Conversation

Notices

  1. I'm extremely frustrated. I have to dumb down my VR training program a lot more than I already have. I'm told that people shouldn't be allowed to drop grabbed items. I mean WTF! They take out all the fun of flinging items around. Also all the dynamic things I've built in should be gone. "Only activate the items that are immediately needed at the moment of the procedure!".

    Tuesday, 14-May-19 13:57:20 UTC from web
    1. @adiwan Grr.... As an end user, I despise when some managerial wonk decides what I should and should not be able to do with the tool. Just give it to me, and let me use it!!

      Tuesday, 14-May-19 15:54:38 UTC from web
      1. @scribus Yeah! They don't know what VR is all about. All they care about is: The workers who have to train with this should not kiwi up in any way possible. It's design by committee.

        Tuesday, 14-May-19 16:17:13 UTC from web
    2. @adiwan ... Your VR training program ? Are you making a video game for uni or something ?

      Tuesday, 14-May-19 21:44:47 UTC from web
      1. @drinkingpony Worse: Industry. This training is about maintenance of a part of a robot arm. In order not to interrupt production VR trainings are created to train off site. My desire to quit and create a game increased rapidly.

        Wednesday, 15-May-19 03:32:29 UTC from web
        1. @adiwan I am afraid that I am having a hard time to put 1 and 2 together here.

          Wednesday, 15-May-19 09:12:10 UTC from web
          1. @drinkingpony Simple: I program a program such that people can gain knowledge of a procedure with this program. They interact (although very crudely) with the program and will be guided at each step of the procedure with a narrated text and helpful highlighting of objects. A VR training has the advantage that it can be trained everywhere (where VR headsets are available) and it doesn't disturb the production line. In the case of the robot arm entering the working area of the robot arm shuts it off completely and the production halts.
            Why is the industry worse: They don't know what they want and if they say what they want it's usually worse. HEY LET'S GAMIFY IT! >Just a simple highscore list sorted by completion time in its "hardest" difficulty (no guidance with highlighted objects or narration). But what kills me is that they want to use the hard coded steps from the guided difficulties instead of offering more interaction freedom (all objects can be picked up and dropped everywhere)

            Wednesday, 15-May-19 13:31:40 UTC from web
            1. @adiwan Oh, ehh. That sounds a lot like people who never even held a controller and make them design their own controller. That or play Sekiro, not sure which is the more hilarious thing to imagine.

              I am afraid that without going into any details I can not imagine how hard it is to deliver on such a weird thing.

              Though I would have probably given up a long time ago and instead just tried to make a miniature replica and let people 'do the thing' on that thing.

              Or... maybe... try to use this thing somehow ? http://rainbowdash.net/url/874441

              Thursday, 16-May-19 00:19:42 UTC from web
    3. @adiwan (I'm late to reply, whatever) What is the purpose of your training program? Is the purpose to give users an overview of what they will be working with so that when they encounter the real thing they are comfortable with it from experience of the analog? Or, is the purpose to train users with knowledge of things that could go wrong and how to deal with it?

      Sunday, 02-Jun-19 05:42:18 UTC from web
      1. @oracle I programmed several trainings in the past. One was about changing a pH probe of a waste water system, then one about assembling a car's front to the chassis, then one about assembling car cockpit elements, and now it's about servicing the welding gun on a robot arm. In general the clients want to train their people to do a series of task by the book, with no deviation, and they are done in VR because
        1) the machine/process doesn't exist yet and it's quicker to train it beforehand such that the time to applying the real process is as small as possible,
        2) training in the real environment costs too much because it halts/disturbs assembly line and such,
        3) train the process everywhere such that there is no need to go to the real location, which saves on traveling costs,
        4) It's flashy and good for marketing
        5) Can be repeated anytime
        6) A very critical error in real life occurs very rarely and it cannot be replicated in real life but can be in VR

        Sunday, 02-Jun-19 06:26:42 UTC from web
        1. @adiwan "I'm extremely frustrated. I have to dumb down my VR training program a lot more than I already have. I'm told that people shouldn't be allowed to drop grabbed items. I mean WTF! They take out all the fun of flinging items around." I understand the frustration, really I do. That being said, depending on the depth of the training, it may not be in the best interest to add distractions or noise to the training environment. The real beauty of VR training is the denoising ability of removing unnecessary items and highlighting necessary ones. You wouldn't want someone to think that something is important when it actually isn't.

          Sunday, 02-Jun-19 06:47:50 UTC from web
          1. @oracle In industrial maintenance, dropping items and tools can cause serious setbacks. My opinion is that a user SHOULD be able to drop items, and that the dropping of the items should create a natural penalty.

            Sunday, 02-Jun-19 06:49:34 UTC from web
        2. @adiwan I am interested in your work here because I have a lot of experience in writing technical training documentation. Mostly with an emphasis in SQF (Food safety and quality) procedures but also in the documentation of my own and my teams poorly poorly written code.

          Sunday, 02-Jun-19 06:51:38 UTC from web
          1. @oracle The main problem is that I'm not responsible for the storyboard of the training. I'm the one who has to convert the gibberish into code. What I get is a very linear script that tells the user what to do and when the action has been done the next instruction appears. I use a giant state machine (in my current one there are about 80 states) for that purpose. Every time an action or an instruction has ended the state machine advances. The storyboard has 3 "difficulty levels" in mind in which the first level the user get all the instructions and the interaction items are highlighted. The second difficulty level is the same but without the instructions, and the 3rd is with no highlighting. In every state only the currently needed items are interactable and nothing more. My current training has also a little quiz in the second level and a thing where the user has to order the steps on a whiteboard. It's on the level of a kindergartener.

            Sunday, 02-Jun-19 07:14:06 UTC from web