**** BEGIN LOGGING AT Tue Mar 18 02:59:59 2014 Mar 18 03:03:05 mranostay: I understand that the timeline is a bit packed, but I plan to lay the groundwork of the project in the coming 2-3 weeks so that the deadlines can be met. I actually wanted to have your opinion on how much of it to be done before I filed the proposal but we could certainly discuss it now. Mar 18 03:03:48 Will you be online when I check in again after an hour? Mar 18 03:32:32 The USB code was debugged by TI and DMA support was added. Mar 18 03:34:16 'Evening jkridner Mar 18 03:37:50 To which kernel versiov Mar 18 03:38:07 *version has it been added? Mar 18 03:48:30 hi jkridner Mar 18 03:48:38 hi Mar 18 03:56:04 hi jkridner I don't know if you read my answer to your idea of cards? Mar 18 03:56:33 I haven't. Mar 18 03:57:56 I guess you have a lot of work :) I wil tell you here if you have time Mar 18 03:58:06 do you agree? or I am taking you time? Mar 18 04:07:01 I'm working on a bit of coding. Mar 18 04:08:30 * DiegoTc doesn't interrupts. He understands that Mar 18 04:19:40 * Abhishek_ understands that too. Mar 18 04:59:10 DiegoTc: replied. sleep time now Mar 18 04:59:29 Abhishek_: kernel version of what? Mar 18 04:59:48 kernel edits should go first against the mainline Mar 18 05:00:15 g'night all Mar 18 05:00:23 okay, I just wanted to ask the context in which you said it Mar 18 05:00:26 g'night Mar 18 05:47:49 mranostay: ping Mar 18 06:46:29 vvu|Log: http://source.android.com/accessories/aoa2.html Mar 18 06:46:41 Reason for multiple Pids Mar 18 06:49:09 I suppose audio mode provides an isosynchronous interface. Mar 18 06:53:57 oh audio Mar 18 06:53:57 didnt even know that existed Mar 18 06:54:20 and it has accessory + adb Mar 18 06:54:48 0x2D05 - accessory + audio + adb Mar 18 06:54:52 I'm thinking Mar 18 06:55:10 Video will be streamed inplace of audio Mar 18 06:56:25 Last night we were discussing about adb interface + adk. I think the above mode might help. Mar 18 06:56:50 vvu|Log: I think this explains why adb over wifi is killed when adk is enabled. Mar 18 08:08:26 praveendath92_: mornin Mar 18 08:19:00 vvu|Log: Morning :) Mar 18 08:19:18 so those pids are for different modes Mar 18 08:19:21 in the end Mar 18 08:19:37 Yep:) Mar 18 08:20:32 maybe this weekened i'll have time to test a bit of code myself Mar 18 08:20:39 really curious how it works Mar 18 08:20:50 first i need to root my n5 for wi-fi debugging Mar 18 08:20:56 Same with me. I will do some tests today. Mar 18 08:21:11 Did you see, there is an adb also. Mar 18 08:21:22 yep, for that one i want to see Mar 18 08:21:23 Probably that is why adb gets killed Mar 18 08:21:49 it should not in that case Mar 18 08:21:52 Okiee. I want to try something basic. Echoing some data to the port. Mar 18 08:21:53 it should keep it active Mar 18 08:22:28 I'm guessing that adb is made available through adk and disabled other forms of access. Mar 18 08:22:49 But adb is there only in mode x3 and x5. Mar 18 08:23:18 I'm looking forward for your test results. Mar 18 08:23:24 :) Mar 18 08:23:42 why do i need to have exams and quizzes while super nice stuff is happening in the open-source world Mar 18 08:24:06 Haha. Been there. Last December. Mar 18 08:24:42 Is it like your semester exams now or just quizzes ? Mar 18 08:24:52 no, just normal quizzes Mar 18 08:25:00 but each of them is like 5% of the grade Mar 18 08:25:09 especially for my math course Mar 18 08:25:37 An assignment to be turned in today at 6pm, and I'm already feeling sleepy :P Mar 18 08:25:37 Yeah. Can't ignore that 5%. Mar 18 08:25:42 indeed Mar 18 08:25:49 that 5% Mar 18 08:26:06 Abhishek_: Let me guess, you haven't started it already ? :P Mar 18 08:26:54 hmm, practically true :P Mar 18 08:27:27 ;) Mar 18 08:28:01 the unbelivable happened, now when i connect an usb device on my BBB while powered up it enumerates it Mar 18 08:28:13 on the last version of angstrom this was not happening Mar 18 08:29:37 Isn't that normal ? Mar 18 08:29:54 now it is normal Mar 18 08:31:21 I broke my LAN port while connecting BB :/ Mar 18 08:34:07 hmm, so for audio, the BBB needs to be a USB sound card Mar 18 08:34:13 is there a g_audio? Mar 18 08:34:24 hmm Mar 18 08:34:31 its not g_ Mar 18 08:34:32 its host Mar 18 08:34:39 av500: why do we need audio ? Mar 18 08:34:46 we dont Mar 18 08:34:48 I do Mar 18 08:34:50 in the car :) Mar 18 08:34:55 only plug 1 cable, not 2 Mar 18 08:35:02 praveen, you mean, the LAN port on the BB? Mar 18 08:35:30 Fortunately not. Just the wall port. Mar 18 08:35:58 ADK mode can be used. Yeah. Mar 18 08:59:42 hmm, sadly the SDK from google does not mention ADK at all Mar 18 08:59:46 ADK2 Mar 18 08:59:48 ADK2012 Mar 18 09:07:37 ADK2012 has some example implementations Mar 18 09:08:31 but mainly from the host side u just need to trigger the android into adk mode then simply send bulk data Mar 18 09:08:49 praveendath92: for adk? where? Mar 18 09:08:49 on android adk will deliver you packets in fixed size of 512 bytes as i remember Mar 18 09:08:56 says in the specs Mar 18 09:09:17 er Mar 18 09:09:18 oops Mar 18 09:09:20 I mean ADB Mar 18 09:09:26 av500: Let me check. Mar 18 09:09:30 I dont see adb in the ADK Mar 18 09:09:36 av500: still starts with A :) Mar 18 09:09:51 jaja Mar 18 09:09:55 Yeah. I couldn't find that. Mar 18 09:10:14 I mean an example in that mode. Mar 18 09:10:38 well, I think if you recompile the adb host daemon for BBB Mar 18 09:10:40 it should work Mar 18 09:10:53 so a test could be to take desktop ADB and run on BBB with an android device Mar 18 09:11:01 then you log into BBB and type adb shell Mar 18 09:11:06 and you should be in the phone Mar 18 09:11:13 it could be really simple Mar 18 09:11:30 its much harder on the microcontroller in the ADK kit as this has no linux/shell/etc... Mar 18 09:11:42 so you could try that Mar 18 09:12:08 with a BBB and an LCD you could make a portable logcat Mar 18 09:13:56 I think there is additional interface for that USB connection where we can do abd stuff. Mar 18 09:14:07 Audio opens 2 bulk interfaces Mar 18 09:17:11 Since they made this for a microcontroller, I think the interface is easier than. Mar 18 09:17:19 yes, the adb is the same as for a standard android Mar 18 09:17:47 adb on a microcontroller is not the issue, its the stuff you run on top of it Mar 18 09:17:53 like shell or push/pull Mar 18 09:18:02 but since we have BBB, we can make use if it Mar 18 09:18:09 I mean adb. They is an additional interface, not sure in which mode - interrupt or another bulk Mar 18 09:18:18 adb Mar 18 09:18:23 as on a standard phone Mar 18 09:18:25 android Mar 18 09:18:36 I dont think they changed that Mar 18 09:18:58 so as I said, running adb on BBB with a phone in dev mode Mar 18 09:19:02 as a first step Mar 18 09:19:13 then probebly that works in adk mode too Mar 18 09:19:46 Yeah. Makes sense. Mar 18 09:20:13 ADK will then be ADB + Other interfaces Mar 18 09:20:59 yep Mar 18 09:21:45 vvu said he will trying it out this week. I'm constraining myself to ADK with kernel module Mar 18 09:21:59 And also I need to complete my proposal. Mar 18 09:23:03 hmmm Mar 18 09:23:04 # adb host tool for device-as-host Mar 18 09:23:07 in adb makefile Mar 18 09:23:16 praveendath92: sure Mar 18 09:26:43 ah Mar 18 09:26:45 commit c318833b359164f25d7a93f6395c0cad787bacd7 Mar 18 09:26:47 Author: John Michelau Mar 18 09:26:48 Date: Thu Sep 23 17:08:34 2010 -0500 Mar 18 09:26:50 Build adb for target-as-host Mar 18 09:26:51 Mar 18 09:26:53 Added a new target for adb, so that the adb Mar 18 09:26:54 host executable is built for the target as Mar 18 09:26:56 well. This allows the target to connect to Mar 18 09:26:57 any Android devices which are attached to it. Mar 18 09:27:02 its adb for android Mar 18 09:33:16 Awesome find ! Mar 18 09:42:07 not what we want Mar 18 09:42:15 that builds adb for android Mar 18 09:42:20 adb host for android Mar 18 09:42:29 we need the desktop build, but for ARM Mar 18 09:42:35 maybe just https://github.com/karfield/adb Mar 18 09:42:38 but compile on BBB Mar 18 09:42:42 is enough Mar 18 09:42:46 I need to setup a BBB :) Mar 18 09:44:05 Yeah. This one is for a general system. Mar 18 09:45:57 https://github.com/karfield/adb/commit/2096ff9a64ae8fe1f0d160870d1bb990fc914d8b Mar 18 10:47:48 av500: i can do it l8er today, have my BBB running already Mar 18 10:48:02 praveendath92_: still need the adb binary for ARM ? Mar 18 11:45:26 2 Mar 18 12:19:54 vvu|Log: Not for now. Mar 18 12:31:01 done with my submission.Time to work on the proposal :) Mar 18 12:45:57 Completed ! Nice ! Mar 18 13:08:14 ds2 : ping :) Mar 18 13:35:40 hi jkridner Mar 18 13:37:17 jkridner: can you check if there are user posts on -gsoc undert moderation? Mar 18 13:39:44 mranostay: I replied to your comment Mar 18 13:44:06 mranostay : ping :) Mar 18 13:44:36 panto : did you have a look at my proposal (I sent you a mail!) Mar 18 13:53:30 panto: You are yet to review my proposal too. Mar 18 13:53:57 Abhishek_, link Mar 18 13:54:04 karki, looking at it Mar 18 13:54:13 it is on the melange portal Mar 18 13:55:21 ok, now I see it Mar 18 14:56:49 av500: hello good morning , I have seen your comments Mar 18 15:46:32 ds2: did you look at the changes? Mar 18 15:54:19 av500: there are still some problems with usb on the Beagle, when i disconnect my Nexus 5 my board just freezes Mar 18 16:02:02 then dont disconnect :) Mar 18 16:02:18 any clue if the syslog gets saved somehow ? Mar 18 16:12:03 hmm Mar 18 16:12:06 wasnt there something Mar 18 16:12:07 in android Mar 18 16:18:35 com.tortel.syslog Mar 18 16:18:55 App package that is. Mar 18 16:19:18 i don't care about android side, just wanna see what is happening on the BBB Mar 18 16:23:09 serial debug? Mar 18 16:26:28 hi jkridner are you available? Mar 18 16:29:19 hi alexanderhiam Mar 18 16:29:22 do you have time? Mar 18 16:31:28 sort of, what's up? Mar 18 16:34:22 hi I have you 3 questions Mar 18 16:36:48 #1. You wrote this comment on my proposal. about the fact that you'd be writing a documentation framework and not just designing a wiki template or something. Mar 18 16:36:55 Probably my spanish -> english translation wasn't so good at that moment, could you be a little specific in that please? Mar 18 16:37:35 I just meant you should emphasize the coding aspects more in the short description Mar 18 16:38:03 ahh more technical Mar 18 16:38:45 mranostay: When will you be available? Mar 18 16:38:48 just because when I first read the short description it sounded more like you would just be writing documentation Mar 18 16:39:59 oki doki Mar 18 16:41:30 Abhishek_: what is up? Mar 18 16:41:46 #2. alexanderhiam what do you think the impact the impact of this project will be to the beagle board community? Mar 18 16:42:12 Did you see my reply to your comment? Mar 18 16:47:20 mranostay : did you have a look at my Higher language support for PRU proposal? Mar 18 16:48:26 karki: briefly have a few questions Mar 18 16:48:48 are you basically making a Higher Level PRU assembly loader? Mar 18 16:49:04 or doing a language compiler to PRU code Mar 18 16:49:15 aka python to PRU assembly Mar 18 16:50:06 DiegoTc: I think it would be very valuable for beginners to have a single place to go for documentation as it's all pretty scattered at the moment, and I think the platform as you describe it would encourage more people to provide good documentation for their own projects. Mar 18 16:50:29 mranostay: FYI, I'm advocating for extending the types of commands you support for lighting to include general application interpretation ala BotSpeak (https://sites.google.com/site/botspeak/). Mar 18 16:50:44 alexanderhiam: I can you this as one of my answer for proposal? Mar 18 16:50:53 mranostay: you probably have already noticed that, but I want to answer any doubts you might have. Mar 18 16:51:12 * jkridner doesn't necessarily know what karki has planned Mar 18 16:51:37 DiegoTc: certainly Mar 18 16:51:53 jkridner : Well my proposal is in sync with pruduino readme; or so I think (please have a look) Mar 18 16:52:15 *yawn* Mar 18 16:52:34 thaks Mar 18 16:54:11 mranostay: I would like to know your opinion on the impact of this (my) project on the beagle board Community. Mar 18 16:54:29 mranostay : I'm not focusing on python or node.js for the GSoC. The idea is to implement a botspeak interpreter for the PRU. For GSoC , the client side will be limited to C. Mar 18 16:54:31 Which I can include as an answer in the proposal Mar 18 16:56:58 hmm never heard of botspeak before Mar 18 16:57:47 The main idea behind the project is to have a generic firmware which can interact with any client side language running on linux. (As far as the the client language has the library to transfer botspeak script fragments to the PRU) Mar 18 16:58:29 Botspeak is a very interesting project :) Mar 18 17:00:16 jkridner: I know I botter a lot, but have you check my proposal, planning to make some changes and what to include yours. Mar 18 17:01:10 karki: there's a big difference in terms of memory requirements between the ascii based botspeak scripts and your other idea to have few-byte macros. Have you thought about where the running botspeak scripts would live? Mar 18 17:03:28 (I'm assuming botspeak is interpreted and not compiled) Mar 18 17:03:36 alexanderhiam : yes I know :( I did come with a one is to one mapping idea for a botspeak instruction to single byte representation! but then It really isin't botspeak anymore. I'm thinking of a way around the problem! Mar 18 17:04:57 alexanderhiam : should have a solution by tomorrow ( a way to keep the botspeak instructions readable at the same time not to waste memory! ) :D Mar 18 17:05:37 karki: yes, please include size estimates as memory is limited on the PRUs Mar 18 17:05:57 DMA? Are there any guaranteed empty chunks of memory? Mar 18 17:07:55 ds2 : Hello :) I did reply via email as I was not sure I could make it here today! but did my email answer the question (wrote it in kind of a hurry!) or have I got something wrong and should come up with another solution! Mar 18 17:09:54 I'll have to go complete my machine learning assignment! (deadlines and college! uugh..) BBS Mar 18 17:10:22 alexanderhiam : I'll look into it, not too sure. Mar 18 17:16:59 ds2: I hope you did review the edits I made yesterday Mar 18 17:25:41 DiegoTc: I think I sent some feedback. I keep looking back at the submissions and updating my feedback. Be sure to look at the feedback I'm giving others regarding the "documentation" project as it gets called. Mar 18 17:26:51 jkridner: I have the idea of the new design Mar 18 17:27:01 instead of using tabs Mar 18 17:27:29 we can have a panel with each gist Mar 18 17:27:48 there's going to be a next bottom that will activate each panel Mar 18 17:28:09 Abhishek_: I looked through them and I don't have any comments at this time other then possibly a little concern with the schedule as mranostay pointed out Mar 18 17:28:45 Abhishek_: prehaps fleshing out the schedule with more detail Mar 18 17:31:56 Okay, and about the two answers I have to seek from you all? Mar 18 17:35:47 DiegoTc: I don't understand the "next button", but I guess you just mean that is the wizard part of going through the steps. I would add that at times it could make sense to view multiple panels (cards) within the same deck at one time. (A gist would contain all the cards for the tutorial/deck. As I described them, each one as stepX) Mar 18 17:36:51 DiegoTc: I don't have it fully thought out, but I'm trying to guide you toward having fixed-sized elements for which we can do interesting layouts to browse them all. Mar 18 17:38:52 so, a single gist would contain all of the elements for a tutorial, the JavaScript source code, the Fritzing diagram, the build instructions, the background information, meta-information including where to buy the component and the tutorial author, etc. Mar 18 17:39:25 DiegoTc: I'm wanting you to update your mock-ups to reflect this and include references to your mock-ups in your proposal. Mar 18 17:39:44 jkridner: ping Mar 18 17:39:49 hi av500 Mar 18 17:39:55 some ppl claimed thier posts were not showing on the ML Mar 18 17:40:01 can you check moderation? Mar 18 17:40:39 k. Mar 18 17:40:55 I run through it only every couple of days. :( Mar 18 17:40:55 thx Mar 18 17:41:07 well ,we should before app period closes Mar 18 17:41:40 I just sent 3 through. Mar 18 17:42:11 jkridner: I was thinking like this: https://docs.google.com/drawings/d/1ZN7AqNXKcurmBKLQBFXdnr4oHlNbw9rqdFjX1D91Wqo/edit?usp=sharing Mar 18 17:42:22 no time right now to modify mockup until lunch Mar 18 17:42:53 av500: I'm not seeing people requesting to join the list. :( Mar 18 17:43:59 I don't like the idea of having all them in one gist. Reason when you see many "fill options" you quite to leave some in blank. But with the next is a lie you know you have to fill it. Mar 18 17:45:43 jkridner: i don't know if you get my point of view? Mar 18 17:47:26 However we display everything, it should handle if we leave them blank. We cannot have a single tutorial be non-atomic as that would make forking unmanagable and prevent a single resource from defining the tutorial. Mar 18 17:47:37 so, no, I don't get your point of view. Mar 18 17:47:48 Did you see the way jsfiddle.net handles a gist? Mar 18 17:48:33 a gist can have many files and files can be added after initial creation. Mar 18 17:50:27 let me see, trying to visualize it Mar 18 17:50:33 yes i saw the Jsfield Mar 18 17:50:54 jkridner: okj Mar 18 17:50:56 ok Mar 18 17:51:24 DiegoTc: here's my example gist: https://gist.github.com/jadonk/9602198 Mar 18 17:51:48 And here's how it looks on JSFiddle.net: http://jsfiddle.net/gh/gist/jquery/1.9.1/9602198/ Mar 18 17:51:53 same data, different view. Mar 18 17:52:14 The gist can be easily forked, edited and shared. Mar 18 17:54:03 jkridner: you have 3 gist there or no? Mar 18 17:54:15 just 1. Mar 18 17:54:21 the gist can contain multiple files. Mar 18 17:54:32 rrr Mar 18 17:54:36 I got you know Mar 18 17:54:37 I was suggesting we could have as many files as we need to build up the tutorial. Mar 18 17:54:46 yes Mar 18 17:54:51 png files, fzz files, js files, html files, etc. Mar 18 17:55:03 and use intelligent naming conventions so that we know how to embed them. Mar 18 17:55:17 and be thoughtful about how we want to wrap the scripts, if at all. Mar 18 17:55:44 I thought each block was one gist Mar 18 17:56:51 so, there will be a consistent framework around them such that maybe they could even be shown as something like how https://gist.github.com/jadonk/9512092 or https://gist.github.com/jadonk/9602198 are visualized on Codepen or JSFiddle respectively, but we'll have our "card view" that acts a browsable element and a wizard for walking through the steps. Mar 18 17:56:56 now i have it clear Mar 18 17:57:25 All of https://gist.github.com/jadonk/9602198 is one gist. If you have your github account, you can simply click the 'fork' button and begin editing your own copy (without changing mine). Mar 18 17:58:29 we should also handle full-on github repos, but the gist is the lightweight mechanism that'll get us going quickly. Mar 18 17:59:22 when you mean card view you mean something like this: http://jsbin.com/OrIWAr/5 Mar 18 17:59:27 instead of speech bubble Mar 18 17:59:31 cards Mar 18 17:59:33 providing a smartly organized set of information within these gists (ie. API definition), we'll enable people to make new tutorials quickly. Mar 18 17:59:45 meh. Mar 18 18:00:01 the visual element should have the outline of a BeagleBone. Mar 18 18:00:36 large enough to contain a circuit diagram as a Fritzing diagram that is reasonably readable. Mar 18 18:01:08 perhaps there are a couple of zoom levels (not arbitrary, just a couple)... Mar 18 18:01:28 we need to have consistent sizes such that people who make these are making something that always looks good. Mar 18 18:01:48 it would make sense if they also looked good on mobile phones, but not strictly required. Mar 18 18:02:01 yeap, let say the jsbin I send you contains bubble speech make with css. I have to design something similar but with the design of a beaglebone Mar 18 18:03:54 right. Think BeagleBone-shaped trading cards. Mar 18 18:03:58 * jkridner looks at http://www.readwritethink.org/files/resources/interactives/trading_cards_2/ Mar 18 18:07:16 Steam trading cards: http://steamtradingcards.wikia.com/wiki/Steam_Trading_Cards_Wiki?file=Steam_Trading_Cards_-_What_are_they%253F Mar 18 18:09:15 jkridner: wnat something like this: https://docs.google.com/file/d/0B73UFFcF-_ZSQVRsR21nQkk4V1ZXLXlLSzVHTHFuRkFKbnpV/edit Mar 18 18:09:20 as card? Mar 18 18:09:40 instead of the background i have the beagleboard logo Mar 18 18:09:41 ? Mar 18 18:10:01 mikehhawley: test Mar 18 18:10:14 I wouldn't use that *particular* layout and I suspect that landscape would work better than portrait layout. Mar 18 18:10:24 I'm just giving the concept. Mar 18 18:10:37 no, that was an example trying the tool Mar 18 18:10:54 my real question goes here Mar 18 18:11:05 we are going to pass that card as CSS Mar 18 18:11:20 or are we going to use the image background? Mar 18 18:11:38 right.... I think going through that tool and creating a card gives you some idea (though I again want to work on the rendering side first, rather than the authoring side and simply author with gist.github.com for now). Mar 18 18:12:07 DiegoTc: I'm not sure what you are asking regarding CSS vs. image background. Mar 18 18:13:05 * jkridner wants to solve the layout issues, but is less concerned about pure-aesthetics in the initial implementation. Mar 18 18:13:56 The gist is going to have multiple files. The files are going to have a card view. This card view is going to be make via CSS or a background image Mar 18 18:14:25 A gist should provide us with a deck of cards.... ie. we might display more than one card at once for the gist or we might "flip" through them in wizard like fashion as we go step-by-step. Mar 18 18:14:50 I don't know what you mean by "via CSS or background image". Mar 18 18:15:18 The cards won't be images.... they will be HTML content. Mar 18 18:15:36 something like the circuit drawing will be an image within the HTML. Mar 18 18:15:55 CSS will lay them out and I can imagine they'd include background images to help style them. Mar 18 18:16:21 the CSS should be fairly isolated such that we can easily drop them into virtually any web page. Mar 18 18:20:03 jkridner: one minute Mar 18 18:21:41 jkridner: I think we (I'm) having a bad misunderstanding of terms Mar 18 18:22:11 this are the cards: http://screencast.com/t/GWprnKbw Mar 18 18:26:11 ah, no, those aren't the cards. those are source elements to help create the cards. Mar 18 18:27:21 http://beagleboard.org/Support/BoneScript/Ultrasonic_Sensor/ is an example of something that can be turned into a deck of cards. Mar 18 18:27:55 There should be some kind of cover-card that gives a quick description for browsing through the various decks. Mar 18 18:28:14 There should be some kind of card or series of cards that give background on the tutorial task. Mar 18 18:28:52 There should be a card or series of cards that tells you how to build the hardware, including an assembly diagram. Mar 18 18:29:20 There should be a card that gives extra information like links to other tutorial decks. Mar 18 18:29:35 There should be a card showing the tutorial in action. Mar 18 18:29:56 There should be a card with source code for the tutorial. Mar 18 18:30:10 There should be a card for visualizations associated with the tutorial. Mar 18 18:30:48 When walking through the tutorial itself, multiple cards might be visible and some cards might only be visible once you complete the steps from previous cards. Mar 18 18:31:17 jkridner: just a thought: would a table of contents fit into this scheme? or perhaps the 'whole deck' view would make it easy enough to jump straight to a particular step Mar 18 18:31:45 http://jsfiddle.net/gh/gist/jquery/1.9.1/9602198/ is a nice view, but it misses several elements including the ability to have "steps" and hardware build instructions. Mar 18 18:32:45 alexanderhiam: I would look for as much graphical navigation as can be done, eliminating much of the need for the navigation menu at the left of http://beagleboard.org/Support/BoneScript/Ultrasonic_Sensor/ Mar 18 18:34:22 I would look for a first-class experience regarding live-running BoneScript demos like the ones I've been showing on JSFiddle with some simple sensors wired-up and including things that require social-logins like sending tweets of the room temperature, but I'd also expect it to be possible to make tutorial decks on configuring wifi or doing OpenCV-based face detection. Mar 18 18:44:48 jkridner: have you checked out the wolfram language ? would be neat to have that onboard too. although, not something that can be done via gsoc i guess Mar 18 18:50:15 anujdeshpande: isn't that a proprietary language? Mar 18 18:50:45 alexanderhiam: it is. guilty pleasure :/ Mar 18 19:34:53 jkridner: Do you have any thoughts to add in the web based interface for the proposed LA tool? If you might have seen the updated proposal, you might have noticed that I plan to include it as a part of the Bone101 getting started as an out-of-the-box application. Mar 18 19:35:42 Abhishek_: including it as part of the out-of-box tutorials will be great. Mar 18 19:36:27 Abhishek_: I'll need to read a bit more to see how you plan to do the visualizations. Flot is handy, but you'll need quite an infrastructure to properly label and zoom. Mar 18 19:36:46 Yes, that is what I was currently thinking about Mar 18 19:37:36 Idling on the libsigrok thread, people had brought about links to this: https://code.google.com/p/wavedrom/ Mar 18 19:39:11 I had also noticed this one and pointed out to ds2 about a week ago: http://www.edaplayground.com/w/ [login is required, but I can send you a screenshot] Mar 18 19:39:48 ah, excellent find. Mar 18 19:40:24 that last one seems like it would be closed source. Mar 18 19:40:48 yeah, it is actually an online viewer for vcd files (standard ones emitted by LAs) Mar 18 19:45:35 * jkridner is really enjoying http://jsfiddle.net/H7nBn/1/ ! Mar 18 19:46:03 quick question- is posting ideas on gsoc page of elinux.org limited to only the students proposing? I did post an idea thinking it may help students want to pick it up Mar 18 19:46:49 interesting link jkridner, thanks for the heads-up :) Mar 18 19:47:15 joel_: your editing of the ideas page was the right thing to do... Mar 18 19:47:32 jkridner, ah ok, thanks for clarifying :-) Mar 18 19:47:33 that page is meant to be there to help students write proposals that would most likely be accepted. Mar 18 19:47:52 gotcha Mar 18 19:48:05 cool, thanks. Mar 18 19:48:05 joel_: please feel free to update other ideas anywhere you feel you can bring clarity. Mar 18 19:48:15 Hello. I submitted my proposal on melange but I can't see an edit anywhere. Where's the edit link? Mar 18 19:48:15 definitely Mar 18 19:49:46 rseethamraju: Once you open your proposal for viewing, you can see the link to edit it. Mar 18 19:50:24 * jkridner is looking for rseethamraju's proposal, but can't quite map the irc nick to the Student name. Mar 18 19:50:43 its rseetham Mar 18 19:50:56 the proposal don't even show up properly on the My proposals table only the organisation column in filled as beagle bone. Mar 18 19:52:03 I refreshed at least now the organisation shows up as Beagle Board Mar 18 19:53:14 Abhishek_: how do you open your proposal? Mar 18 19:53:33 rseethamraju: it did get submitted, I see it here Mar 18 19:54:00 rseethamraju: I see it as well. Still having problems editing it? Mar 18 19:54:20 but how do I see it if I want to edit it the title itself comes like an option on a drop box Mar 18 19:54:28 yes can't edit it Mar 18 19:55:00 i can only see the mentor view, other students here? Mar 18 19:55:19 don't need to now anyway. I'll do it tomorrow after looking at all the comments. Assignments to finish. Mar 18 19:56:30 It'll show u tomorrow right? It did this after I submitted the verification the proposal link showed up after like 2 hrs Mar 18 19:56:57 anyway 'Night Mar 18 19:57:12 Oh time difference!! =) Mar 18 20:05:10 jkridner: Do you have any suggestions/advice on how to get it scroll? WaveDrom seems to work well for static waveforms but will have to see on how to get it scroll, zoom and pan. Mar 18 20:06:12 ^^ I edit a few minutes ago. Mar 18 20:07:00 vvu|Log: ping Mar 18 20:07:06 praveendath92: pong Mar 18 20:07:37 How do we change to ADK mode ? Mar 18 20:07:48 you do control trnasfer to the device Mar 18 20:07:53 i showed the python code to you Mar 18 20:07:58 They have a different set of Pid and Vid Mar 18 20:08:12 every device has a different vid/pid Mar 18 20:08:21 and when in ADK mode they have the same things Mar 18 20:09:09 I read a complete book on USB drivers today. Mar 18 20:09:18 link ? Mar 18 20:09:22 cuz i'm interested too Mar 18 20:09:33 http://lwn.net/images/pdf/LDD3/ch13.pdf Mar 18 20:09:37 +1 Mar 18 20:10:04 aaa remember this thingie, got an eye over it last year Mar 18 20:10:11 when i was searching smth about URBs Mar 18 20:10:17 From what I read, the vid and pid are charecteristics of a device Mar 18 20:10:48 true Mar 18 20:10:57 Yeah. I skipped this a lot of times for a quick solution but today I decided to go through the whole of it Mar 18 20:11:32 So, does the device connection be registered as one device ? Mar 18 20:12:06 If so, it should have a different vid, pid right ? Since adk mode is not the default mode Mar 18 20:12:14 u have one device connection on which you have multiple interfaces as i remember Mar 18 20:12:19 and each interface has a number of endpoints Mar 18 20:12:27 Yeah. True. Mar 18 20:12:32 and the endpoints have a direction host->device device<-host Mar 18 20:12:43 Yes. But pid and vid are for devices Mar 18 20:12:52 ep0 is used always to set up configs and stuff Mar 18 20:12:55 Yep. Mar 18 20:13:05 yes, vid/pid is the identifier you are going to use Mar 18 20:14:05 But the device when first connected has a differnet vid and pid. Not adk_vid Mar 18 20:14:09 and adk_pid Mar 18 20:14:16 yep Mar 18 20:14:34 So, in the link you sent, line 31 Mar 18 20:14:46 as i told above, each Android device has a VID that is specific to the vendor *TI, Asus, LG* and a custom PID. Mar 18 20:15:02 but when it goes to adk mode it switches these values to some predefined ones from Google Mar 18 20:15:07 Yep. Got that. Mar 18 20:15:42 That is clear. My doubt is, how does switching command looks like ? Mar 18 20:16:01 w8 a sec Mar 18 20:16:08 I mean, default - some mode with default *samsung (say) pid Mar 18 20:16:14 Okiee. Mar 18 20:16:17 vid* Mar 18 20:18:00 praveendath92: http://android.serverbox.ch/wp-content/simplectrl.c Mar 18 20:18:04 I wrote a basic USB driver and started to think how do I ask the device to change it's mode to adk. Mar 18 20:18:12 static int setupAccessory Mar 18 20:18:14 that function Mar 18 20:18:27 he does there control transfer Mar 18 20:18:31 using libusb api Mar 18 20:19:44 this is super basic code, just put device to adk mode no communication afterwards Mar 18 20:20:42 Just to clarify, this part wasn't done explicitly in the py code right ? Mar 18 20:20:57 *** Looking at the C code *** Mar 18 20:21:04 no, my bad for sharing that code Mar 18 20:21:39 Actually, that will help once I setup the mode. Mar 18 20:22:41 _av500_: ping Mar 18 20:22:48 praveendath92: set up a git repo to keep track of your things Mar 18 20:22:53 ds2: ping Mar 18 20:23:21 Yeah. Planning to create a new repo for gsoc-testing codes. Mar 18 20:23:29 Should do that right away. Mar 18 20:23:54 BTW, was that the code that you tested today ? Mar 18 20:24:45 nop, i have adb running on BBB now :) Mar 18 20:25:37 i wanna put my nexus in adk mode to see if i still will have adb there Mar 18 20:26:19 Nice ! Mar 18 20:26:45 i need to set up the BBB nicely, still running on default user Mar 18 20:27:12 Can you send me details on how you did that ? Mar 18 20:27:19 adb on BBB ? Mar 18 20:27:49 can we do that 2morrow cuz i am not on my main desktop now Mar 18 20:27:55 jkridner: ping Mar 18 20:28:02 pong Mar 18 20:28:08 hi, have been wanting to implement node-webkit getting-started app? Mar 18 20:29:04 jkridner: i started to write my project application. Can you give feedback? https://docs.google.com/document/d/1bD8qjc5KhJXfew0q7hlH5-DKQ4C5NlZZdllpk7iCdFw/edit Mar 18 20:29:47 jkridner: on the new debian image uSD is accessible if booted from eMMC ? Mar 18 20:32:34 Sure. I'm already too sleepy. Mar 18 20:33:46 av500: I updated my proposal. Need to add more details on the Displaylink part Mar 18 20:34:41 The usb connection and drivers part is now all clear :) Thanks to vvu|Log and http://lwn.net/images/pdf/LDD3/ch13.pdf Mar 18 20:35:06 anytime praveendath92 Mar 18 20:35:36 :) I will get going. Good night guys :) Mar 18 20:36:04 vvu|Log for praveendath92's project is any libusb magic needed on the android side? or is the plan to use the USB classes on Android etc? Mar 18 20:36:23 nop, no libusb Mar 18 20:36:30 just USBManager class from Android Mar 18 20:36:38 vvu|Log, gotcha that I was thinking, cool. Mar 18 20:36:43 it should work on all Android devices, no root no nothing special Mar 18 20:36:55 sweet Mar 18 20:36:59 yep :) Mar 18 20:37:00 vvu|Log: yes, you can still use the uSD when running from the eMMC (or vice-versa) Mar 18 20:37:17 * vvu|Log needs to learn how to manage partitions right Mar 18 20:37:24 WDNCh: yes, I'm continuing to read and review proposals in any time I have. Mar 18 20:37:25 vvu|Log, perf-wise not too bad? Mar 18 20:37:52 in comparison to libusb Mar 18 20:38:18 libusb is still in discussion now about android, on their ML is a thread on how to make it standard for android Mar 18 20:38:44 jkridner: OK. I give permission, so you can add comment. Thank you. Mar 18 20:38:48 cuz now there are so many libusb implementations for android, forks and with custom changes and nobody know nothing about them Mar 18 20:39:04 ah ok, thanks for the info Mar 18 20:39:04 and to use libusb 1st u need to have root access to your android so it breaks everything Mar 18 20:39:28 yeah the rooting part definitely sucks Mar 18 20:39:32 take the fd from usb manager class then pass it to libusb or smth like that Mar 18 20:39:46 and it involves some JNI code which for me at least is super magic Mar 18 20:40:14 so why would someone want to pass control to libusb if they're using usb manager anyway Mar 18 20:40:20 I'm hoping the answer is perf Mar 18 20:40:45 i have no clue here, did not stay much to think about it Mar 18 20:41:07 when i started to understand a couple of things about libusb on android University started and did not have that much time to spend on it Mar 18 20:42:45 vvu|Log, ah ok... Mar 18 20:43:11 joel_: also do you know any maintainers for libusb around #beagle ? Mar 18 20:43:26 i am trying to find some info about libusb on osx and how to take over devices there Mar 18 20:43:49 vvu|Log, no the only person I know working closely on USB stuff is felipebalbi Mar 18 20:44:04 and ds2 prolly Mar 18 20:45:18 vvu|Log, shouldn't libusb abstract the heck out of it? ;) and allow you to pass a vid/pid or some such Mar 18 20:45:36 on Linux it has libusb_deattach_kernel_driver Mar 18 20:45:46 which takes care of this for me, passing me control in userspace of the device Mar 18 20:45:53 but on osx that function is just a stub Mar 18 20:46:03 with a nice comment in it "Patches are welcomed!" as i remember Mar 18 20:46:30 bummer Mar 18 20:46:50 and you need to write a kext which has no code in it and specifies what to be done when that device is connected to osx. mainly i got lost in Apple's docs...their obj-c names are just way too long Mar 18 20:48:47 if the multi-boot project will get accepted i will look more into it, everybody hates Mac :) Mar 18 20:49:07 Hello, are there any mentors? Mar 18 20:50:42 I updated my proposal for Cross platform USB boot according to the feedback I received yesterday Mar 18 20:50:46 vvu|Log, you probably saw this but found a ticket on libusb about this marking it invalid (last comment): http://www.libusb.org/ticket/33 Mar 18 20:51:22 and I am stuck on the part where I have to descibe the impact on the community having feedback from the mentors Mar 18 20:51:43 and then some folks talk about using hidapi: http://www.signal11.us/oss/hidapi/ Mar 18 20:51:51 Dont have much idea about it, sorry. gtg too Mar 18 20:52:04 no worries joel_ will figure it out somehow :) Mar 18 20:52:13 :) Mar 18 20:58:02 jkrinder Mar 18 21:01:06 alexanderhiam Mar 18 21:01:19 hey Mar 18 21:01:22 hey Mar 18 21:01:27 just left another comment Mar 18 21:01:40 Okay, will look at it now, ty Mar 18 21:08:31 alexanderhiam I provided the answer Mar 18 21:08:32 alexanderhiam: can u look at my propasal? (beta version) Mar 18 21:10:02 WDNCh: is it on melange? Mar 18 21:10:11 alexanderhiam: i didn't upload yet, but you can add comment https://docs.google.com/document/d/1bD8qjc5KhJXfew0q7hlH5-DKQ4C5NlZZdllpk7iCdFw/edit Mar 18 21:11:28 it would be best to submit it there, then you can get feedback without having to catch mentors on irc. You can keep editing it there until the 21st Mar 18 21:18:47 ok i'll in 1 hour thank you Mar 18 21:49:26 alexanderhiam: I uploaded my proposal on melange. Could you please give feedback when you are available? Mar 18 21:50:15 ok, I'll take a look soon Mar 18 22:02:28 Hello, I need to discuss my idea for GSoC for BeagleBoard? Mar 18 22:02:37 Can I talk to anybody for consultation? Mar 18 22:02:47 tell us what wonderful things you'd like to propose. Mar 18 22:04:09 I have developed a Web-based home control and automation system using Beagle Bone Black. Mar 18 22:04:41 what is the highly-reusable open source component? Mar 18 22:05:16 Regarding the software, the NodeJS code would be very useful. Mar 18 22:05:28 Regarding hardware, not much I think. Mar 18 22:05:46 simple integration tasks aren't likely to be selected.... whereas reusable libraries and tools that help teach/enable integration are. Mar 18 22:07:06 jkrinder: after yesterdays feedback I have modified my proposal on cross platform USB boot, if you have a spare minute, could you please look at it Mar 18 22:07:47 I would, but I don't think I am qualified enough to opine. Mar 18 22:08:42 as well, what exactly should I write in the part where I have to describe the impact on the community having the feedback form the mentors Mar 18 22:08:58 Can I talk to an official representative of BeagleBoard please? Mar 18 22:09:09 electro_: ??? Mar 18 22:09:39 jkridner, yes? Mar 18 22:09:51 it doesn't get much more official than me. Mar 18 22:10:11 Okay, your introduction? Mar 18 22:10:11 * jkridner is the BeagleBoard.org GSoC Administrator and co-founder of beagleboard.org Mar 18 22:10:28 Okay sorry, honest mistake. Mar 18 22:11:03 Could you look at my proposal in a document form? What is your email address? Mar 18 22:12:00 best e-mail address to send to is beagleboard-gsoc@googlegroups.com. I've been trying to go through the posts there every day. Mar 18 22:12:23 Okay, I'll send it right away. Mar 18 22:12:25 for sufficient bandwidth, we'll need to pull in some other mentors anyway. Mar 18 22:12:51 Should I send videos and photographs of the hardware as well? Mar 18 22:13:35 hmmmm Mar 18 22:13:42 hardware? videos? Mar 18 22:18:27 jkridner is also known as mom here :) Mar 18 22:18:30 * mranostay ducks Mar 18 22:18:54 * jkridner primps. Mar 18 22:19:27 electro_: I do hope it isn't hardware specific, but additional information that shows you know what you are doing is always welcome. Mar 18 22:21:20 electro_: you said "developed", is this a finished project? Mar 18 22:21:35 It does involve specific circuitry that is plugged into BeagleBone's IO ports, I could send you the circuit design (made on Proteus) or I could send you pictures of the actual circuit. Mar 18 22:22:05 can you give a quick 2-3 line description right now? Mar 18 22:22:31 alexanderhiam, yes. It is a finished project, but I'm willing to develop it further. Mar 18 22:23:27 @ds2, my project allows you to control your home appliances and power supply from anywhere around the world. Mar 18 22:23:40 would you have 11 weeks worth of work? Mar 18 22:23:51 Plus you can monitor you power usage and temperature of your home. Mar 18 22:24:03 Yes, I think I'll make it into something bigger. Mar 18 22:25:06 how is that different from say... a CM11A connected to the serial port of a Beagle ? Mar 18 22:26:50 I don't know what a CM11A is. But from what Google tells me, I think it is a module of my project. Mar 18 22:27:11 I have used an opto coupler to serve the same purpose. Mar 18 22:28:32 You can switch the appliances On/Off and increase their current,. Plus you can use a temperature sensor, a voltmeter and an ammeter to view your power usage. Mar 18 22:29:47 this is sounding more like a personal hardware project atm, send your proposal to the mailing list and we'll take a look Mar 18 22:31:09 it is a uart to X10 interface Mar 18 22:35:04 panto: I'm editing the timeline of my app, request you to take a look again when I am done. Perhaps it will be able to answer your comment Mar 18 22:38:39 alexanderhiam: I have emailed it, kindly take a look. Mar 18 22:58:44 jkridner: I wanted advice on how to pass data from a native C application (in this case using libsigrok) to a node.js server that would serve data to the browser client Mar 18 22:59:07 if you had a C library, you could use node-ffi. Mar 18 22:59:38 you can also use sockets or tmpfs. Mar 18 23:06:17 hmm, the idea is to create a bidir interface between the client and a C application that would allow setting triggers etc, and once the data is captured by the app, then forward it to the web client.. Will it be more efficient if the C application integrated the server functionality, or would it be better to use tmpfs between C app <-> node.js server <-> web Mar 18 23:06:17 client? Mar 18 23:06:55 * jkridner isn't familiar with the functionality of libsigrok. Mar 18 23:07:18 is it for the backend data collector or frontend display? Mar 18 23:07:43 * jkridner would rather see a node.js app for the server. Mar 18 23:07:49 * jkridner is likely in the minority Mar 18 23:08:37 it is for the backend data collector, after all data has been collected and processed Mar 18 23:09:05 mranostay: RFC Mar 18 23:10:01 * jkridner sees http://sigrok.org/wiki/TODO has WaveDrom as a to-do Mar 18 23:11:25 yes, I picked up that discussion from #sigrok Mar 18 23:11:31 adding a C API to libsigrok that output JSON for WaveDrom would be a great thing to wrap with node-ffi. Mar 18 23:12:45 I guess the hardware accessing functions would have to be wrapped too, for use Mar 18 23:12:48 Ideally, libsigrok integration would provide a node.js EventEmitter that could be used to push updated JSON content using socket.io to a web page. Mar 18 23:16:17 This integration will be accomplished by writing a node.js library? Mar 18 23:27:12 * vvu|Log hates a bit all these vid/pid identifiers Mar 18 23:28:47 Abhishek_: that's how I see it. Mar 18 23:32:40 Hmm, with just 1 day remaining, this is one aspect I am looking at right now. I would give implementation of this 3 weeks' of time, and 4 weeks for the web client. Mar 18 23:50:25 jkridner: Since the project will deliver both a front-end and a back-end, would it be better if development alternates between front-end and back-end every week? Or it would be better, say, spending first 3 weeks for the front-end and then 4 weeks for the back end and so on? Mar 18 23:51:00 we have an existing front-end we can use with something like sigrok-qt, right? Mar 18 23:51:19 I'd say finish the backend first, then do the front-end, but that is just me and I'm unlikely to be the lead mentor on this one. Mar 18 23:53:02 There's PulseView in sigrok, but the idea is to deliver a web client based front-end as the LA PRU inputs would conflict with the HDMI framer pins, hence connection of an external display might not be possible. Mar 18 23:53:32 We can use VNC too, but the web client front-end seems to be a better alternative Mar 19 00:02:29 ds2: ping Mar 19 00:02:57 jkrinder: what exactly should be under the part of where I have to write in the proposal the impact on the community having the feedback from a mentor, i.e. can you provide me with that feedback? Mar 19 00:42:16 jkridner: Hi! i talked to alexanderhiam yesterday regarding the license issues with starterware. he maintained that it might be possible to use it for the project. Mar 19 00:42:27 great. Mar 19 00:44:03 i mean, do you have any updates on this? he used this as ref: http://e2e.ti.com/support/embedded/starterware/f/790/t/252760.aspx Mar 19 00:44:18 he said as they were not replying, maybe there isint any issue Mar 19 00:46:24 might take a day or two to get a reply. Mar 19 00:47:15 the date for the question is March 18, 2013 Mar 19 00:50:03 also i am yet to get your feedback on the possibility of having the project with its focus on creating support in energia Mar 19 00:50:27 if this is acceptable, it would help me in writing the proposal :) Mar 19 00:56:07 http://opensource.com/life/14/3/google-summer-code-2014 Mar 19 01:02:29 jkridner: can you please provide me with feedback on my proposal after the modifications I've done after yesterdays talk? Mar 19 01:02:43 URL? Mar 19 01:03:02 http://www.google-melange.com/gsoc/proposal/public/google/gsoc2014/abarbarosie/5629499534213120 Mar 19 01:04:09 praveendath92_: ping Mar 19 01:04:31 praveendath92_: just tested, you will have adb working while adk is running Mar 19 01:04:45 just done now put my phone into adk mode and did adb shell Mar 19 01:07:44 abarbarosie: do you have the same proposal shared on the mailing list? I can provide general comments in response to the proposal using the tool, but I can go line by line if you send it to beagleboard-gsoc... as long as you don't mind sharing it with the other students. Mar 19 01:09:19 No, I don't have it on the mailing list, I will post it and notify you. Mar 19 01:12:01 mranostay: ping Mar 19 01:12:40 jkridner: posted Mar 19 02:17:46 responded **** ENDING LOGGING AT Wed Mar 19 02:59:58 2014