commitment to diversity of opinions and perspectives.” With an Opensim attitude of “dont criticize, get out of the kitchen or fix it your self.”
I am not criticizing Opensim, I think Opensim is great, but really just warning of the problem of dealing with Education.
The guest avatar/account thing: I think security demands account credentials, even for guests. The intuitive work around on this amorphous issue is to create several decent looking default avatars, one to be chosen at account creation. If you don’t have to start at ground zero on your ‘look’ you spend more time acclimating to being in-world, less time focused on your inventory contents.
The education arena necessarily is a walled garden. Teaching is not done by the Dali Llama, so any lesson must be bounded by parameters of involvement. That said, 3D representation of things only imagined in some classrooms is irreplaceable. I remember a play along game in history class where we all pretended to be stock traders in 1929. A 3D world would have pushed that well past the ‘I don’t want to pretend today’ attitude that many received it with. So, in a short note, making scripting/building/and interaction with in-world ‘sims’ should simplify the educators task of using OpenSim to .. well, educate.
Accessibility issues will resolve themselves soon, but work to allow low end viewer hardware to co-exist with high-end services is worth doing. Anything on the OpenSim side that makes it possible to run low end viewer hardware helps in that respect, even if it means that for lower end compatibility, you must load special OpenSim modules and are limited in which grids you can connect to/with. So, you might find that it is agreeable to have OpenSim-EduLite version.
All that I learned when starting skydiving, could well have been practiced in a 3D world. Repetitious training saves lives. OpenSim could do this, so there are special education needs that we often don’t think about. Modeling a home for firefighters to demonstrate how safety works is yet another area for training that becomes easily absorbed in a 3D manner.
There are MANY ways to apply the moniker of ‘education’ to the list of ways that 3D environments can improve training. No one will make money out of a video game that teaches home fire safety, but you can do it in OpenSim without EA footing the bill and expecting a return.
Well, a few thoughts, thanks for all your work, it is truly appreciated.
[…] Educators – what would you most like to see next in OpenSim? | justincc.org (tags: education virtualworlds opensim) […]
Stability was a major stumbling block last time we worked with OpenSim (we had students working in sims without supervision, and they crashed it so often we ended up writing a shell script to check if the whole thing had died and restart automatically). That was a couple of versions (0.6.6 I think) ago, though, so might have improved since then anyway.
Other things:
Documentation: Either having behaviour, especially scripts, match SL documentation, or improve documentation on OpenSim-specific behavior
External authentication sources: Let us point OpenSim at an LDAP server and have avatars auto-generate when users log in the first time.
Parcel permissions: We gave students each their own parcel of land, but couldn’t actually block other students wandering in (I think we discovered that listing every other student in in the banned list work, but the standard “Allow public access” option didn’t actually stop people crossing into the parcel.
I remember seeing talk of easy ways of adding content to the default library, open the OpenSim dev mailing list – that is likely to be useful, as it means as institutions build base content (clothes, avatars, textures, etc.) we can easily share them to provide a more useful starting point…
[…] Educators – what would you most like to see next in OpenSim? | justincc.org […]
* A proper guest system. I remember logging in to a grid (I think it was the Intel one, iirc) with my gmail address — that’s great for museums, exhibs et al. A more decent mapping to a readable avatar name system for that, i.e. “zonja.capalini gmail.com” instead of “http://www.google.com” followed by a lot of giberish. And finally, access control on *that*, so that you can close your sim, parcel or the whole campus when you want to impart a class or seminar, and still have people coming without explicit registration.
* Yes groups and search and everything which is considered “core” should be directly implemented by Opensim. Having to implement an extra XAMPP or WAMP is painful and cumbersome.
* Easier configuration could be resolved by having one or several predefined packages, like Diva’s distro, say a “small academy” package, a “university” package, a “museum” one, etc.
* A predefined set of tools in the inventory implementing basic needs, like tools for presentations, showing [streaming] video, etc. So that people don’t have to reinvent the wheel every time or lose time shopping. This could come in the Opensim library for these distros, or better, in all versions of Opensim.
* Stability and reliability are quite good nowadays for standalones. What I see as urgent is a strenghtehing of the script engine so that “normal” scripts from SL do work (i.e., color changing scripts and in general scripts which use loop techniques).
* Finally, some decent items in the library to clothe oneself, and a decent set of predefined avies, so that people don’t have to be always ruth — and, I don’t know whether this is client side or not, but, if at all possible: enlarging the “away” timeout automatically to a very large number. I’ve been too many times myself in The University of Away Ruthes and it’s an awful experience 🙂
groups, i would love groups . . . =)
Bonsoir,
excuse mais je ne parle pas anglais.
juste te remercier pour tout tes infos pour opensim.
Ce post n’est surement pas au bon endroit
From a technical perspective:
– stability stability stability
– easy download and install
– security and groups
– better user manager (not just add, but view and delete)
– many struggle with SL, working with alpha SW is a different ballgame
– accelerate development, obtain corporate funding, move out of alpha
Most of the teachers that use it or want to use it are not from the IT discipline. They represent other content areas (Business, Communications, Education, Marketing, etc.) Then there is the K-12 environment. Budgets are small but ideas are big. Technical expertise varies, but not to the extent required to interpret “non general user” instructions. We provide pre-packaged systems for a marginal cost with complete instructions for ongoing support.
After this, leave it up to the professionals to implement the pedagogy requred to deliver learning outcomes using the tools available. Basically, OpenSim should become just this “software” that is running “over there”. Separate the technical software jargon from the ultimate end users..the teachers. They do not need to know the technical jargon.
Needs to be a separate education hosted service that is pre-populated with educator tools, services, and pre-defined pedagogical models not even remotely found in some of the underperforming other alternatives to SL. It is not about “technical, technical, technical”. It is about pedagogy.
Also…too many half-complete “going nowhere” viewers out there. Most will require changes once new Linden viewer arrives. Hopefully theirs will suffice.
Three things necessare for a successful build (and usually 2 are missing). Without a balanced combination of these items, something gets lost. The proper combination is what has hindered Linden Labs for years, and may continue to do so in the future.
– Pedagogy – without this there is no learning
– Science – includes art, programming, wind, lighting, and more. Without this knowledge, the result is junkyards and the inability to discover new pedagogical models. Hence: PowerPoint in virtual worlds!!! If it was all fopcused here, then we have a lot of what we have today…NON Usability!
– Game Theory – Proper game design theories need to be implemented. It is what makes the video game industry successful. WIthout this, interest wanes in a short period of time. Too much of this, well, then it is just a game, and not an education using game theory to implement.
Hi, Justin, really nice this your post. I remember the very first reason I arrived at opensim was to have an open system which can be run in a school allowing interactive desktop sharing (I read in that period a lot of enthusiastically posts on this topic).
In Italy there had been setup some educational grids involving dozens of educators actively trying to use opensim as a primary elearning platform. What they are facing right now is the following:
* Instability (sims are crashing really too often, especially when using a minimal script). At least need to have an automated watchdog checker which can restart automatically the sims if they crash.
* Lack of interactive tools. VNC sharing appears not working. Voice FreeSwitch is difficult as hell to set up (and thus it is avoided).
* Groups appear not bundled and difficult to set up.
* Lack of specific sloodle integration. Yes, sloodle is working (at least a bit), but I needed to manually modify many scripts before they actually did do something useful.
* Bundled ready to go installation
As somebody asked here there is the real need of a bundled installation web + opensim + sloodle oar that can be given to a school and telling them: hey install this, configure this and up you are up and ready to link to a moodle course and start a lesson. Without that
many people will simply be frustrated.
Possibly something can be a “nice to have” like VNC and to some extent voice. Regarding voice it might be interesting to use skype according to some forum they can reach 25 people at the same time (http://forum.skype.com/index.php?showtopic=412991). Skype is much more easy to make people not computer savvy to install and use.
Regarding screen sharing what about yuuguu?
OpenSimEdu integration might be something not necessarily on opensim developers shoulders, but possibly “mashed up” by other educators, but surely there is the need of a strict cooperation among them and opensim core developer.
And Justinc, thanks for bringing this topic to open. Hope a REAL *beta* opensim version can be out there in some months. THAT would be the REAL kick off for opensim to be adopted more extensively. If you are not able to avoid crashes, at least allow for autorestart and for automatic sending of crashes reports so you are able to have statistics of how reliable is the platform in a real usage.
salahzar
Salahzar,
Regarding VNC…if you are talking about VNC on Ubuntun 9.10, there is an issues, but it has to do with the remote desktop inherent to Ubuntu…. If you go into Appearance Preferences, turn off the visual effects, and VNC works fine. But there is an issue that will be resolved.
Back to the content of these posts, the theme will repeat itself – separate the technical from the user to the largest extent possible. Educators do not care about some of the technical “buzz”. It should just become a server that folks use, and not “monkey” with it all the time. I have been in the IT industry for 37 years. I can get through the stuff in the documentation, but there are times it can make one’s head spin, and that is really not necessary. My servers run pretty nicely since the nightmares of the early days of OpenSim. They never crash. Yeah the featurs above are missing and need to be incorporated. Well they issues / requests above should be a requirement / priority along with the bug fixes and other enhancements.
Lots of press letely about virtual worlds and their value to education…see this one: http://bit.ly/aeiBa9
I think there is a critical timeline as to whether the technology will fly or not. Just as a mere user, many educators struggle with the interface. A push needs to happen, or it will just go on the shelf. Let’s not make the same usability issues that Linden Labs has made, and they are a stable environment. There are reasons why LL has lost about 90% of its customers since inception. Usability and HCI are high on the reason list.
Once again, the best thing that could happen is: Linden Labs Lower the prices for Islands to educators. At the same time they should become an OpenSim integrator, and provide grids where ecucators can connect in a safe and well established way. And it would be for educators / Schools only. They are too scattered now in SL.
Yes there are places available outside of OpenSim, but some are questionable. And yes, folks need to be creative, and learn the technology (as a user), but there also needs to be pre-established builds that showcase the environment and the capabilities of the technology. It will serve to both impress and also to set a goal to which folks can strive. Seeing nothng but junkyards from region to region is fine, but keep them in the background. It can be really discouraging.
I think something is coming as far as Linden Labs and OpenSim integration (well, I know there is). I would wait and see before jumping ship.
Thanks for the thoughts everybody. Very interesting stuff – I’ll comment more extensively when work lets up.
But certainly there’s a lot on the wishlist and I don’t think that a lot of this stuff will happen soon – even if all goes extremely well I still think we’re very near the start of a long journey.
Thank you for this topic Justin!
I’ll give you a short list:
-spatial voice
-LDAP integration
Other than that I think it’s the viewer that’s holding it back. No school is going to run their own sim without someone technically proficient and at least a little obsessed with developing it. But that’s okay, someone has to have the enthusiasm to get others to see the value.
I think it’s stable enough at this point (0.6.9) and releases like Diva make it much easier to maintain.
Justin, you are on the right track. It is a lot of work especially when developers are, for the most part, donating their time.
Erik is correct about the viewers, but on the other hand, all the functionality in the viewers is needed. But to a casual user, someone who is not interested in all of the power, the viewers can be intimidating.
Just make OpenSim stable, and include some of the functionality mentioned.
Again, for me, it works great, never crashes, and the builds we do certainly push the system. There is only one issue that is baffling, and I sent you an email on this. It would be nice if this were resolved. See the email.
Mesh based object import is a must for us now.
Well, I read through everyone’s responses and it’s quite hard to pick out a theme! For instance, some people say stability is good enough while other people think that it needs to be improved (in which case I’d urge you to try out 0.6.8 if you haven’t already – it really is a lot better in that respect).
Also, parcel permissions have been improved now and this will show up in the next OpenSim version. Diva also added a mechanism for adding items to the default library by simply popping an IAR in appropriate directory – will try and dig out (or write…) some documentation when I have time.
Everything else covers a lot of ground (portals, simplified viewers, meshes, better admin tools, better documentation, external authentication, bundled groups, easier configuration, etc.). I think some of these we will see in OpenSim itself and others (such as admin tools) will be left to the external projects that surround it. But none of these are simple – they’re all pretty big ticket items that I expect will take some time to come about, though if adoption of OpenSim accelerates this point could be much sooner than a simple linear extrapolation of current development rates would indicate.
Again, thanks very much for the comments everyone. This is very interesting for me and hopefully for others too.
I have been doing teaching in Second Life since 2007, and I think a package OpenSim (with LAMP or AMP) + Sloodle + (Shop + Community) website (drupal) + Currency (maybe paypal) will be useful for any educational institution or business.
Teachers and lecturers will get into teaching in Virtual Worlds if they can set up the system with one click (just let us see how web design has evolved and now with a few clicks non-technical users can create their websites and the increase in websites has been exponential since then as they dont need to do any technical adaptation). The same happens with Moodle, it is very easy to set up and maintain. Therefore organisations could create corporate virtual worlds that can be connected with hypergrid.
For example, universities can develop short courses or entire modules and micro-charge to their users (as it happens with Moodle and paypal). Also business can adapt this package for specific needs, like Moodle has been adapted to business and not necessarily for e-learning only.
Today I was thinking that probably the currency in SL wont be necessary with the new feature of the SL Viewer 2.0, where it is possible surfing on the internet in-world by using flash 10. One can click in any button, so it might be possible clicking on a paypal button inworld. However, I am trying to think in options of delivering 3D objects or allowing entrance to a room/region/parcel/island after the payment with paypal.
Congratulations for such a great work done with OpenSim.
@Andrew – I think such a package would be a great idea too. Hopefully once OpenSim gains enough critical mass such projects will be viable.
[…] a post by Justin Clark-Casey he asks what teachers want to see in OpenSim. if you have not added your two cents, consider doing so. Justin is one of the […]
Security. It is the most significant barrier to using virtual worlds, both protecting students from wandering out and from inappropriate content coming in. Decision makers need to feel secure that a VW endeavor will not end in a lawsuit.
Security!
The most significant barrier to using virtual worlds is the FEAR of what can come in and what can happen as students wander out. Decisions makers need to know that the environment is safe and that the district will not be sued.
Duty of Care
Educational institutions have a duty of care to their students so:
Profanity filters in text chat
Chat logging
URL blocking for built in browsers / web on a prim
asset upload vetting / approval
anti-griefing measures / griefing report
the abilty to isolate students from guests, other users and other groups (both in textchat and speech).
Technical
VOIP: speech is the most natural form of communication with my students and the fastest.
Web on a prim: is a must (with Flash / Javascript / AJAX support) not asking for much really 😉
Groups: in a free roaming world I need to be able to track, grant permissions and interact with my students as a group.
Currency: Not necessarily a ‘real world’ money one but being able to reward students with currency and or permissions to do things would make for a good ‘carrot’.
A good web UI to manage all of the above
Non technical
Documentation:good info on how to set up and run an opensim is widely scattered over the web and difficult to know if it is relevant to the current release
An asset starter pack would be welcome, as dropping into a small blob of land in the middle of an ocean, butt naked (figuratively) as a woman left me wondering where the hell to start 😉
Not all educators are going to be technical or artistic, so some form of asset exchange seems like a good idea.
I am in the early days of playing around with opensim so please forgive me if any of these features exist and I just haven’t found them yet 🙂 I also know many of the above are in progress, some with external projects but I am looking at this from the view of the feature-set I would need in an educational PACKAGE.
@Don – URL blocking for web-on-a-prim/moap can be achieved via the viewer’s security settings. Some of the other features (web UI, groups, currency, voip) are available via 3rd party modules. Many of your duty of care issues could be achieved via region modules but I don’t know of ayn implementations yet. Some of the non-technical issues are project and ecosystem maturity questions that should improve over time.
@Grid Jumper – thanks for the thoughts. I suspect that security is going to be a massive challenge in the Hypergrid. Of course, with OpenSim you can run everything behind the firewall and avoid many of the security problems.
I am starting my journey towards a Ph.D. in Computing in Education. Starting with my Masters Thesis, I’ve been studying the creation of a virtual science labs for online secondary students. The virtual science labs (I refer to them as “Authentic Virtual Labs”) must at the least create learning outcomes equal to a classroom lab but hopefully beyond that which is achieved in the classroom.
To accomplish this, a virtual, 3D world MUST support the ability of students to control their avatar and the inworld objects with enough capability to construct and control real lab experiments. That means that the “scripting” capability must provide for full programmability and event-driven interaction between avatars and inworld objects. For Physics (my subject) the Physics engines are OK but not really sufficient for Mechanics. But also need fluids and thermodynamics, etc.
Then, the “viewers” must support 3D displays and eye glasses and kinesthetic input devices and 3D motion sensing systems.
You asked …..