 Okay, welcome to the 25th of September my profit deficit So we're roughly halfway through our sprint Maybe some small children in the background So let's let's go through and give any status updates especially focusing on any problem areas Chris Bayer You are in the center on my screen. So The center square Yeah, everything is nothing has blocked me right now I spent today Doing some here and there type tasks Like I Kevin on to On a Jira, and he's poking around there Change to the Docker file or white count skills getting that done I'll probably do the discord. Well, I probably shouldn't do the discourse thing today So I'll do that next week fit But yeah little things I did I did also change the conversion process it to To not SSH out as much it now runs a thousand records in a second In seven or ten minutes. So that's that that's an improvement So, yeah Next week will be starting on I've done some work on good. I'm getting up to angular 10 I've not gotten all the way there yet. If there's a several step process I need to do I'm working on that and we'll continue working on the tag or next week. Okay, Ken Oh, I'm already unmuted Yeah, so I did a full request Chris. Did you have just that one or were there multiple full requests for me? Well, you reviewed one the other day and I merged it and then you reviewed the one I set up this morning Yeah, so that's good. So I'm good. I'll merge. I'll merge it today Yeah, so that's the last one I have for now I think the conversion code is all in just a Python notebook right now. It's just throwaway code. I don't want to commit it Okay. All right, so I did some full requests. I also Got the my PI forward just now flashed with the pie crop image Kivvy image so I'm going to go and see tasks on that and I also got a chance to look at the VF control USB commands and sent an email to you kept on Michael regarding GPIOs and I to see us these five of us are all available Through those commands. So that that makes it a lot easier really the only issue outstanding with the interrupts over the USB port and That's not absolutely necessary to get everything working. So I think we're good if we choose to go that route in the future Yeah, and that's where I'm at Okay, great Yes Yeah, I went on a bit of a PR closing rampage And got our call at least down quite significantly We're now at 20 open PRs with only a small number from prior to 2020 And Yeah, the the road map stuff that I that I posted in in chat Which was halfway through responding to to Michael's comments So I'll finish that up After this meeting Oh I had it I had a look at the Exposing the the logs from our void com for test runs and Chris interested in whether you think we should go That deeper route. I don't know if you got the message on the ticket Yeah, I like the idea of being able to show the logs for each Test, you know related to that test. I guess really good idea. So if we can figure that out That'd be awesome. Otherwise, I dig through thousands of lines, but yeah, yeah, we'll see it's it's worth a look. I think So essentially the allure reporting framework that we use as a have an attachments Feature and so you can attach either just strings of text or attach actual documents like specifically to Each test and so we could theoretically attach the At least a portion of the microsoft logs to the test where it failed and so you can kind of see or just attach the entire log Yeah tricks gonna be parsing the log and figuring out what what part to put with what test. Yeah Anyway, so I'll See if I can figure that out the backup plan is just to Attach the entire logs to Well package the entire logs up with the allure output so that when they get pushed across to the reporting To the reports host then that goes along with it and then we'll just provide a link in the PR comment or in the email or something Or I guess we could upload it to something like Zero x zero dot xt that we use for some other things Which is just like a new pace pin type thing But we shouldn't need to do that Yeah, so roadmaps the ass stuff and the logs That was my main stuff in terms of the I In terms of the Also did a we did the Mark to update. Oh and another blog post about our partnership coming partnership with Talk socket, that's what they call talk socket So that should go out today theoretically anyway and Review a bunch of PRs and thumbs and bugs and fix those That's me Okay, so is there anything in particular that you're waiting for from anyone else at this point? Have you No Well in terms of the road map I guess If anyone has opinions then then jump forward otherwise, I'm just gonna keep moving forward with it and I think so specifically about your stuff Michael I feel like the The how the ability to add separate boards for you know sub projects I think is really good and I think we should use that, you know Where there's a bigger piece of functionality that's gonna obviously have multiple parts to it I feel like the community verse Internal team like splitting out to roadmaps for an internal road map and And a community road map kind of creates more distinction than is necessary at the moment there is a label that's Applied to any PR or issues. It's like help wanted and so that kind of indicates that you know Either someone's working on it and would like additional help or no one's working on this at the moment So it's it's available for people to pick up And so I've you know in terms of If an issue is being worked on by me or if the issue is being worked on by Jarvis or by, you know, okay, or you know, another community member I don't know that we Get a lot of benefit out of splitting that onto a separate board Okay, yeah, it wouldn't necessarily be a separate board Maybe it's just a different column or maybe, you know, maybe it's a label like you said They just wanted it to be clear to the community where our focus is You know, what our community Projects well, and that's kind of intended with the so one of the one of the columns is Is Nice to have and that's in the intended to be Anything anything where It really is a it's not going to be a focus for the internal team at all. So, you know In you know, we're not going to work on it, but also It's probably going to take it's probably going to take a slight backseat even in terms of PR reviews and things because it's not You know, there's just other other priorities where we if we if we need to review PRs that we're going to get to those first And I know it's a really hard thing too because obviously people are putting a lot of work into into different things like, you know, there's a guy that's That's been porting things over to windows, which which is huge and he's been making a lot of really good progress. And so I really do want to get to, you know, everyone's contributions, but there's just there's just a limit to how much we can do in a day. And so We need to stay focused on the things that are going to get the mark to out the door and then get to windows support when we can get there. Will that be a separate repository? Well, it spans a whole it spans, you know, precise and a bunch of repositories. It's just he's sort of created a central issue for porting Microsoft windows within the Microsoft core. So it's just sitting there for a moment. But yeah. Okay. All right. Well, thanks for that. I'll, well, yeah, I'll take a look and give any further feedback as you know, as you make progress on that. But I like what I've seen so far. So I think that'll be cool. Well, I think it's going to keep evolving over time. Like, you know, they're all living things. All right, Derek. There any updates for us? Yeah. Sorry, no video. I've got some internet issues today for some reason. Yeah, so I got a couple of things that I've got in review. I just invited today for Chris may not have seen it yet, but let's try to meet up maybe Monday to review the GUI. Made some of the changes we talked about for the first, the first round of the tagger. Sorry, the GUI tagger. And I've had a little progress in the sourcing side of things. I got the DSI display at quoted at $20. Which is what we're using right now. So one of the things we've been discussing was, you know, we can save some money if we switch the DPI, which uses the 40 pin GPIO, but uses almost all of it off the pie to drive the display. You can get a display around $11, you know, $11.50 or so. If you go that route, but there is, it's not just plug and play like the DSI is, which is quite nice about the DSI you just plug it in and go, there's no setup at all. But with the DPI, not only do we need some converter boards and such, there is going to be some amount of software stuff and hardware stuff that needs to be done to get that to work. It all seems to be fairly well documented out there, but it's still not just plug and play. So anyway, that means I think that kind of helped us decide with sticking with the current design for a little bit longer and using the DSI and not moving on the DPI right away and saving that as a cost savings change in the future. I'm also making some good progress on the speakers. I should be getting like 20 samples of those a couple of weeks. If you guys recall the speakers are kind of expensive, like $11 a piece. I had kind of waited, I kind of halted a day or two on printing the first FDM design because of this discussions we were having of possibly moving to the DPI like quickly for the display, which would actually change the physical aspects of it. Not a huge amount, but enough that multiple parts would change. But now it seems like we're going to stick with that. I'll get back to printing more devices or more parts to finish that first. This has a question. Just when we say like, moving to the future, changing in the future, just to 100% clarify, we're talking about after all the kicks started back as all the kicks out of my twos have gone out, right? Yeah, future revision. Yeah, cool. Yeah, that's that's kind of the Michael question. But yeah, there's like, he talked about that in our matter most chat, there's kind of a point where the cost savings versus the NRE are going to make sense. But it looks like kind of speaking for you, Michael, like you've got to crush the numbers and thought that was that was okay to put off for a while. Yeah, I just wanted to also make it clear for people that might be listening along from home, you know, we're not going to change the display in two weeks. Yeah. Yeah, I mean the thing for us, and also we're kind of discussing there too, is that means we're going to have to enclosure versions, potentially in the future. One that has a, you know, set up for DPI and one set up for DSI and maybe some other little things between the enclosures that are a little different, but you know, across that bridge, I guess, when we get there. So, yeah, that's, there's a lot of things we're talking about too, maybe Michael probably fill us in on that on the hardware side. But for me, it means I get to go back to, you know, the version I was working on without heavy modification. So that's good for me. Yeah, so that's, that's about it for me. I don't think I'm really blocked by anyone. Did I see that we got Kevin in that? I thought I saw a comment. Yeah, he's injured now and broken around. He's already come down a few things that pointed him to the Mark II project and to the Microsoft board, so what's going on. I told him a lot of the stuff he was doing is probably assigned to you, Derek, right now, so to feel free to find stuff, stuff that he's doing, that kind of thing. Yeah, cool. Yeah, maybe I should meet with him and talk about some of the stuff. Okay, cool. Well, yeah, that's it for me. I'll see you guys have any questions on hardware front. My only questions on that front are, you know, the status of getting the bomb put together and getting price estimates for quotes for the various pieces. You know, where are we along on that front? Um, I'm doing pretty good on, on like the purchase parts, you know, the plastics are still there, you know, that's just kind of a ballpark estimate. And I'm waiting for Kevin to put in his latest estimate. I think that was also kind of just a bit on hold until we decided how we're going to do the third spin. Um, but yeah, so maybe, well, let's see, Kevin's going to be out like half a next week, isn't he? Yeah, I'll try to get a hold of him today and see if he can, before he leaves, give me his latest and greatest bomb for the PCBs. Okay. That's good. Thanks. Yeah. So for my part, on the hardware side as has been alluded to, we're going to do another spin on the board. This will correct a bug and make a really minor change to the audio data path, which will greatly simplify our software. So, and hopefully this will be the one we go into our first production run with. This will be, this will be good up to, you know, tens of thousands of units probably. You know, we start going higher than that. We'll probably want to do a revision for cost reasons and, you know, perhaps for other reasons as well, functionality. But, but yeah, the design that we've got right now is pretty solid. In fact, the other thing that I've asked is we've got I think something like six working boards right now that have a couple of bugs on them. And the, so I've asked Kevin to take a look at reworking those boards so that we can get them up and running, you know, maybe by next week at full functionality, you know, with the amplifier fix, which is really just walking a couple of capacitors and, you know, running a couple of wires from the XMOS over to the sound card to enable that data path so that we're, you know, we're not having to send dual audio stream to USB. So, you know, it's really just, it's literally two wires. So if he can implement those in there and they're stable enough to work with, then that'll be great because then we'll have boards that are exactly the equivalent of what we will be making in volume. So that would be great. And I'd be able to get one of those to replace the board I have, or should I send my board in to be fixed like that. Yeah, that's the question. We have four to go to roll over, correct? Right. Well, those are, they got the, the re-speaker based ones instead of the SJ201 versions. And we, I don't think we have enough to replace all four of their boards, but, but yeah, as soon as, as soon as we have them, I'd be, I'd love to replace them with the new version. So, but yeah, to answer your question, Chris, the idea is that everybody except maybe Ken would send, you know, all of their boards back to Kevin for him to rework late next week. He's going to be out through Wednesday. So it won't necessarily be the fastest turnaround, but it'll certainly be faster than, you know, the three weeks that I'm expecting for the next spin of the boards. So with this version, I'm confident enough that it's actually going to be a working version. We might do a higher volume run. And the goal is to have them fully assembled by the, by the assembly house. So, you know, there's no, there should be no rework required by Kevin once they come back to us. They should just come back from the factory ready to plug in. So, so, you know, we might work more than 10 this time. Josh has been experimenting with some pretty exciting TensorFlow accelerator units. And one of the neat things about the current design is that you can just plug one of those into the USB slot. And it'll be available to use. So he's been playing around with that. He's got TensorFlow late working on it. And he says it speaks things up by 30x. So that's pretty cool. That won't be incorporated as a baseline part of this design, but certainly it's a pretty low cost option for, for people to add in if they want to play around with some extended functionality. So we have, we give those of us who have the boards we have now an address to send them to. Yeah, I haven't gotten a confirmation yet from Kevin to agree to this plan. So as soon as I hear from him and get confirmation that this is something that is reasonable to do, then I'll let you know. Let's see what else you guys want to hear about. I've mostly been spending my time on fundraising. So we're not going to talk about that here. And yeah, when we mentioned the audio popping issue, perhaps over the last week or two, that has been resolved. And that was just, that was a blue wire issue because there was some additional capacitance on her. Yeah, we're getting ringing on the line because it wasn't, you know, properly impedance matched with the source. So anyway, it's basically an odd issue once we wrap that, that line on the circuit board, it'll just go away. Yeah, so on the hardware front things are looking pretty good. That's, that's all I've got for today. So how are we looking overall on this milestone? I know that when we set out earlier this week, this milestone this spring, when we set out this early this week, it looked ambitious, but it was all stuff that seemed necessary for the upcoming milestone. How are, how are we feeling, you know, with regard to progress about that now? I'm looking fine. Okay. Very few of my tasks are actually related to milestone zero. So, you know, I'm doing okay as far as getting them done. Really wait to look to that specifically. Right. Okay. Okay, well, we'll just keep it on Monday. Check in again. Unless there's any other questions, comments or concerns. Nope. All right. Wait, hold on. Hold on. I had an idea anyway. I just want to throw out. I got to mention this to guess. I think it would be nice to come up with something that we can use to test the audio on bring up. Something very simple. Because what we're seeing a lot is, is a failure of, you know, micro service or, you know, as we're bringing as we're working on the hardware, you know, this probably could be a failure on the hardware side as well. So something very simple, like we play a tone during the boot up process and we put the mics on and make sure that that tone is heard. And, you know, we'll do a little check that goes say, Hey, okay, hardware is good. Bikes are working. Sound output is working. You know, we're past that. Power on cell test. Yeah, exactly. And I don't know the, you know, level of effort on that, but, you know, that could save us a lot of headaches going back and forth and like, is this software is this hardware is this software, you know, that kind of. Especially for me. Surprisingly, my, my old unit that I have over there. I altered the startup bash RC to do exactly that because of the issues we're seeing with the speaker board. So when it first powers up, it tries to play a wave file and mp3 file. And I can always tell upon boot by listening to it, whether it's working or whether it's busted. Now, the problem is that when it's not working, it thinks it is the only difference is I don't hear anything. So I don't know how to encapsulate that in a test that would give you a yay or nay. I just know that it's an audible validation for me that it boots and okay, I don't have to reboot it. But I don't know. I mean, because it's tricky what you're asking because the audio, the audio subsystem, when it's not working, at least with the speaker, it thinks it is. Well, I think that the idea is, like you said, is you turn on the mics as well and listen for it. So it can listen to itself. Exactly. Well, not if Kevin is successful. Well, you can actually surpass the unfiltered audio back in. It actually records both. So yeah. I don't know how to do it through a programmatic perspective. It's one of the channels coming back from the XMOS is the unprocessed input. I feel like there's two things here. This would be an excellent system for the final image that goes on devices so that it's doing it when people boot their device and then we can run some troubleshooting process or try and recover from whatever's gone wrong and that sort of thing. But I feel I wonder if for our own hardware development purposes, whether we actually want, well, whether we can just create a completely separate image that has nothing to do with Microsoft so that when we have new hardware, we boot a test image that goes through and tests all of the subsystems, all of the hardware subsystems in some way. I haven't actually looked yet, but I imagine that other people who are voting with the pie device at doing similar things. But yeah, I mean, this is this is certainly something that we need to get to before commercial release, but it does sound like it's going to have some utility pretty buggy as well. So I've created a generic issue. And I've ascended to Kent because, you know, he looks, he looked like he wanted it. He made the mistake of opening his mouth again. So he's learned by now. It's fine. I'll look into it. I do agree. It's, it's a value short term and long term. You know, it's tricky. I'll figure it out. Yeah. So, I mean, possibly also I'm thinking like for project rollover or any other potential pilot or partner program we do in the future. It makes debugging hardware a little bit easier too, because we can say, okay, try running this thing real quick or doing something. You know, especially if it was like something we can run via voice or whatever, like do a diagnostic test or something. Run a level one diagnostic. But anyway, okay, just something I was thinking about. Yeah. Okay. Thanks. That's a good thing. Any other great ideas out there? Nope. Ken doesn't want any more great ideas. All right. I do have a question. I'd like to discuss the audio filing issue you have out there for this next week. Is there a good day for you? Probably Tuesday looks fairly open. Okay. That's this week. Yeah, it looks like Wednesday. Wednesday. I'll try to get something on schedule for them. All right. Okay. Great. Well, thanks everybody. This is a semi great idea. I'm using it right now and it's working pretty well. So I found an extension for Chrome meetings where if you're muted, you can press the space button and you get unmuted so you can talk. So it's like a push to talk feature. So, so far as we're not really pretty well, you all want to try it. It's out there in a Google Chrome store, but I've, I've not talked over my mute button yet today. So that's, that's progress over other meetings. Yeah. That's a good tip. Thanks. And I sent out to the links for a couple other extensions that I thought we might want to try. One of them being this closed captioning system that might make Emily's job a little bit easier or perhaps redundant. Pretty her up to doing more interesting things than writing down all the Dolby goods we say. And then the other one is, you know, this one that I thought I've seen guys using where you can, you know, raise your hand or give a thumbs up to something. You've got something to say during, you know, somebody, you know, especially myself was just blathering on. So yeah, I think we all need to have that installed to see those notifications come through. I think that's where that works. Not really sure. Does everyone see my hand being raised right now? You have to have it installed to see it on everybody. So I can see yours because I have it installed. Right. But don't have it installed. You won't see it. Okay. All right. Yeah. So I encourage everyone to install those so that we can, we can use those. But if you don't have it installed, then I, at least I do. And so if you're having trouble like getting your voice heard, you can install that and that'll be a little bit more noticeable. Right. Yeah. So, yeah. So thanks for that tip, Chris. And yeah, and check the, check these other extensions out. So yeah, that's it. So I guess everybody have a great weekend and I'll talk to you on Monday. Oh, we'll be on Monday. I guess.