12 January 2024 (Andrew Onboarding)
Frode Hegland: Okay. Hello. Hello. I’m just finishing a message here. Boom. Done. Right. So, yeah, a couple of things before the meeting. And notably, I’m talking to you before Dini. That’s random. There’s nothing meaningful about that other than, you know, Dini and I aren’t trying to organize anything here other than how to get you and Andrew to be happy with your different levels and specific things of work. And then there is a complication that I want you to have at the back of your head. Alan mentioned this. So I was stupid enough to buy it. And even more. There is a thing in here quite quickly. Using index cards to think within a group seems to be very central to this. So of course we’re going to try to find something online for our group to do the user stories. But it really strikes me that the user group we’re supporting for this project as it stands, and this is up for discussion on Monday, probably is a, an academic dealing with their own library. So that really narrows the user stories. It’s not going to be learning saving the world. It’s about dealing with your own stuff. Right.
Adam Wern: Yep. But even that is way too broad to have us a development or design target. So so the. Yeah, you know, all the difference between a kind of an undergraduate student trying to read a paper assigned by the professor to a PhD or a Upward’s trying to do a very, very literature review for the for something.
Frode Hegland: Oh, no question. I didn’t mean it in that as a counter to that. Absolutely agree.
Adam Wern: No no no no. But so but already by yeah we have a much tighter scope than start from the start. But we need to do to be much more clear just to see what we’re dealing with, because I think the actual uses will be quite different from thinking of a PhD student to wanting to do some sort of advanced thing.
Frode Hegland: I agree with you. I’m kind of talking outside of this, and that’s why we’re talking really. So the reason is I’ll just do a quick screen share. Actually this is not to properly discuss today at all. I just want to give you a heads up. Yep. So yeah, thinking with index card. This is something we’ve all discussed and thought about for years. Right. So I just thought, okay, if we’re going to make our own because that’s how I am, I just want to make my own one thing we could do, and I’m thinking 2D and XR almost the same thing. Right. Conceptually. So we have a tile so you can put things on because usually they are tiled. And then beneath this you have a pile. So it’s like, hang on. Even that adds the complexity of how do you define how big the tiles are. Right. So then there was the question of maybe just have a horizontal divider where everything above snaps, everything below doesn’t. So that was kind of neat. But the whole thing is, hang on, you know, I do actually have this. And it works to many extent. It does a lot, but there are some severe limitations in this. And so I find that an interesting thought piece, but what I’m really thinking about is. Well. Text at its best, is an encapsulation of a human expression. That can be frozen and moved around, right? Yep. Yeah. So at linear document is important because it makes a point in a sequence. Foreign keep it. But I do think that if we in our thinking with our library and all of this, if we manage to make a way where we can develop these index cards for thoughts, I think we will do an amazing thing.
Frode Hegland: There’s a lot of companies doing little nodes and stuff. I think that with the right user story, we could actually do better because it tends to be over complicated with all kinds of stuff. And furthermore, two fun aspects of this. One of them is imagine that the. Document type is PDF, but now it sounds stupid. But imagine that each page is a card. So that gives us the opportunity to store this stack as a document and then choose to open it up. Furthermore, if we use the way, you know, we can copy a citation using visual meta. So you paste as a citation. If we have a basic metadata for this, we should be able to work on something, copy it and paste it in someone else’s. And as a last thing to mention, now I could imagine if we have such a system and it can one thing can be embedded in another. We could do something. As you work on your map, you give it to me through text, email, who cares, right? I drag it into mine and it appears as that in other document. It knows that I wasn’t the author. So this is Adams. So I double click on that and it opens your map. So it’s now embedded I can do things and then later on you can say hey I’ve updated my map. You just send me a new one, I drag it on top of the old one and it updates and archives the old one. Instead of having live, which you may or may not want, depending on the kind of work you’re doing right. So I think we shouldn’t discount. Discount such a thing. What do you think?
Adam Wern: Yeah. I’m. I think I’m with you for the most part. Well, there are always the implementation details and whether PDF is the right choice generally. And but I agree with most of the principles around it. Maybe and then the technicalities on how actually you want, but I agree with what you described from a user perspective. I think that would be awesome. And in a way, we don’t have that kind of. We have. In general, we have a hard problem of sending collections to people, collection something, things we can send one file. It could be a zip file, but just as a mark have problems with sending the whole data set. And I sometimes have a problem of sending kind of three. A few different experiments. I mean, the descriptions and the images. A collection of some sort. I have a where the collection is more valuable in its whole than the individual parts. Just getting an image is not enough. And now I resort to putting it on slack. Or perhaps a slideshow, but it’s not really. Well, a slideshow can do some some parts of this that you can send a slideshow and you could theoretically move slides around and you can move the objects in a slideshow around. But it’s it doesn’t feel as freeform as what you described.
Frode Hegland: Yeah. The That. It’s crazy to me that such a thing doesn’t really robustly exist. So while we’re talking, this is a normal document in the library, right? So this is so I select text now. And then I do a copy quote. And as you know, if I now go to author and excuse me, it’ll paste it with whatever was there. Not a lot. Yep.
Adam Wern: Nope.
Frode Hegland: But crucially, if I now go into plain text and paste it, it also tries to give useful metadata there. So that’s really the approach I’m looking at. The idea is that I just want to make sure I can see again. There you go. That we develop because in a way the difference between a PDF and this kind of index card thing should be minimal. It should live in the same environment, right?
Adam Wern: Yep, but a PDF is one type. An index card is quite similar conceptually.
Frode Hegland: Yep. Right. So that’s I’m thinking that. What I will talk to in eight minutes when we start the meeting. Is that you and Andrew? Have to get a PDF and metadata into the headset ASAP. There’s no you know, that just has to happen. But that should not in any way cut down on our wider island type discussion of what are we actually trying to build? And if we’re trying to save the world rather than just do this, part of it has to be how do we deal with thoughts, right? Yep.
Adam Wern: And shared thought.
Frode Hegland: Exactly. And it’s so easy. Anybody can build great software. That is a walled garden. It’s not that hard. Nope. But to build it in an open way and open not just legally and technically, but open in the sense that any other developer easily can add stuff. I think has to be the thing.
Adam Wern: Yeah, yeah I agree, I agree. And It’s in a way, we almost had that with the web. But then it became everything became went from being docs that you could actually copy. At one point you could actually copy a folder of HTML with images, and it was kind of a doc that you could keep. And you didn’t care so much where it lived on which domain, because that wasn’t important. The links were just internal, basically. Basically within that was a document that worked. But then everything became JavaScript and highly enriched, and then it became apps of everything. And it’s not a bad thing because we have fantastic web apps nowadays. I just today I suddenly got a mail. Postal mail that was ten there was posted by the Swedish Transport Authority ten days ago and it arrived yesterday. But I opened it today and they said that my car was what do you call it, grounded when you don’t, when you haven’t sent it in for inspection? So it was grounded on the 2nd of January and I didn’t know. Right. Yeah. Because. Yeah. Well, but a few second or a few minutes later, I managed to book online. A time for an inspection. And today, an hour ago, it was inspected. So it’s fantastic. All those apps we have doing things. But but we lost the documents along the way.
Frode Hegland: We did. And using the term document, which is nice to hear because what I’m thinking about is entirely document centric. Right. So the whole point is that we don’t have to worry about archiving or network or online or nothing because, you know, on my computer, whatever software I use would be my master space. And then I can choose to import your things with maybe we can have a script in the future that updates automatically. But that’s not the point. The point is that in an online session, we will have one of these documents in XR or even on the web, do our thinking together, and then we save that and we have it just.
Adam Wern: Yeah, yeah, I am totally with you there. And that is. What we want and what we should strive for. So I think for Alan, I don’t think we should develop this thing first in order to make our designs, user stories, because that would be kind of putting the horse before our car. What is a cart before the horse or or something or invent? We have to kind of do that in the, in whatever tool that is chosen and just do it to get some user stories and or jobs to be done or whatever we call them. But the. Yeah. And then so we have some design targets to design to think about because it’s easy to invent things that have nothing to do with academics or perhaps nothing to do with XR, because I’m, I, I’m imagining that. A few people will try out the Or. Many people with a in A after February will try out Apple Vision Pro in some sense. At the company or at at an event, but there will still be a lot of people saying, so what? Why what does this have to do with. Why can’t I do this? You present it on a regular computer, so we need to have something that absolutely can’t be done. Not just lots of documents, because even that could be done on a computer as we. So we have just internally we know know this. But I need we need something that is kind of impossible to do at least one thing.
Speaker3: So.
Frode Hegland: Yeah. No, you’re absolutely right. I’m going to do a quick share screen again before the guys come in. So you’ve probably seen this, but let’s just open the library, go to.
Speaker3: A.
Frode Hegland: Document has nothing to do with me. Close the library. And I do this. Ask. I and then I run names on it, and it prompt is to extract names from the documents. And yeah, it takes a ridiculous amount of time, but it’s okay. These things can be scripted, you see, at the bottom. The windows has saved metadata, right? Yep. This is taking too long. Maybe it’s.
Adam Wern: A. Yeah, but I I’ve seen I’ve seen this thing. Yep.
Frode Hegland: Okay, so here I can go through, see if there’s anything to delete that’s not relevant. No. So I click save metadata. And it actually works. That’s the that’s the thing that’s kind of amazing. Now at the very end of this document. Yeah, it was a big document. Wow. In addition to normal visual meta it has this. Right. So the list.
Adam Wern: Of keywords or people or.
Frode Hegland: It’s a. Hi guys. Adam and I just finishing a little fun thing. Just two seconds. Yeah. So also, Andrew, maybe worth you just having a look at this, and. Sorry, we’re just having fun. Inside this visual meta, it has the names. It has the result of that I query. And this is readable by the system, like anything else is. So that means it can be used for searches. The reason I’m mentioning it to you now, just for fun, is the notion that it can be used for clustering the documents later.
Speaker3: Yep.
Adam Wern: Anyway, we have to be. We? Yeah, we have to be aware that the last year, that kind of or even much more advanced vector databases that analyzes, you know, the whole AI explosion thing or LM explosion thing that many systems exist for analyzing PDFs. And they are quite sophisticated in that they get they do clustering in much more advanced ways where where a concept with many names have similar a similar kind of yeah, they exist exist in a concept space, so to say so similar concepts that are similar are clustered together. So you don’t even have to search with the right term. But one term brings everything that are close by. And so we have to be because that could be a kind of if we present something like this, the state of the art right now. And it’s kind of it’s much further on into that direction. So we don’t present something too simplistic.
Frode Hegland: No. Absolutely. Adam. And also Danny and Andrew, before we properly start, just to finish that off, it was just an indication that.
Speaker3: You know, we can.
Frode Hegland: Do stuff with the documents that can be used in a bigger thing that was just stick figures compared to a thing. Anyway, guys. And Andrew, I hope you yeah, you had a wood fire, you said.
Andrew Thompson: Yeah, we’ve got a wood stove. So it was came in handy.
Speaker3: That’s so cool.
Andrew Thompson: Powers still been flickering ever since it came back. So I’m here right now.
Adam Wern: You’re on on a generator and a satellite dish and self-sufficient. Now, tomorrow.
Dene Grigar: Is the bad day. Tomorrow we’re going to get the huge snowstorm.
Speaker3: Yeah, it’s bad again.
Dene Grigar: So bad. So beware. Andrew. Hey, Andrew, if you need if you folks need a place to stay, let John and I know.
Andrew Thompson: Thank you. Yeah, I think.
Speaker3: Later. And.
Andrew Thompson: We should be all right. It will probably get cut off from the world again. But we’ll. We’ll have heat and we’ll survive. But if I’m not here on Monday, that’s what happens, you know?
Dene Grigar: Well, Martin Luther King Day is Monday, so I’m. Oh
Andrew Thompson: That’s right.
Adam Wern: Then we send you wood. More wood then send wood.
Frode Hegland: Please send wood.
Dene Grigar: I’m going to the store after our my meetings today because I go until about noon and because I need provisions.
Frode Hegland: It’s. It’s bizarre. My wife’s best friend. She lives in Texas, and they are completely off grid Tesla batteries and solar only while the place is also properly before we start. So I was interviewed by New Scientist magazine this morning on the topic of audio only interfaces. It started with the whole thing of little little things that are coming out like the rabbit and all of that stuff. So I’m not going to bore you with a lot of the details, but it struck me in the conversation that with those kinds of interfaces, we’re pretty much at the same place as VR, AR, because this is the last time that we’re going to be live. And this is why provocatively said, no silence in the future. Of course, that was just to get, you know, reactions. We’re going to be at the point where someone like our kids, they will always be able to just speak something, and there will be an AI to hear what they say and to deal with it. Unless they choose to turn that off. That’s really bizarre. That’s going to change a lot of what and who we are. So we had a great discussion. Maybe you mentioned something on Monday, but finally I did tell him about the work we’re doing, and I said, make this audio stuff a cover story. It’s huge. And then do you want an XR? Right. So glad we’re all here. And Danny I think you’re probably the best person to convene the proper meeting.
Dene Grigar: Which thing?
Speaker3: The proper.
Dene Grigar: The proper meeting.
Speaker3: Yes. Okay.
Frode Hegland: To convene us today. You are the person.
Dene Grigar: I guess I am. Okay, I’d like to cover some business. First, I want to talk about the question you asked about equipment. Send that on my agenda first. Then I want to talk about process and Andrew’s hours. I want to get these things kind of knocked out today. I see this talk today as more operational rather than theoretical. Right. Just nuts and bolts. So we shared the grant information with you, which included a a justification of the budget. So you should have seen that. Andrew. Adam, by now, but we’re going to be buying the we’ve got two headsets going to Frodo and one to me. Andrew. And reading through the information about the headsets, we need to have it geared for you. So when we buy this, you’re going to have to be the one to go in the Apple Store and have your eyes looked at through the apple. Let me finish. Just let me finish the the news this morning was that you open up the new Apple app and they it scans your face.
Speaker3: Exactly.
Dene Grigar: That’s what I mean by Apple Store. So it scans your face. It looks at your eyes, it measures your head. And we can customize this Apple headset for you. You’re looking at.
Andrew Thompson: That’s fascinating. I would have assumed it would be adjustable, but. Okay.
Dene Grigar: No, that’s why two would have been good for us. Because I could have had one. You could have had one. But you’re the prime person that needs it. So we need to have you the be the one scanned on the Apple Store.
Andrew Thompson: We should both scan and see how close it is because, well.
Dene Grigar: My eyesight is worse than yours. We’re going to go for you first know that when we buy the headsets, Frodo, you and I, when we we. The university owns the headsets. Andrew will take the one from my lab home. Frodo will have two for you, and one can go to Adam if you want him to have it through your lab. Now, know that the university owns these headsets. My my comment back to the university when I was discussing this budget item with them is, by the time we’re finished with this project, those headsets will be outmoded. So, you know, what are they getting back? And that is outmoded technology. That is, I’m going to collect, right? Yes, Frodo.
Frode Hegland: So a few questions on a very meta top level. This is your money and my money. The university is handling it. So that’s why I don’t quite understand why they own either any IP potentially. Of course we’re going open source or any equipment. It’s not their money.
Dene Grigar: The university. Any university in the United States. That’s a research institution that takes these grants in the relationship between the funder, Sloan or Mellon, or any of these NIH, any of these organizations. And the university is the is a. Is the prime is the R is the prime relationship. The university represents us. Right. So the reason why Sloan wanted a US universities, because they are a US organization and they need to be able to be in a relationship with the US organization. So they give us $250,000. They know that it’s going to an organization that can handle and manage their money. Right. So the university, any university in the country would have the same stipulations because that’s how Sloan that’s how Mellon functions.
Speaker3: Yeah, I.
Frode Hegland: Kind of we.
Dene Grigar: Can buy it back. I mean there I do buy back stuff myself. So I have a lot of material that the university was going to outsource, and I bought it back. Right. So that’s that’s always there. But the university outright owns it because they’re holding on to the cash for us. And then at the end, if we want to buy it back, they may just give it to us. But the bottom line is we have to go to the university for that. We will be shipping two to you. Froda. And you’re welcome to to share that with anybody you want through your lab. At the end, we’ll have a discussion of two years, what we’re going to do with the headsets.
Speaker3: Right.
Frode Hegland: Yeah. So on the specifics on the hardware, first of all as I texted you privately, Dini Alan, I think we should get him a quest three, but not as payment for work. He’s a complete volunteer, but as a somehow a loan. So he can do the work with us as a piece of useful equipment. Because, you know, he’s. I don’t know. Also, he I think he’s unemployed at the moment. He was working at Twilio, so he’s a bit weary about spending money. So if we can, I think it should come out of testing budget because it truly would be testing.
Dene Grigar: Can’t the way those I’ll find out if there’s any wiggle room, but my experience with working with grants for these 20 years is that when you have goods and services, that money has got a lot of wiggle room. But if something is under a certain category, and testing right now is considered labor because we have it set up like $100 a person, we have set up our testing as physical human beings.
Speaker3: Okay. So. All right, I mean, I.
Dene Grigar: Want to can I mention something else? One of the things I think will be helpful is when you come here and we can do our onboarding. We were talking about a long time ago. You onboarding the first week. That couldn’t happen. So when you come in and we on board, we’ll sit down with the budget people, the grant people, and let them walk us through this. And that will be part of the onboarding. These questions are. I mean, it’s understandable that you are asking them, but they’re kind of pretty. I mean, they’re they’re standard issues the way grants happen.
Frode Hegland: That’s that’s that’s fine. And that would be great to do. Okay. So when it comes to the actual headsets, I do the one through whatever means the one that I’ll be using, I want to keep because it’s a historical device.
Speaker3: Right? Me to have.
Frode Hegland: Value in the future, it’ll have much greater value than it has now.
Dene Grigar: Keep the box and the whole damn thing to.
Speaker3: Like the first.
Frode Hegland: Iphone, there’s no question. Yeah. Whether there’s Adam should have one or not really comes down to the level of involvement he wants to have. And I do believe that. Adam, do you want to speak to that? I have initially, I felt you were just going to help Andrew get a transport mechanism done and then go have fun. But I later get the feeling that you want to work on this at least 3 to 6 months, maybe even longer. What is your perspective?
Speaker3: It.
Adam Wern: Well, we haven’t discussed that that much. You just asked me if I could kind of get the plumbing done, and I said, yeah, potentially. Let’s see if we can work out some details for that. And we were there right now, in my mind but I’m always experimenting and have been even though I haven’t even shown you everything that I’ve experimented with, I show one fifth maybe. So that’s why. And also, I’ve been talking to Mark over the last years because we did the we did the visualization of his data in 2D in a regular browser. And I it has been nagging me that we haven’t tried that in, in XR before. So that’s why I did the XR version. And I also kind of I’ve been doing more experiments for personal investigations, I think. So Regarding a longer time period. Or or more more involvement. Well, I was waiting to see where where this was going and where I, where I would fit in to this project or if I, if I was going to take the position of more of a muse, doing crazy things and doing experiments and user interfaces like I have been doing, because that’s very interesting to me. To, to kind of support the idea generation phase or to, to, to, to do the wild experiments that may get into the project or may not, but it may inform the project. So that’s another position to take or kind of a role to take to be. Yeah.
Speaker3: Adam.
Frode Hegland: This is not a dating game. We want you. So this is really for you to say how much you want to be involved, because your expertise over the years has become clear. Your honesty, even though it can be very annoying at times and your commitment to the project is clear. So you really just need to tell us what you want to do and what you want out of it.
Dene Grigar: And I recommend something. Adam. We have we have $1,000 a month for two years for consultancy. And so we can set this up in phases so that the first phase is to get Andrew on board, to get him up and running on this. In this project, my ultimate goal is to have an Frodo, and I both feel like we need to have something really important to show next fall. We need to be able to show it in September at the conference and in November at the symposium. So we front loaded Andrew’s hours so that he’s working more hours in this first year than the second so we can get up and running. And once he knows how to work with this stuff, I think he’s going to be fine on the hours he has left. And we can also augment his hours through other means, which I plan to do. So we can think of this in terms of phases, and we can think of this as a three month at a time, so we can set you up for three months of consulting starting February 1st to be February, March. So February to.
Speaker3: How about.
Frode Hegland: January?
Dene Grigar: Well, by the time we get the paperwork done, I mean, he could start working, but the paperwork today is the what, the 12th? It’s going to be end of January.
Speaker3: So so for.
Adam Wern: So it would be at least on paper, it would be a three month period from February. And for example.
Dene Grigar: Yeah, I’m thinking it’s going to take to the at least two weeks to get you set up. So that’s January 25th maybe. And that would be January to February, March, April. So that’ll get you to April 25th for the first round or 24th. And then we can evaluate where we are and re-up. Once you’re in the system, it’s nothing to re-up you. It’s just getting you into the system. Especially since you’re in Germany, right?
Speaker3: Sweden. Excuse me.
Adam Wern: It’s Denmark. Not so.
Speaker3: Sorry.
Frode Hegland: That was a Freudian slip.
Adam Wern: Sweden? Yeah.
Speaker3: Yes.
Dene Grigar: I’m sorry. Well, anyways. Yeah, the three month period. And then that way Andrew can get up and running. By that time, we’ll have our headset, and the testing will be a lot easier for him to work through. In the meantime, he’s he’s working and reading through the documentation for web XR. Yep. And getting getting wet.
Speaker3: Very good.
Dene Grigar: That sound, does that sound reasonable?
Adam Wern: Yeah. From a eagle eye view perspective. How would that work in practice? Do given an EU citizen with a. Well, I have a company as well. How would that work in practice? What what? And what is that? Is it kind of a consultant? Like, is it a company doing those services? Is that the preferred industry?
Dene Grigar: It’ll be an individual and you’ll get individual. We’ve got Frodo on a contract. He’s just got two questions to get fixed, fixed on it. But essentially he’s going to get paid in quarterly, like a certain amount of money at the end of each quarter. So at the end of like, say, February, April 24th, you’ll get a check for $3,000. American.
Adam Wern: And and would that be kind of what does that count as as in the US system? What is that? It’s the salaries.
Speaker3: Stipend consultant.
Dene Grigar: It’s consultancy.
Adam Wern: So kind of a business business transaction I guess, or transaction.
Dene Grigar: But it’s to an individual human being and a researcher.
Adam Wern: I’m a bit. I’m unfamiliar with that. Lands in the European or Swedish tax code. But that’s to be figured out. It’s. I’m sure there are thousands of people doing that right now, so.
Dene Grigar: Well, you imagine we have so many people we work with and well, you know, we we’re a Stem university, which means we have a lot of business in Asia. Yep. Working on technologies digital technologies to be one, one part of it. But we buy a lot of things from Europe. We work with a lot of people in Europe, a lot of people in Asia. So we do have, as we do, have knowledge in how to work these situations out. And so we can. But I would put I would put you in contact with Allison Logan. She’s our contact person and she will work with you on setting up this contract and we’ll do it. I’ll tell her it’s going to be in a three month increment, step by step, phase one three months. And if it if we need more time, we can up it a month, three months. We can go for a year if we need to. But essentially what we want to do is have something to show in September.
Adam Wern: So. And so that’s the kind of practicalities of it. The only kind of concerns that are a question mark for me right now is kind of the the design process, I think, because as this is kind of. And the sums are not that high. So it’s more of a passion, a passion project that may lead into other interesting work. And and you know, we’re here because we love doing the things. But that means that I also put higher. The higher standards on actually having a good work environment because I do it kind of part one. It is kind of part voluntary work, of course, even though you get to some sort of payment for it to cover some costs. So then, so then we need to have a kind of a clear design process structure and and expectations on what what to deliver.
Speaker3: That’s relatively easy.
Frode Hegland: First of all just stepping back a little bit before addressing that, I would like you to start 1st of January because it’s cleaner and you’re going to be putting in the times anyway. So a little bit of retroactive is just not a problem. You know, that makes sense, right?
Dene Grigar: We can’t do that.
Frode Hegland: No no no, we we can’t do that okay. We can’t do that. Dean says we can’t do it. Dean knows this much better than me.
Adam Wern: I will solve that. I think it’s one is administrative and one is. Hours.
Speaker3: Yeah. Worked.
Adam Wern: Perhaps we will solve that, I guess. I’m sure it’s not.
Speaker3: If we could do better.
Dene Grigar: I mean, if we could have done that, then Andrew would be paid through. I mean, I paid his salary through the through the 10th. He’s now officially being paid by the university, through the Sloan Foundation. But until the 10th, I was paying for him because we don’t do retroactive so well.
Speaker3: So.
Adam Wern: So what is I’m sure we’ll.
Frode Hegland: No. Sorry, madam. Please.
Adam Wern: No, no! Go. Go ahead.
Frode Hegland: What’s this thing in the chat here about Monday and Tuesday?
Dene Grigar: Well, Monday is a national holiday here. I’m just wondering if we’re. I mean, I’m not planning to have my my own lab meet.
Frode Hegland: Okay. I guess I can.
Dene Grigar: I’m happy to come to yours, and Andrew can come to if he wants to. Yeah.
Andrew Thompson: I’m available. I don’t have anything going on. It’s a holiday on paper, but I’ll still be working. So if we’re still meeting Monday, that’s great. I just based on the conversation, it sounded like it wasn’t happening. So I wanted to clarify.
Frode Hegland: But let’s not change because. Change. Yeah.
Adam Wern: And we said that Mondays will be general future of text meetings, a bit more general. And so it could be a good time to go back to that and to just let people talk generally about text.
Dene Grigar: And so I see money as theory.
Frode Hegland: So the big question.
Speaker3: Theory and practice.
Frode Hegland: Was about about the design process. I think we’ll be talking a bit about the design process on Monday just to get that clarified, because the way that I see it is there’s two things going in parallel. The first thing, as I’ve said many times, obviously you guys, the A team here, you work on the system to get PDFs from a folder on a computer onto an XR environment in the headset, because that has to be done no matter what. Right? So you in dialog with the rest of us, do that in a manner you think is not extravagantly complex, but it’s robust and can be done with things later. That’s super important. Other than that Alan’s taking charge of more of the user story. So he was highlighting this book the other day, which I picked up, which is actually really quite good, and it uses a lot of This kind of stuff. This is what Adam and I were talking about when you guys came in.
Speaker3: Thinking on index.
Frode Hegland: Cards. I think that’s a process we have to do. I think it’s good. But. Sitting in the along with Denny the kind of project manager seats. The need or pressure to deliver and build is of course very different from some of the people in the group who come in to discuss and just want to do things right in a different way. Well. So having said all of that, that means The design process. Adam. You really need to be a clear driver of.
Speaker3: So.
Frode Hegland: What I mean by that, I guess, can partly be exemplified by last meeting we had. Mark and I had an argument which was extremely unpleasant. Didn’t last that long. But Mark and I have a history, and I responded to him and what I thought was a direct and fun way. Looking at it afterwards, I was quite brusque. Which I didn’t mean to, but, you know, we’re close friends. Not everybody knows that. So it wasn’t pleasant. But what I’m saying is that if you either one of you feel that you’re not getting through in this group. When it comes to the technical, you are in charge. It is kind of your job to remind Dini and me that no, no, no, this is important. We listened as hard as we damn well can. But if we don’t, please understand. You own that.
Adam Wern: Yeah, there are two parts to kind of the technical design and that that is clear that it’s good to have some someone’s who is technical doing that. But then we have actually what we’re going to do. And the exact target group were imagining and what the jobs they that we imagine that they are doing. And down to the specifics today I want to look at spatial hypertext in this corpus. And I want to see if there is a good paper to to cite for my to, for my master thesis. And in some interaction design, for example, that’s a kind of a clearer thing that you could design around to see what kind of what interactions we want to do from that. Or, and also who will have the headsets on from Sloan and from in in Poland and so on. And what are they? Expecting or what? What could they be? What could they like or. And so we have some very clear idea on what we’re actually building. And that is of course your. Your thing as well I can’t do I have yeah yeah.
Speaker3: That’s going to be phase one.
Dene Grigar: I see phase one just getting the PDF in the headset. Let’s get that done. And if that goes fast then the next, next thing we could do in that first phase is to begin to add the second, the second PDF. So there’s two that you’re looking at simultaneously. And I know there’s some limitations currently I think Frodo, we were talking about this in the slack channel that you can’t really see two docks at one time in the application. Is that correct?
Speaker3: Right.
Frode Hegland: Adam. You can render many of them.
Speaker3: Yeah, yeah you did.
Adam Wern: So the risk of using Apple’s inbuilt operating system. And there you can probably where you’re just getting you could probably open up your your regular PDF reader a few copies of it I guess. So that’s, that’s a regular app in the, a mac app for iOS app in the vision. But what we’re developing is a web three dimensional web app, so to say. And there we are at the liberty of doing whatever we like, including opening 20 PDFs or maybe 100 PDFs at the same time. So that is not the limitation right now. And in fact. When I’m speaking to you. I have a thing I’m testing here, and I have. Here I have about. Well, 20 PDFs pages open in webXR that I’ve tested on the quest three. And it works. So that works. That kind of volume of texts.
Dene Grigar: Let me then mention that then phase one will divide into two phases. One a would be to get the PDF in the headset. Yep. B is to get two or more. And then I think the the thing that goes with B is to set up the, the kind of processes, the step by step directions so that we can have anybody do this with a headset. Right. That would be the outcome of the first phase. Phase two would then to start thinking about the the UX UI aspect. Right. Human interaction. Like what happens when the humans are doing this. What. And that gets us set up for manipulation and navigation and all those things that come in the second year. Right? We can really move pretty fast on this if we can get the first phase moving. So phase two could be like the UX, UI and and answering questions about design. And human experience, right? That kind of stuff.
Speaker3: Right.
Frode Hegland: This is probably under what Allen is trying to shepherd us into to decide these things. We all do agree getting a folder of PDFs, one or several into a headset. Absolutely. Number one. No question about that. And then we need to while we work on that, we need to do the user stories, which is exactly what Adam said and what Allen requested question of where to put them as an issue. I’m writing up my own present them however we want. And the user stories are going to be incredibly interesting, because one aspect of the user story is this notion of an academic going through their own library. That is a very constrained thing. Of course, specific uses are within that, you know, much more detailed. No question about that. But if we are also looking at the more important, bigger picture of helping people think, that is actually a different question. And that is what Adam and I have also been talking about of in the PDF document environment, we should also be able to have things that are thoughts, and this is not something to discuss today. All I’m saying is that the A team goes off and makes the uploading stuff. And as a community, we do user stories to discuss what should be done. Some of the user interactions are probably relatively obvious. We have some already in our Mac reader software that we could try to copy if people agree. You know, there are questions such as you come across a citation you want to know is this is this person who wrote this saying someone we can trust? You know, there are some specific things we can relatively quickly go to in discussions.
Speaker3: But yeah.
Frode Hegland: The user stories. But I do think that we can get a heck of a lot done by Easter, and then we can really get into what ifs and what ifs.
Dene Grigar: Yeah. So that means it begs the question, then how many hours do we want to put into this phase? On your end at Adam Andrews got 27 hours a week, four of which are meetings with Monday, Wednesday’s groups. Right. So that leaves some 23 hours a week. Yep. That’s you know, and so we want to work although you’re in Europe and we’re in the West Coast, we can we want to have time where he can work directly with you. And it seems like maybe Friday mornings may work. So he’ll work independently outside of you in my lab, and then he’ll work with you a certain amount of hours a week or a week. On the processes, right? So would Friday mornings work for you?
Speaker3: Like now. Like so.
Adam Wern: So I’m nine. Oh, yeah. Yeah, this could work. It’s to to here. It’s 5:00 or 530. So it’s a Friday starting Friday evening. But that would work. It would work, it would work. We have three, three small kids. So I have that. We need to keep that in mind. But I think that it would work. Yeah.
Dene Grigar: Okay. So we could say that then that means on Monday and Wednesdays we have the, the, the lab meetings with everybody. But on Fridays, you and Andrew will meet and go through the actual processes and getting the thing done. The most important thing for him to be able to set up that Apple laptop so that he can serve his own media to the headset.
Adam Wern: Yeah, absolutely. And I think it because if he does everything on his own, it will take a while to get to. So it’s very good to get onboarded by someone who’s been through the same process before in a few years ago. And so it’s it would really save a lot of hours for him to use a few of my hours, I guess.
Dene Grigar: So we’re looking at maybe maybe an hour on Fridays together. Two hours. Okay, so two hours on Fridays with Andrew and then the process at Frodo. So let me just finish this thought. If he if Adam and Andrew meet on Fridays, which is 5:00. Adam’s time for two hours. Right. They work out the processes. Then Andrew has Monday Tuesday to to build out what he wants to build out. And on Wednesdays will be show and tell. So not Mondays. Mondays are theory days, but on Wednesdays he can start to show what he’s doing in the in the project, on the project on Wednesdays. And that gives him time from Friday. And then we give him feedback, we talk all that stuff. And then he and Adam meet again on Friday. They do the next piece, Andrew works again, shows off the next Wednesday so that this is kind of how my lab works already, right? We do an introduction process, show and tell. And this puts us on a real good schedule. Okay. Andrew, does that work okay with you?
Andrew Thompson: Yeah, that should work. I don’t have anything else. Friday mornings. It’d just be contract work otherwise, so that’s fine. I can make time for that.
Dene Grigar: So that means you have 21 hours the rest of the week on the project. And so the work you and I need to sit down and talk about when you and I are going to be physically together to work and when you’re going to work at your own place.
Andrew Thompson: Right. Because we’re not sure what days on campus you need this week or this semester. This week’s going.
Dene Grigar: To be different. But I mean, in general, I’m on campus now this semester, Monday, Tuesday, Wednesday, Thursday, so we can work out, out, like you might say, ten hours a week in the lab, ten hours at home or 11 hours at home.
Andrew Thompson: Right now. This is probably a discussion for another time. But we also have the lab hours separate, right, for other things. Yeah.
Dene Grigar: You’re getting paid. So you’re getting paid additional money through my lab for the work you’re doing for L.
Andrew Thompson: Right, right. And that but that’s separate stuff.
Dene Grigar: It’s separate. Totally separate. And you’re keeping hours with Holly for that one. So you’re getting this income plus what? What you would be paid for the lab.
Speaker3: Okay. Yeah.
Dene Grigar: Your salary is more than what? The. What you’re getting through. Sloan. To your setting?
Andrew Thompson: Yes, yes I know, thank you.
Dene Grigar: Good.
Speaker3: By the way, Adam.
Frode Hegland: Your salary is exactly what I’m getting. So we’re in the same boat here?
Speaker3: Yeah.
Dene Grigar: Nobody’s making a bunch of money off this. Right now, I’m working for free.
Frode Hegland: This is not a cash cow. No.
Dene Grigar: Not a cash cow.
Speaker3: So.
Frode Hegland: I was lucky enough, as we all know, to work with Doug and Jeff. Rulifson was his head coder, and Jeff was. He’s probably the most arrogant person I’ve ever met in my entire life, but he also utterly deserves it. And he was talking about building up to the 68 demo where Doug would come in and demand absolutely ridiculous things, and it would be done. That is not us, obviously. However, I know that I can be very unreasonable as a designer talking to technical people. You know, my nature is to say, why don’t we try this? So I hope we will have a very good working relationship in dialog, but please be aware that I will probably every once in a while say, why don’t we build the Eiffel Tower on top of that PDF? Sorry, Dina, your picture is inspiring me.
Adam Wern: Yeah, and on that point, if I may interject, I think it’s good that you’re unreasonable with the with the user experience things or what? What could be done or so. But I think you have to hold back in, in kind of providing the solutions. The course well, especially technical solutions and even area technical solutions. But so we have I think it’s, it is a have unreasonable expectations of what could be done with the medium. And maybe we get halfway there. Sometimes we can do it or, and most times we maybe can’t do it. But I think it’s good to maybe not in scope. We can’t have an unlimited scope, but we may have some magical features that we
Speaker3: Yeah, yeah. No, it’s nice.
Frode Hegland: To hear you say that, because my experience with talking to programmers, sometimes I want something really simple done, and it just can’t be done. And then I mentioned something that would be basically magic. And they say, oh, give me half an hour and it’s done. You know, I don’t always understand the capabilities. But also right now, from the two of you, I really would like to have a feeling of how long is it going to take from right this second until all of us in the community can put on a quest or whatever and points to a folder on a hard drive, and we can see at least one PDF floating in front of us.
Andrew Thompson: I think that’s working right now. Like in the prototype stuff. It’s it’s not elegant. But like, just simple stuff of loading a PDF into the VR headset with no interaction that’s working. It’s not doing the server stuff. It’s just streaming through right now. The quest link for me. Which means we don’t have the interaction. So eventually we’ll have to find a way around that for testing. Technically, that’s where we’re at. You can’t do much with it, so it’s kind of useless.
Dene Grigar: Can I circle back because I only have 30 25 minutes before I get to class? I’d like to get I like to get some things finalized and then we can go back and talk the more detailed design issues. The thing I have down, I have four to do’s today. After my meetings this afternoon is set up. Adam in the contract so Adam will get that started. You’ll be hearing from Allison Logan who who oversees the contracts.
Speaker3: And and.
Adam Wern: She can answer all my questions or or investigate how the EU interaction works and so on.
Dene Grigar: Yeah, she’s getting really good at that after froda.
Speaker3: Okay. Yeah, good.
Dene Grigar: Although he’s not EU there, but we still are. We’re used to working with EU and if she doesn’t know the answer she can go to someone else. Yeah. The second thing on my to do list is to finalize the two questions you’ve asked me. Froda. And that is about intellectual property in your contract and your insurance. You have insurance already. You don’t need to purchase more. And then secondly, we’re making an open source item so the university doesn’t own a damn thing because nobody gets to own it. And that’s under the Sloan Conn Sloan contract. So I’m going to remind him of that. Number three. I need to get the laptop to you. Andrew. That’s ready to go. We just need to get.
Speaker3: So do you.
Adam Wern: Need just on on the copyright thing? Well, even to to establish an open source license, someone still retains a copyright because you grant with that copyright, or you have the copyright and you grant a license, open source license. So someone still owns the thing, and they could potentially. Grant different licenses or commercial licenses. So that situation may be we may need to. Just clarify that situation. Just to see can what can we do? We can do do do if I contribute. For example, if I write code. And this is not only going as its cross country, but we’ll solve that. But Who has the right to reissue new open source licenses? Can can the university still commercialize these things? Can I commercialize code? And the even even though it’s also open source? Because that’s a quite common scenario, that things are both open source and commercialized. That’s a very common thing. You can’t.
Dene Grigar: Foundation. We had to guarantee it was going to be open source, noncommercial, ever.
Speaker3: Okay. So there’s some collect on.
Dene Grigar: This loan side. So I think if they had read through the Sloan Foundation materials they would be able to answer their own questions.
Adam Wern: Yeah. So but and then, but I guess the the university owns a copyright then still. Even though they are only according to the contract, they are only allowed. The copyright is always owned by someone. Even though they are under. Yeah it is. You still have the copyright because.
Dene Grigar: The creative and you dropped in something in the, in our channel and that is Creative Commons. There’s a in the US we use Creative Commons and and when you register something in the Creative Commons then nobody owns it. It’s open. It’s open to everybody. And this is what the Sloan Foundation envisions. So I need to remind our university about this. So I think. I think I’ll work out the details on this, but I think it’s going to go through Creative Commons as per Sloan Foundation requirements.
Speaker3: Yeah.
Adam Wern: So even Creative Commons for the most of the licenses they are, the copyright is still retained because someone is giving that Creative Commons license. Unless you put it in the public domain, which in the US is quite different, is different. That notion doesn’t really exist in Europe, but we’re. Yeah. So so it’s a bit complicated, but I don’t think maybe it’s not a problem in practice. We make it open source, and that’s good enough for everyone, I think. Just.
Dene Grigar: I’ll clarify that today. Hopefully there’s no no closing of school.
Speaker3: I wonder if.
Dene Grigar: Andrew’s.
Speaker3: Like, oh.
Dene Grigar: I need to make a few about hours like the rest of the hours. How we’re going to work this out during this during the week. So those are the four to do list for me. Am I missing anything?
Speaker3: Okay.
Adam Wern: So just. Yeah. Just a quick for my hours. I have let’s say two hours a week. On Fridays. But when I am, am I expected to do if I do back end things like the server things?
Adam Wern: Where is that?
Dene Grigar: Your own? I’m guessing that you’re going to put in. If you look at the $1,000 and I’m imagining that we’re paying you $100 an hour. American.
Speaker3: Right.
Dene Grigar: That’s ten hours a week. So if you give Andrew two, then the rest is on your own time. Yeah. I’m not going to we’re not going to ask you to log your hours at. No, no.
Adam Wern: I have my it is we have the total. And then I do it with Andrew. And then I do make sure that things get done. Yeah.
Dene Grigar: You get another eight hours or so of your own time at your own own hours.
Speaker3: Yep. Okay. So, and we’ll work.
Dene Grigar: For three months. I’ll do a three month contract.
Frode Hegland: So an important question, Denny, the dates for when I’m going to come visit you. You basically need to tell me when suits you, and I will reply, and we’ll get the tickets booked.
Dene Grigar: I mean, that’s the next thing I need to get tickets done for Frodo. Yeah.
Frode Hegland: It’ll be in California on the 24th, but that’s the only preference I have.
Speaker3: By the way, as I.
Dene Grigar: Said, if you’re coming in on the 24th to California, then the next week makes more sense.
Speaker3: Yeah, the.
Dene Grigar: 29th, 30th, 31st and first and 2nd of February. February.
Frode Hegland: Just looking at the.
Dene Grigar: That’s a 29 through second, but you’re arriving on the 28th a Sunday.
Speaker3: Yeah. I mean, coming in.
Frode Hegland: Yeah, I probably want to come in on the 22nd, 23rd to to there and then stay anyway you may can.
Dene Grigar: Yeah. You’re going to California first. So you’re going to fly out of London. Let’s say the 22nd arriving California. You shouldn’t be too jet lagged coming through this direction. Going west is much easier. That puts you ready to go on the 24th. You hang out in the Bay area. The 25th, 26th. You’re staying with Rob Swigert anyway. You can come in on the 27th here or the 28th, but I need you here and by Sunday the 28th, and I’ll book your room starting that Sunday night.
Frode Hegland: Yeah, that sounds good.
Dene Grigar: And then I have you here that Monday, Tuesday, Wednesday, Thursday, Friday. And you could fly out Saturday morning on the third.
Speaker3: For the second.
Dene Grigar: Friday if you want to leave that on Friday.
Speaker3: I’m guessing.
Dene Grigar: Good Fridays are good days for me, because I don’t teach good the whole day to just work.
Speaker3: Perfect.
Frode Hegland: So I’m guessing the tickets will be booked from London to the Bay area, whatever airport is cheapest, and then separate tickets to go up north. Right?
Dene Grigar: Yeah. I’m going to say you probably want you’re not going to fly out Gatwick, you’re going to fly out Heathrow.
Frode Hegland: This makes no difference to me.
Speaker3: Okay.
Dene Grigar: And then we’ll put you to San Francisco and let you go from there.
Frode Hegland: So any Bay area airport is fine. Makes no difference to me.
Dene Grigar: And then San Francisco to To Portland. And we’re just we’re just we’re short. And my house is seven minutes from the airport. It’s I love my location. Perfect. And then PDX back to London nonstop. Perfect.
Speaker3: Okay. Well, no, no.
Frode Hegland: No, I would prefer to go via the Bay area because I think the tickets will be a bit cheaper because it’s just two return tickets. And also, that means if I meet anybody on the first few days, I can set up a meeting in a week. As I’m hoping Bruce Warren and other people will introduce me to all kinds of people who can help us.
Speaker3: Okay.
Dene Grigar: You’ll have it. You have a certain amount of money that you can spend that’s in the grant, and then anything that’s over that you’ll have to work with Allison to give her overage.
Speaker3: Yeah. No. When I.
Dene Grigar: Fly, we’re booked on the cheapest flights, and then I. I purchased my upgrades because I don’t fly in coach unless I have to. I usually sit in business class. I’m too old, too damn old to sit in the main cabin anymore.
Speaker3: I’m okay.
Frode Hegland: Depending on depending on the airline, I’m okay sitting in coach, but whatever’s cheapest making it around circle, triangle, whatever or two returns. That’s fine. But before you go, Denny back to the A team. The. The first requirement to make sure we all agree is. A new person comes to our community. We give them a link. I go to that link and they are able to see at least our PDFs, right? Yep. And they put on the headset and there they are in a default formation, whatever it might be. How long until that can be a reality?
Andrew Thompson: So you want it preloaded with the PDFs?
Frode Hegland: Yeah. The process that has been discussed in various ad hoc ways has been, Adam, this is me being tested. If I’ve listened to you and Brandon. Right. Somebody uses a Chrome browser at least, and they go to a thing. What we have made click a button and that button asks them to find a folder on their own hard drive. Once that is done, there will be a link that anybody with a headset clicks on. It will open those documents with the assigned code. Now, a really important thing is that the end user who has this link needs to be able to choose what code and what documents, because I could imagine Andrew, Adam, Brandel, Fabian will all be doing little experiments, which is fantastic, but it should be really, really easy to make sure that they can specify is it their own documents? Is it Mark Anderson’s big corpus or is it something else?
Dene Grigar: Didn’t we say that’s the first phase? Didn’t we already laid this out as phase one? No, there’s.
Andrew Thompson: There’s some details here. If I could say something and, Adam, I’m going to lean towards whatever you have to say afterwards. But I would say, what if we partially support that? Like if you are retrieving the PDF from an online URL, you can then package that with a link. But if you are hosting files locally, like you’re grabbing from your own library, putting it into your headset, that won’t travel with the link, because this gets very large and you’re having to deal with a dedicated server space as soon as you are sending personal files with every link, and it’s going to get bloated fast.
Frode Hegland: I understand that, and there is this is what Adam is aware of. I talked to my friend Bjorn. There is a beautiful get command as a get list. So that’s what I meant to say that because of the metadata we’re talking about we’re not talking about necessarily sending a huge amount of PDFs every single time, but it should be at least a list of what they contain. That should be quite instant. And then we can look at separately how to do the document. Some of our own books are like 200MB. So absolutely right about bloat. There’s no question about that. But without me giving the solution, as Adam rightly pointed out, I shouldn’t. The user experience has to be. The end user chooses a corpus of documents on their own system, and they choose the code they should run on it. So you as a coder and Adam and Fabian, all of you should be able to provide. Hey, look what I did this weekend type thing. And the end users should be able to run that on their own stuff.
Andrew Thompson: If it’s just for demonstrating prototyping and stuff. That’s simple, because we can have our testing documents on a server. That’s fine. If you want, like. Or are you talking about an end case where anyone can do this? That. That’s what my mind jumped to.
Frode Hegland: I believe we should think about in case where anybody can do this. I believe that when Dini and I are in Poland and we’ve done many different kinds of experiments, we put we show someone a laptop, desktop, whatever, and they see documents they put on the headset, and it’s those documents and they get all amazed because we’ve done amazing things which we don’t know, know. And they say, this is brilliant, how can I use it? Even though we’re not making a commercial product or anything like that. I do really feel that the how can I use it should be simple even now. To specify code and data.
Adam Wern: Yep. To all of this can be done without too much magic and too much programing. There is. And I have prototypes for all. All of this. Basically not the magic interactions, our prototypes. But as you know, prototypes are not the same thing as a rock solid or solid thing. And it needs to be refined and the refinement takes 90% of the time. So all the technology you talked about is demonstrated internally by me. I have most of it running as a tech X. Yeah, demonstration for myself, but that is not it’s not packaged for for you or for others yet. And that will take some time. And and that can be done in stages that But so what time frame are we talking about here am I? Are we talking that I’m talking about 1st of January and forward? Is that the kind of starting point for me and for this or.
Speaker3: Okay.
Adam Wern: So time frame.
Frode Hegland: Time frame wise I obviously would prefer this done Monday morning. All of it.
Speaker3: Yeah.
Frode Hegland: I’m sure we all feel the same. But the deadlines we have is the 24th of this month is the 20 the 40th anniversary of the Mac, and most likely I’ll be in California. I may even meet Bill Atkinson, who wrote HyperCard and people through Bruce Horn.
Speaker3: Whether I get your tickets.
Dene Grigar: Did you get tickets to go?
Frode Hegland: I’m waiting to find out, but that’s it would be nice to go, but since Bruce knows these people personally anyway, I aim to see them on the side, which it’s much more important. And I even got a special lens to do the video interview of them, so it’s really well done. So exciting. So, so the point is that when I’m in California, I don’t want to be another vergana person. I want to be able to. If I’m at a Starbucks and I meet some Apple person or some other person and put on the headset, even if it’s super basic. I just want them to feel that we figured out the reality of, here’s the stuff in the computer. Now it’s on the headset. Yeah, maybe you can move the document to read it. That’s it. Because then they can see we have a starting point.
Adam Wern: Okay. So as you mean in 12 days, you want want this. The problem is that what we’re the timeline we talked about today is that I made this.
Dene Grigar: This changes everything. So phase one is to give Frodo what he wants in the 21st of January. They start getting paid. And that would be the thing we talked about earlier. So this is different.
Speaker3: Yeah, I mean.
Dene Grigar: It’s impossible to get it done in 12 days, and.
Speaker3: That’s what I want.
Frode Hegland: I’m not saying it’s reasonable or achievable.
Dene Grigar: I do want to say I have to go in seven minutes. I have one more question before I go. Frodo, this is for you. You said you wanted to get a quest three to Alan. We don’t have the money in our budget for this unless you decide to do what I did. And take your goods and services for one of your headsets and convert some of that money to a quest three. Okay, then we can’t do that.
Speaker3: I’ll find a way.
Frode Hegland: To just pay for it myself. Okay. So now I understand that, and I really appreciate your understanding about these things. Work to see what we can and cannot do. So thank you. Yeah.
Dene Grigar: Wish we had more wiggle room. I wish we I wish we had more money in the grant. I say this every time I get a grant. It’s like if I had just another $500.
Speaker3: Yeah.
Dene Grigar: Okay. And then that that begs the question. The final thing is getting our headsets. So Allison’s going to do the best she can to get them bought, but once we buy them, we got to make sure that you get access to the information you need to be able to customize your headset to your body.
Frode Hegland: Well, that’ll be fine. We have Greenville. We have Bruce. We have all kinds of people with deep knowledge. But the thing is, if she doesn’t want to wake up at five in the morning, which of course is something we cannot demand of her, we’re not going to get it. They’re only making something like 80,000 the first run. So people who want this will get it. And then there will be many, many months of delay. So unless she can think of a mechanism through which obviously I can’t, I don’t want to use her password. But if there’s some way to do it, it’s got to be done within the first 15 minutes, half an hour. Otherwise it’s just not going to get it.
Dene Grigar: Well, I’ll talk to her about that. Maybe she’ll let me use the p card myself. Since I’m university employee.
Speaker3: And I can buy three.
Dene Grigar: Right away.
Frode Hegland: Because obviously I will be in America around that time for pickup, should it be necessary?
Dene Grigar: Okay. So I’m going to bug out. I’m going to see your dissertation member.
Speaker3: Yeah.
Frode Hegland: Please ask him what the hell.
Speaker3: What the hell? That’s my full.
Frode Hegland: Question.
Speaker3: What the hell? What the hell?
Dene Grigar: And then I’d like for us to have. If you guys can work out. If you can. Phase one is possible for the 24th. And then we already have phase two. We need to get this typed up. I’d love to be able to present this to our team on Monday. Here’s our timeline for the first two phases.
Frode Hegland: I don’t think we can because of Allen. Allen is putting a better structure on the design process of this. So I think we need Monday to figure out what the design process will be. You just sent me an email a few minutes ago that I obviously haven’t read, asking a few questions about this, and I think Andrew and Adam needs to be deeply involved with how we choose to do that.
Dene Grigar: The design is the next phase. What we’re just talking about grunt work, getting the stuff in the headset that doesn’t involve Allen. So Allen comes into play in the phase three.
Speaker3: Yeah. Yeah.
Frode Hegland: Exactly. Exactly. Yeah. Sorry. Of course. Of course I agree with you as well.
Dene Grigar: So we’re going to come up with the timeline to all the way to the 24th of April. And then Allen’s going to come up with the timeline for the phase three, which takes us for whatever he envisions that to be.
Frode Hegland: Let’s talk about that on Monday.
Dene Grigar: Well, actually, I’m going to I’m going to say this what I’d like for us to be able to do. Is be a little bit more organized in these meetings so they don’t turn into. Disagreements. I mean, you and I, especially Frodo, need to be on the same page as Project as Pi. That’s the point of being a pi is that we have kind of a vision of how things operationally should, should happen. And have a plan in place and we take feedback, but we need to start with, with kind of a structure. And then people add to that structure when it becomes a free fall discussion, nothing gets done. And we really have hard deadlines.
Frode Hegland: Yes, absolutely. That is why we’re separating the uploading of stuff from the what are we going to do with it stuff. Yeah. What we’re going to do with the stuff structure is currently in Alan’s hands under our direction. And so I think that on Monday we need to make sure that we agree with how Alan wants to run that and that we can contribute to it. We’re not going to contribute as a we don’t know what’s going on. It has to be based on what we’ve already got approved from Sloan. It has to be based on what we want to do. So yeah, I think next week will be very productive Monday and Wednesday to really firm things out because my little argument with Mark over a disagreement being important, we can’t afford that kind of stuff again.
Speaker3: No, no, no.
Dene Grigar: And we need to have our ducks in a row. And even if you and I have difficulties, they need to be behind the scenes. Nobody knows.
Speaker3: Well, I mean, Andrew, you’ve been.
Dene Grigar: In the lab a long time with me, and you’ve never seen us. Anyone in the lab? Holly and Greg and I and Richard. Ever. Ever argue vehemently against each other about issues. Nothing ever becomes difficult behind the scenes. We tussle on on issues, but it never bleeds into the the larger lab environment because it hurts morale.
Adam Wern: Okay, yeah. And it will be especially important here as we have a remote team. These meetings are so important because it’s the only time we will. Well, of course we can talk offline, but in a larger group this is what we have. So we really need to put a great value on this and take as much as possible offline. That is not necessary in the meetings. Having video calls one on one.
Frode Hegland: Yeah. No, I completely agree with that. And I do feel the need to
Frode Hegland: Sorry.
Speaker3: That was Bruce.
Dene Grigar: Hi, everybody. I’m going to my meeting.
Speaker3: Yeah.
Dene Grigar: Thank you.
Speaker3: See you. Danny.
Frode Hegland: Yeah. So just in front of you guys, I just want to explain a little bit of the Mark Anderson situation. So I know Mark really well. Mark is autistic Asperger’s. Most people wouldn’t tell. Of course you can tell, because he’s very sociable, very well liked and very good with language. So he’s kind of the opposite of your normal Aspie, so to speak. And that’s absolutely fine. We have a relationship where he will waffle on forever, as I might often, and I can interrupt him and stop him if he’s going too far, but he will then instantly say, hang on. No, I’m actually doing fine. You know, face to face that works. But sometimes he doesn’t see someone else having the emotional investment in an issue. And that can be very difficult. When we were in Rome on the day before the last day. Andrew. We had dinner. Mark said something that I found really incomprehensible and quite offensive, and that’s fine. It can happen with friends. But he then proceeded not to want to discuss it. And that’s the problem. So in our little meeting earlier in the week where he said a thing that was based also on what he’d written in slack, and I said, oh, we disagree on that strongly. I think that’s wrong. And then he cut me off. The cutting off thing is something we must be very careful if people are going into different areas. Yeah, we say that’s a discussion for another group. Yes, that’s important, but this thing of just saying your position isn’t valuable or I’m going to I’m saying it is wrong and we’re going to move on. It’s annoying.
Speaker3: This.
Adam Wern: But on the same. And I’ve known you for a while now at. At least in the meetings. And sometimes you take the. The critique on ideas as personal critique that you haven’t. And I think that is something you could, on your end as a project, lead to kind of tone down a bit, because I know it’s an emotional response. Of course it is. And I know that it’s natural and human to do it. But as much as you can to try to not feel that any critique on an idea is a critique on your abilities or you as a person, that would be wonderful as well. Because sometimes, even when I’ve said things, you take it personal instead of taking as a critique invalid or valid on the idea. And just so we have that as well. But I agree with you, because Mark is getting provoked as well, and he’s kind of getting agitated. And then, yeah, all we had that situation a few quite a few times.
Speaker3: Yes.
Frode Hegland: Absolutely. And I. I was angry at my son two weeks ago. He’s six and a half, and it’s the first time I’ve ever been angry with him. Right. We all have different kinds of patience for different kinds of groups, right? And Mark has a great ability to get under my skin because he likes to complain that, you know, no one takes his different kind of cognitive ability seriously. And support teams like Mark, for fuck’s sake. How long have you been coming here? We all support each other, so? So that is something I really have to watch now on your comment on criticism of ideas and so on. Adam, you can be very brusque. You know, with Peter two weeks ago, you said, Peter, you’re suggesting things that are blah, blah, blah. You know, it’s like, whoa, that was, you know, machine gun fire. We all have to be open and clear, but all have to watch that. And yeah, and I fully understand that my job now is 90% human. It’s about making sure we have a happy community. You know, I completely understand that. So when it comes to specific ideas.
Speaker3: You know.
Frode Hegland: Sometimes, you know, let’s get in the playpen and duke it out. Yeah. But hopefully as nicely as possible.
Adam Wern: Yeah. And. Yeah, well, I’ll try to watch that, but we need to have the discussion with Peter, for example, because he suggests things that are not related to XR at all. He suggest his favorite technologies that are not at all relevant. It’s a kind of irrelevant. And he can’t. He has. Yeah, it will be problematic if those get inserted into every conversation. Choosing technologies that are not even XOR technologies or or usable or.
Speaker3: Yeah, no.
Frode Hegland: That’s that is an issue. But the thing is.
Speaker3: You know.
Frode Hegland: Sitting in California as I used to, having, you know, with a beautiful temperature and coffee and talking to people about saving the world. Maybe many of them. Doug Engelbart.
Speaker3: Sorry Andrew.
Adam Wern: Have a kind of off topic discussions if you really want to.
Speaker3: Yeah.
Frode Hegland: Anyways, so. So what I mean by that is
Speaker3: Sorry.
Frode Hegland: Bruce Harness keeps interrupting me. You know, my head’s going all over the place. We are actually doing something here, right? We’ve looked at all kinds of If things in the past, but now there is a little bit of payment for you guys, which I think is really important. I find one of the biggest problems in our society is that people who try to do good things have to do it on a volunteer basis, and that means that people already have enough money, are the ones who can work on it, and that’s just stupid. Yep. Right. So that’s that’s really important. But but the other thing is, because even you guys are volunteers to an extent, you know, you’re not getting paid enough for this. And the other guys are. Oh. Sorry, Bruce. Thank you.
Speaker3: I’m on a call.
Adam Wern: On a on a on a side note here. Or a main note maybe. Andrew is are are there any details of technical or conceptual that we should take now to just to to just help you or us work together?
Andrew Thompson: Are you talking like, theory, or are you talking like practical code?
Adam Wern: Yeah. What are your main just to get your more onboarded at this very moment, even though even if it’s just 5 minutes or 1 minute or just a getting to know what what kind of are there technical or conceptual or anything that you that you would like to.
Speaker3: That’s the key question right here.
Andrew Thompson: This so kind of it’s I guess I should give a very, very brief background here because we don’t really know each other. No. And like my role in this, I guess I, like Denise mentioned, I’ve, I’ve worked in VR quite a bit on development stuff, and I’ve worked in web quite a bit. But I’m typically on the front end of web. And for VR, I’m in game design, so I’m working inside engines. So I’ve never done XR before this. So I’ve been working on it the last couple weeks learning as fast as I can. So there’s definitely a gap in my knowledge for this. So in some cases I can know what I’m talking about, and in some cases, I don’t know what I don’t know. I’ve got. Something’s working. Forgot pdf.js loading in the pdf. It’s then using the webXR to throw that over using three.js for that. I think that’s what we agreed on. Throw that over to just a simulation, which I’m not bothering with server stuff because that’s back end, which I don’t know as well. And that’s going through the quest link for testing. I don’t have interaction yet, so that’s what I was going to work on. You used to.
Adam Wern: When you used test quest link, are you basically then running a browser locally that you stream a stream to the headset? Yeah, that’s.
Andrew Thompson: Pretty much it. It just so.
Adam Wern: You don’t open the Oculus browser in the headset to access that code in any way? No, that’s that would.
Andrew Thompson: Be what the server would need. And I don’t have that set up. I run a local server just for just so the code works properly.
Adam Wern: Yeah. So what I would suggest then, because I do that as well. But I access that one most often through cable. I know Brandel and and Fabian have set some set up some sort of wireless connection between them. I use a cable because it works well enough for testing for me, even though I’m tethered. It’s not perfect, but it’s fine. So with that.
Speaker3: When I am.
Adam Wern: When I’m in developer mode on the quest headset, it automatically tunnel any local host request on that headset to the local host on my. Laptop, which means that it looks as as if the server ran locally on the headset, but it actually runs on the laptop. It tunnels everything, and that makes it so. I’m using the quest browser. To run the code. And I think that’s where we where we’re going as quickly as possible for you, because that is to in order to make it fully standalone and to to make it even though I use it tethered, it is theoretically standalone. The server could be online or it could be or we have those alternative means of streaming, streaming things which is kind of WebSocket web RTC things that makes it even more disconnected or, or peer to peer or so. But that’s the next step. The first step is then to start using the quest browser in the headset and to get that up and running, because that is how all of us are using it. And that’s the end goal, I think, to to make it more standalone and not be not depending on quest link or anything like that, because that is kind of.
Andrew Thompson: Yeah, we’re doing that to get the finger tracking right because that part doesn’t run through the quest link.
Adam Wern: Yeah. And finger tracking, but also kind of the idea of stand that it’s a bit more standalone. Or it is standalone. Apart from getting the app code or the JavaScript things and the documents is standalone, but you need to be served that in some way. But apart from that, it’s standalone. So you can just go to someone could potentially then go to an URL with the request browser directly from the browser and get things served the whole package to the app, so to say the XR app and the potential documents from there. Or some other place, but that is the kind of the immediately immediate goal, the north goal for us soon to skip the whole quest link thing. Right.
Andrew Thompson: That’s that’s what I’ve gathered from the discussions I’ve been hearing. And that’s definitely something I don’t know how to do. No back end server stuff is not my area of expertise. I could learn if.
Speaker3: I need to.
Adam Wern: Yeah, but you basically, for almost all my prototypes, I don’t have any back end at all except for serving. Except for having a web server up and running, serving those HTML, JavaScript, and PDF files and an occasional JSON with the metadata and or kind of the database, so to say. And so that’s for read only things. As soon as we want to write things back. But that’s a step two saving workspaces or saving annotations or so or something. We need some sort of saving mechanism, which could be. Yeah, we have and I have another solution that doesn’t really require a server. And in the same way.
Adam Wern: Should we talk about that now? Did you get that kind of chrome thing? Chrome idea we had?
Andrew Thompson: A chrome chrome idea.
Adam Wern: We have an idea. Or I suggest an idea if we want to serve files from a from a private collection that you have on your laptop, for example Chrome and some other chrome edge and some other browsers have a JavaScript API for serving local folders. So you can request if user clicks the button on a web page, it could actually open up a file dialog and you pick a folder on your laptop or, yeah, on your computer, which may contain PDFs, for example. And those can then be served. Or those can be accessed from that JavaScript page or that web app that you have locally or so those can be sent over web RTC, this kind of video and data connection that we use all the time in different settings. When when we talk on zoom and the app, the web app client zoom they are using WebRTC to send a video, audio and data between between basically browser tabs and, and other browser tabs. So what we’re doing then is that we. However, we have a web page under a browser tab serving all the PDFs to the headset. For the time when, as long as the laptop is open and the web that page is up, you can show whatever you want over that web RTC. And that’s a peer to peer connection and completely encrypted. It’s encrypted. It has some it it has. Low latency. Because it uses the same network, it established the closest connection it can get between two browser tabs, basically one in the request and one in the on the laptop. And you can serve everything there. And that’s very that’s kind of a big shortcut and also a privacy thing that you get. It’s really using the technology to its fullest and it works. That’s and that’s.
Andrew Thompson: Brilliant.
Adam Wern: And it can serve very big files because it’s only traveling within the local Wi-Fi. If you are in the same room as a headset. So that’s a kind of a shortcut thing that we’re thinking about potentially at, at one, at one point we could do a kind of bundle a browser and do an electron thing app that does all of that. So it’s a it feels like an app on a, on a mac or a windows, and it does everything automatically. And we could do that if we want. But for now, I think just picking a picking a folder that you want to read. Folder PDF. It could be thousands of PDFs. I, I I’ve tried it with the full collection and it works. I can pick whatever document and stream it over and and render it on the quest browser. Okay. And this is not super quick. For some reason it feels like a web server is lower overhead. But But it’s not slow either, so it’s You can.
Speaker3: It’s.
Adam Wern: Good enough for this.
Andrew Thompson: Yeah, that sounds like the perfect solution. With the one exception of what Frodo was talking about earlier. Where you then send those to someone else with a link. And if we’re just sending things locally, you wouldn’t be able to send them globally, right?
Adam Wern: Not Well, potentially, if I open my server and I. And I put the code somewhere online so you can reach it, you could potentially come in and visit my library that I have open here across the globe, because WebRTC doesn’t care if we’re next room or if we’re next to each other, or if we are across the globe, but the latency and the capacity will be a bit slower. But as long as your browser.
Andrew Thompson: Is still open, right? Open.
Adam Wern: Yeah. So so the other scenario is that we want to go to a known collection, for example, a demo collection or a workgroup group collection that the our academic work group has collected documents together. And then we need to to serve that perhaps from a server instead. Yeah.
Andrew Thompson: That sounds very reasonable. But that server wouldn’t be just open for any user to know. No, to.
Adam Wern: Know. And then we didn’t need authentication of some sort to get into that page and to, to start requesting documents. So that’s another architecture. But I don’t for a, for a face a or the first face, we’re not really doing a shared work group. And with authentic heavy authentication or user lists and so on. But but what we want to do, I guess, is to have a demo collection online. So you could just visit a URL and not even care about. Is setting up your folder or picking documents here, just a demo URL. And for that for that version, we could just use a standard web server serving things from that server. They have the PDFs, they have the index dot HTML and the and the our great web app.js. Running, sending everything directly to the quest browser. So in theory.
Andrew Thompson: That all makes sense to me. Now that you’ve explained it I don’t know how to do that. No, no.
Adam Wern: No banana and I have no problems with. I like when people are just say what they don’t know, and I don’t Yeah. So just say ask the most stupid questions. I won’t judge you. I’ve been there recently enough myself, so I can remember some of the I didn’t understand two years ago, I didn’t I haven’t hadn’t seen a single webXR line, and it was totally not, not even 3D programing either. So everything was new to me there. But you have lots of experiences in similar fields, so we could. Just when you get to know how this exact system works it will be quite quick, I think.
Andrew Thompson: Yeah, I think that that’s true. The concepts will stick, I hope, quickly. It’s just the actual implementation.
Adam Wern: Yeah, yeah. The implementation. It’s hard to actually just get a hold of or understand the webXR. How do you where do you put the data? Where do where does the computation live? Where is it served? It’s a bit.
Frode Hegland: Now I need to go.
Speaker3: Yep.
Frode Hegland: Do you guys want to use the next 40 minutes? I’m happy to leave this running. Of course.
Speaker3: Do you want to go? Well you do.
Frode Hegland: It’s really up to Andrew’s desires and needs and also Adam’s time. I don’t know how you want to use the rest of the time today, if at all. Excuse me.
Andrew Thompson: Since this is like a first meeting, we may not need the full time. Just whatever works for you. Adam I know that there’s still quite a bit of, like interaction stuff I can also work on right now, so I’m not lacking stuff to do.
Adam Wern: What do you prefer? Do you want to. To work now? Right now work with interaction or do you have any more kind of things, your ideas or or concerns or questions or anything? That would be nice. And I think that.
Andrew Thompson: I think The order that I try to develop stuff, I feel like I need to get the VR interaction working. As the next step. So that’s that’s what I would want to work on. If that makes sense in the workflow that you’re you’re rising yet?
Adam Wern: I really think. And when you see an interaction, you mean something more tangible for yeah, some graphics and hands and or something that you that it makes it tangible is that.
Andrew Thompson: Interaction I might even worry about, like the visualization of the hands later. I just want to get. So right now it’s just a PDF sitting there in VR, you have to click the mouse to do anything. So I want to get like actual buttons and then grabbing and stuff like that. All the stuff that’s built in to the engines that I’m used to, I want to like, yeah, get.
Adam Wern: Yeah. And we have to Webcore and three.js does some things for you. So it’s not, it’s not like from the ground, from the very bottom up. But but many of the gestures are not for free. You have to kind of for, for every, perhaps every frame or so, you have to to, to test whether whether the thumb index bone the position of that meets if you want a special special gesture, you have to kind of match them together. What’s the distance between them? If it’s close enough, then you trigger some sort of event and you keep that. So.
Andrew Thompson: So you have to manually hook up interactions.
Adam Wern: Yep. But that’s the it’s fine and it takes some time. But it also great beauty of it. Some I think a click is there. So if you have a basic click unless you override it. But you can. There is a built in click and maybe a there are two gestures built into webXR the webXR standard, I think click and the squeeze.
Andrew Thompson: Okay. And all we need for now.
Adam Wern: Yeah, it is. So for most of my experience, unless it’s a real hand interaction thing with the high fidelity, you could just use those for, for like moving pinch basically a pinch. There is a pinch start I think or select start which pinch triggers that and and a select end which. So when you so with that you can do very many simple interactions I think And there is also, I don’t know where you are in the process. There is when you enter exile, there is there are events fired, and with those events fired, it could be nice to to make sure that the 3D version you have is placed in a good space. Space relative your body. Or the room you want to.
Andrew Thompson: You haven’t done any of that yet. It’s all zero.
Adam Wern: Zero. Yeah, yeah. So that is also something to keep in mind when you whenever we enter VR mode in a webXR or AR mode, when, when you see your own walls. And I really like that, but maybe it’s not for this project. Yeah. You need to move.
Frode Hegland: I have to go downstairs and check on the family because we’re supposed to.
Adam Wern: Yeah, I do that. We leave this running and we will talk a few more minutes and then we’ll drop out. You could you can pause the recording now if you want to stop the recording, because.
Speaker3: It doesn’t matter. By 930, I’ll.
Frode Hegland: Come back up before I leave, so I’ll say bye. Yeah?
Adam Wern: Yeah. So so let’s see what Brandon said. So sent some Excel templates. Have you tried those?
Andrew Thompson: Do you mean the boilerplate that he posted?
Adam Wern: The boy? Yeah, yeah, the boilerplate. And also on three.js. The examples are quite good to see the the the different events. There you have the pinch events or select events in some of the examples. Yeah. There you have to jump between the different examples because they are quite different and use different things. And you can also.
Andrew Thompson: Find I found like what seems to be a bug with, I don’t know if my browser or webXR itself but I can only run one simulation before I have to shut the whole browser down and reload it, and then I can run another. And I don’t know why that is.
Adam Wern: You mean a simulator?
Andrew Thompson: Like one of the web pages actually entering VR? It only works once. Then I have to restart the browser.
Adam Wern: You can’t leave that mode and
Andrew Thompson: Just. It’s weird, but, I mean, I found the workaround, so it’s not stopping any work.
Adam Wern: I don’t have that problem. So it’s
Andrew Thompson: Probably just from me.
Speaker3: Well.
Adam Wern: No, that shouldn’t be the standard.
Andrew Thompson: Okay, well that’s good. I wanted to make sure that wasn’t just like a new issue with webXR or something. So.
Adam Wern: So how I’m developing right now is that I have S3. Three.js is also it has quite good support for doing 3D and a flat browser. So many of the things like positioning things I do in on a regular laptop and when it’s interacting interaction, you have to keep it to kind of halfway up on your face doing a reload web page and put it down in the browser. So it’s kind of this down and up. But there is a and you also get the debugger when you some things are must be debugged for hand tracking and so on. Back to the browser. So. You will need ADB. Which is the Android debugging console installed. Or that’s a good thing to have at least.
Andrew Thompson: Yeah, I may have installed this in the past, but I’m sure it’s out of date. It’s been so long since I used it.
Adam Wern: Yeah. So you need you need to update that one. And and. And if you go to Chrome colon the devices. Okay. Let’s see. Chrome. If you connect, if you connect your Oculus with with the cable in the beginning, at least you can get the away from that. But if you go to Chrome you may have to. Let’s see. Can I send you this? I send you this on slack. I think I will do that. Or do you prefer to use iMessage or do you use. What’s your preferred method of communication? Just slack keeps.
Andrew Thompson: It all in one place.
Adam Wern: Yeah, yeah. That’s good. It’s hard to get your. Of personal messages mixed up with So let’s see.
Speaker3: Are.
Adam Wern: So if you go to. If you go to this page do you use Chrome, by the way?
Andrew Thompson: I’m using it for this. I normally use Firefox, but for this project it’s Chrome.
Adam Wern: And when the when the quest headset is in developer mode, is it in developer mode? You know that. You know if it is.
Andrew Thompson: I believe so. I turned on all of the development stuff I could find when I first got on. It’s it’s blocking this URL.
Adam Wern: Now. Yeah. You have to copy it and copy it, and then you can’t click it because it doesn’t allow hot copying.
Andrew Thompson: It blocks it to. I have to like, type it out.
Adam Wern: That’s so okay. So it really, really protects you from phishing attacks here. Yeah. The do a bookmark on that page because there you find will find. You will find when all the settings are right in the headset and you have a cable connected the headset will will show up there and you could remote debug all the pages. From there you can see the error messages from the headset browser quest browser in here.
Andrew Thompson: Like the console log.
Adam Wern: The console log will get streamed over to to to a window here. And you can you can also reload the page from there. So so command click R will remove the quest browser in there, which is quite kind of convenience because you could kind of change a code press command or there and then take on the.