cancel
Showing results for 
Search instead for 
Did you mean: 

Michael Running Wolf - Launch Pad "The Madison Buffalo Jump VR Project"

rngwlf
Explorer
Hello! Here’s my first weekly update. 

First a bit about me: I am a member of and grew up on the Northern Cheyenne Tribal Reservation in Montana. I grew up in a tiny village named Birney. My rural community had erratic electricity, water, and got the telephone when I was 15. This naturally led to me getting a Bachelors and Masters in Computer Science.

Along with scholarships and fellowships I financed my education by inadvertently training myself to be a full stack web developer in Linux and Microsoft Windows: this was a result of Montana’s small IT economy where you’re expected to know and do everything. Originally I set out to be a OpenGL developer in my undergrad years but webwork was more plentiful and my interests drifted to database analysis and natural language processing.

Though C#, and other Java/C languages are comfortable territory for me, shifting to Unity and visual IDEs is new to a VIM developer like me. Luckily distant memories of Quaternion Spatial Rotation math must be like riding a bike. 

Virtual Reality, along with the Oculus DK2, reignited my interest in 3D programming and how it can help my personal and educational goals.

I pursued my masters degree because I want to help build software for Native American language revitalization. There are many technical problems involved, such as corpus construction with sparse datasets, but the biggest problem is time. Indigenous language speakers in America, and the world as I’ve learned in my international travels, are dying of old age. Despite decades of dedicated work by linguists and community advocacy, there’s been very little progress to save languages. Dozens of American Indian languages die every year. 

However the situation is not entirely dire: a few tribal nations have managed to reignite their own languages through intensive social and financial investment into advanced language education. Native Hawaiians, for example, can educate a pupil from kindergarten through a PhD entirely in their own language. My wife and I were invited speakers at the International Conference on Language Documentation & Conservation in 2015, where we were inspired by the Native Hawaiians’ success. Though our enthusiasm was shaken by how far behind Natives on the mainland were.

Unfortunately most tribes are just barely scrambling to begin their own language revitalization programs. I believe VR can be a shortcut to kickstart an effective language education program. 

My project, with an somewhat uninspired name, “The Madison Buffalo Jump Virtual Reality Project,” is an experiment to create a multilingual indigenous VR experience. I’ll provide more details as I progress, but here are the bullet points of my project goals:

* Use a real landmark sacred to Montanan tribes: the Madison Buffalo Jump State Park
* Demonstrate pre-Colonization life through a communal buffalo hunt
* Link elder knowledge through advanced technology relevant to Native youth.
* Create audio content to use multiple tribal languages in the VR experience.
* Make it difficult NOT to learn the fundaments of a language.
* Have fun being a VR technovangelist to tribes in Montana’s backyard.

My dad checking out my Gear VR:

civj5xlbjczj.jpg

16 REPLIES 16

akababa
Protege
Love your postings!!!

rngwlf
Explorer

Week 7: Unity Assets


So my project is a multi-prong effort: get Montana Indigenous voices and the technical side. Both feel slow. I kinda wish it was winter so I could spent more time at the computer, but then I'd be wishing the roads were safer to travel so i could record more elders...

I need another 10 hours a day so i can sleep.

 I need to build a sweet cheek Buffalo Jump manually using the default mesh editors in Unity. Another Alternative is to import GIS data as a Depth Elevation Map. But I once, in a fevered delusion, tried to do that with Blender years ago. It was crazy making simply getting the correct data into a GIS system (ArcGIS and it's $12K/year price tag and Grass GIS) to then export a DEM image.  The United States Geographic Service has terabytes of free data in a dozen incompatible formats. 

At any rate, I've come to the conclusion that i'm not going to spend too much time on either path: I'm going to buy a Unity asset to will either get real world data, or allow me to create a reasonable facsimile. There's a few I'm looking at: Relief Terrain Packn (RTP) and Real World Terrain (RWT). both are a bit pricey for a miserly Indian like me. 

RTP will allow me to manually create a jump, while RWT tantilizingly offers the possibility of just clicking and bam an instant buffalo jump! Terrain and world building is it's own little world, there are so many packages that do similar things but these two appear the most popular for what they do.

 I'm cynical, so i'll probably buy both out of necessity. Otherwise my learning of Unity continues, I recently bumped into application size issues, my Android Google Cardboard demo app exceeded 100MB and isn't allowed in the google play store, (while Apple App store lets you have a multi-gigabyte binary, yay?). Luckily there's a checkbox in the player settings for android deployment.

Real World Terrain Asset: https://www.assetstore.unity3d.com/en/#!/content/8752

Relief Terrain Pack: https://www.assetstore.unity3d.com/en/#!/content/5664

My goal is to get a prototype world built and start putting my animated buffalo upon it!

dakini_3media
Explorer
Michael I really love your work! .... This is a really wonderful stream!   Hello also to Caroline!  

aprilarrg
Protege
Loving what you are doing with this project! 🙂 

rngwlf
Explorer

Week 8: Visual Evidence Conference and Madison Buffalo Jump

ti2lbh1009c8.jpg

Didn't have time last week to do a report, was in Bozeman doing a panel for the International Visible Evidence documentary film conference. I met some apparently big wigs in the documentary world (Dr. Brain Winston anyone? Seriously he seemed scary to all the film folks, he was short and pudgy in my opinion).

tkw1hz14ddko.jpg

 As usual we slapped out a presentation we've given before:
1a2ratjor6e9.jpg
The most interesting was a speaker from Guadalajara doing Geolocated VR/AR for the Metropolitan Museum of Art in New York.

81z7k8tqsn1f.jpg
He was so jazzed by what we were up to he wanted to follow me around town.

However I had a prior date with the Summer Art Walk and a comfy chair.xeb7kdkarv2b.jpg

The big take away from the conference was that it's a totally new field and the film world calls what I'm proposing to Oculus Interactive Documentaries. 

The weekend continued with a visit with Montana Part system to discuss a small grant we got to help build a new trail system. The Board of the Friends of Madison Buffalo Jump and the Montana Park system met with Gary, a 20 year veteran of trail design.e4euj1xnyknk.jpg
Madison Buffalo Jump

b9wagc8foueg.jpg
Gary nerding out with the State Park regional managers

ok8mshngbtr5.jpg
Gary pointing out erosion features to avoid
 

txs5fsq1hcb7.jpg

I'll shortly do a technical update I apologize for being semi-erratic, weekends aren't spent on the computer it's when I travel for my project.

rngwlf
Explorer

Week 9: Terrains... 

and discovering the dreaded "You're hardware does not meet the specifications of this application" Android Error.

This was one of the hardest things I've ever done:

k8p0unzz3wc4.png

However I built the app and then *black* and a little while later an error, the comforting unity splash screen didn't even pop up. Back tracking I figured it was a new UI asset i plopped in (remember coders GIT is your friend), i was able to get a debugging error, after scrubbing the offending asset: 
You're hardware does not meet the specifications of this application
My terrain appears to overload the Gear VR!  So my work around was to export the data and spit out a mesh that I decimated until the OVRCameraRig had a reasonable frame-rate. Which isn't as impressive, sadly. This is a rough approximation of weeks of work toward a singular goal then it goes all awry in a fit of panic as I realize "this might not work" *sad face*  But my workaround appears to be OK, though not nearly as impressive (I think I need a Rift to properly showcase the Buffalo Jump)

But given the timeline I have little choice but use my work around, the march continues. I imagine one of those $1000 terrain optimizers for VR will do it justice, but blah.

Now my next goal is to get video and audio integrated of my drone fly by.

My time going forward will be dedicated to the application and the demo build.  I'll try to post my lessons learned next week!

rkoman
Protege
Wonderful work!