 We are having a meeting here. This is actually more entertaining than above. A little bit of what the video team is doing. We have several teams here working. The teams are this abstract for the proceedings for our camera team. How do you mix it? Do we have extremely authority? Extremist is the most important. Not knowing and archiving is not really... I'm not sure if this will happen from Mexico. And collecting of the slides of the speakers, which needs to be done and producing in the east. These are the areas of the team that we have. We also have some... Tomorrow, today, we have network here in the morning. I'm not sure there are some other open questions. Should we use the meeting? We started the recording for the last conferences. For other stuff, like for example, we talked with Foster, but there are also other talks, which might be interesting for Indian people. The same goes for other conferences. One of the issues is that at the moment, in the last year, I was the only one maintaining the archive, uploading stuff there, putting slides there, organizing stuff, and I would like to have a team doing this because in a year or two years, I won't be fed up with this if I do it alone. Also how to inform about new content. Last year, I mostly used my blog to say there are new talks available. I was thinking about creating an RSS feed or announce mailing lists, or I don't know. Another question is what to do with the source files. Today, we've gathered 100 gigabytes of data. I think we can throw away 30% or something, but DV is still the best quality. Should we keep it now and upload it in three years, or should we throw it away or whatever? Another thing is backup of the complete archive. I upload it to one machine, from which it gets r-sync to another, but if it's deleted on the one machine, the r-sync mode automatically deletes it on the other, so I keep a backup of everything I upload on my hard drive at home, and I'm buying hard drives. I've bought two hard drives in the last year, I think. License stuff. At the moment, we use a software license. We discussed and forced them to create a DepCon video team license, which was when you wanted to do that to... I haven't done that, but I should think. I can reword the MIT license slightly, so it's failure rather than software or something like that. Yeah, I knew it was easy, but I didn't remember what to do with it. The last thing from my list from this proceeding was sponsorship. Maybe find sponsors for DV cams for video team members so that we can actually test setups before conferences and not in the last evening before. And also, the last conferences I did, I borrowed DV cams from different friends and copied all the tapes, and the thing what makes... The cameras... Playing DV tapes is more... Does the camera hurt more than recording, so I don't feel comfortable always asking friends for the cameras to grab stuff from the tapes. That's the list of things I had. Yeah, I don't know how to go on with this buff at the moment because I'm completely tired. So one thing which we not discussed here and were really a volunteer would be needed would be collecting slides from all the speakers so that we can upload the slides together with the talks. Usually 30%, 40% of the speakers sent in their slides, then you send a request mail for the slides and another 30% sent them. Yeah, hand over the microphones. We cannot do pictures as last year in Helsinki. In Helsinki we had a hardware DV mixer and we don't have that thing now because of the lack of DV cams also as a reason. It wasn't a DV mixer, it was an analog mixer. Okay, in Helsinki it was an analog mixer but we thought about preparing a DV mixer or anything and we couldn't do this because we have no cameras. And at some Spanish conference in Mallorca, I don't remember which it was, the SCP, the slides, the screen shot from the speaker's desktop to a web server so people on the net could see the slides. This is one thing, but the other thing is to actually get the slides, get the PDF or the sources to upload them together with the movie. This is a different thing. So this would also be another job which would be nice if somebody would do this which also involves going to the speakers, installing the small script to upload the slides but it would also be good because as we see today some speakers don't test their laptop with VGA out and come here and don't have a way to display their slides. So this is also a task. Tullima is going to all the speakers and tells them that they should test the notebook and they should speak with the audience and they should do whatever speaker information and so it would maybe good to hook up with him and to go to all the speakers and prepare this SCP thing. So do we want the SCP thingy? I think it would be nice. The streams are not so good quality and some slides are not really readable. Like in the Intel talk, she used very small forms. But if we have no internet here, what's the point? We do have internet here. We do. Yes, it's just turned down today because we did not manage the bandwidth. I can work in that. Yeah, it would be cool. But as you said, if the... Yeah. I should have tested it better, I guess. Yeah, but I don't think it's so difficult to do. Well, sometimes it makes the whole screen flicker with every capture this script makes and that's very completely... I can try. You can do it. Yeah, then somebody would be nice if somebody would say, I will collect the slides and upload them. Yeah, cool. We could do it as a team. Exactly, but just, yeah. Have a mailing... Okay. No, he doesn't have to... Yeah, we all have a... You not, but everybody in the video team and we can add you as an account on Homa. Maybe everybody has an account on Homa now. And then you have an account on Homa, you have an account, I know. You got mail with your password. Oh, that mail. And there we can just create a shared directory for the video team where we can upload all slides and put them, yeah, collect them. The other thing, we need to work on the encoding of the videos. Like now we have the raw DV files and we need to write some scripts which will encode them to different codecs and different qualities, which should not be that hard. The script is easy. I've written those scripts for other conferences and we just need to define what qualities do we have and then we need to decide if we can keep those DV files, if there are enough, or throw them away because we don't have the hard drive space. I think we should see this in a few days. We still have at the moment 900 gigabytes free space left on this machine and on the other machine and the heckler is also one terabyte. So that should work. This is a borrowed machine. So we will presume we cannot take away the disks. So unless we have disks to copy to, we cannot keep those original files. Yeah, I was already counting on your hard drive and my hard drive, which is together half a terabyte, that we will most certainly need to... There won't be enough for the DV, will it? One hour of DV is 12 gigabytes. So... And with the encoders... We will have more than 40 hours. Yeah, and with the encoded stuff in different formats, so we can say one hour is probably 20 gigabytes, including raw DV, maybe it's only 18, but if we have 40 hours, then it's 700 gigabytes, so we will need another 300 gigabyte hard drive to get everything... We'll keep the tapes. We can keep the tapes, but then we have 40 tapes, which will be, again, lots of work too, but that would be a way to store it and copy it later, but I don't think... The tapes are... The DV and the raw DV is going to be just a backup, because no one's going to want to download that. So it's a backup if we really need it, I think. I think nowadays nobody will download it, but in five years maybe it's... Then we have the tapes. Yeah, but then we won't work with the tapes. Then the tapes will be in some dusty corner of some cupboard, and I won't be interesting in 40 hours copying the tapes. It's not 40 hours work, but it's 40 hours. I think we should definitely keep the tapes, but what if in five years devices that can copy off DV tapes faster than real-time will be actually affordable. Right now they're not affordable at all, but maybe we might in the future be able to borrow or buy one relatively affordably. Tourer hypotheticals are very nice, but I don't know if that's really realistic at the moment. No, no, no, but the point is in five years we will still have the tapes. In five years we'll have another five years worth of tapes as well. Sorry. Can you turn the camera away now? Okay. Tourer will just build this machine, no? Tourer knows about tapes. We have the setup with the streaming that we constantly produce streams and write it to the hard drive and get files which are labeled I think tower minus date minus hour or minus time. And that would be another job to on the end of the day to rename those files into useful file names for the encoding. Otherwise we'll get much more files which we'll need to rename. I think this and also delete the other stuff so that we get... I think this would be good to have it in the video team tasks schedule like we have now tasks for audio mixers, video camera operators that there is every day somebody who volunteers to go through the files and we need actually to have a way of viewing the DV files as it is. There's... As Marge is currently configured there's no way of actually viewing the DV files from there. We would need somewhat to be able to serve those to... I mean you could serve them to the grabbing machines and install a... Toge and Morph DLC or whatever on there at the moment that isn't done. We also have the problem that we... The hack lab will be closed... At least the tower will be closed at some moment in the evening and we only have a 10-ambit link to the rest of the network so we cannot really watch DV files on the other end. Or can we? No we can't. DV certainly not. So we basically need to encode everything and then decide after encoding what we throw away, right? We can figure it out fairly well by the timestamp, by looking at the modified time. You have to tell us when the recording ended. Like the way you work now you interrupt the recording when the talk starts and interrupt it when the talk ends and so we have one 10-minute break or something so we can throw away the small fires. No. We actually have one break between... We tend to have one break between talks. In practice, the way it's been done I've tried to do it this way but I don't know if it's been done this way consistently but I interrupt it before each talk and after each talk which then produces a bunch of small fires which we can then throw away or just echo nil into it so that your serial number generator won't be confused because that just iterates over each serial number. Isn't it so that in the current schedule there are no consecutive talks? There is always a break between the talks. There will always be a break between the talks because at least the speaker changes. Yeah. I don't see where is the point because you don't have two talks following each other yesterday but I mean the coming days? Yeah, but there will always be a small break like people will get in and get out so that's not really the point that's just the point that we will have small and big fires which we know we can throw away probably but I would still suggest to encode them first to view them shortly. Another reason to continue making the small files is that as long as the breaks in the file are short enough we can continue streaming them so that you only you don't stop the stream which would force the clients to reconnect you just stop recording or just make a sort of mark in the recording. But if you stop if you stop the recording now then it automatically reconnects do the clients then have to reconnect? Not if they don't run out of buffer space I believe from my personal experience sorry from my personal experience what further away what am I doing wrong from personal experience when I run an ice car stream without the source briefly it sometimes will if it's briefly enough it will not reconnect but the way it's set up now was how brief is briefly enough? less brief than what we're doing now because what's happening now is that when we interrupt it's just a quick rerun of the files then half a second probably you're always just saying I'll see that they did have to reconnect between Denise's talk and this one so it depends how briefly that was yeah but we encountered some strange network related problems if you scroll up here you will see that it fails to log in to the server several times but the way this DC6 record whatever script is called it just automatically restarts the capturing and the streaming is this break short enough? yeah it should be anyway but we cannot make it any shorter anyway it's automatically started instantly actually there's a one second delay in there simply so that you can interrupt it properly so that could be removed but it wouldn't be very difficult to stop the loop then it would be difficult to to break the script we could just reduce the delay, do you know sleep allows for floating point delay numbers if you just say sleep dot one seconds you might decrease the chances of that happening I believe so yeah this is getting a bit into details yeah it is yeah but I think it would be nice if we try it but this is the details of topic or details but we already have I have not counted the files 20 or 30 files which we cannot review tonight because we won't have the bandwidth to view them well we can if we encode them first yeah if we encode them fast but how fast is the machine here I don't know I think it's just a single processor with three Gihats or something there are 27 files by the way there are at least two processors so in my experience if you load the quality the encoding will run much much faster and you don't need much quality for review yes but we will need to have to set it up and I think it's unrealistic that we'll start the encoding things before dinner maybe we should just do it before dinner so during dinner it gets encoded then we can review them after dinner or maybe we should have a task force tomorrow or one person or whatever to review them we can try starting encode them which can we decide quickly which formats and we can just encode them now without deciding so what we streamed would should be enough to review them and do proper encodings later we just throw them away fine there are 27 files 7 of them are of zero length and how many are below 2 gigabytes most of them 12 minutes hang on I'll just grab 6 of them are large as in over 1 gigabyte and how many talks did we have today 6 yeah that basically works out then cool I love it when that happens is there anything else we need to decide now I'm interested in your ideas of team maintenance of meeting archive dbnnet I'm not quite clear what how much work that involves what the tasks are do you mean the encoding I mean the archive well all the recorded talks are stored all the depthcon 5 videos and the foster videos and the other conferences videos because at the moment these are mostly talks or conferences where I have been and if you go to a conference to make recordings there at the moment there's no way to upload them to announce that which I've right so it's not so much that there's more work than you can it's just not so much that it's too much work for you it's just that it's an access issue yes also that but there's there's also leftover work from depthcon 5 the we had a nice overview or we still have a nice overview page on the dbn wiki which I think would be good to have it moved there there are still some slides in my inbox but people send them to me and there are still slides missing and there are some DVDs were not produced there's stuff other people could do and this is basically just a call for your participation participation and how the access issue then we will be solved can we discuss and practice when it's due okay you want a mic so let's stop now might be a time now might be a time to assign people to task for tomorrow yes cry the noon I just remembered by the way that today on IRC lots of people for various buffs so I think it's interesting to stream buffs and I think we have the setup so we just do it in the other hack lab it's only operating a camera basically no we have not tested this today as planned yesterday by the way but we tested it which is the same setup so it should just work with little preparation and tomorrow there's this welcome we went at 11 I think and then there's the first thing scheduled here at right after lunch I don't know I don't think so there's the schedule with the all the handwritten editions to the schedule I don't know if there's an online version I hope so yeah there's something in the wiki for all the buffs and stuff in the schedule there's actually one short session here before lunch yeah tomorrow Open Selahat and Debian can we be friends there's also just before that the welcome thing which is going to be here of course aren't you looking at the schedule Neil not at the schedule at the I'm looking at everything that's in here here being the parliamentary tower we need to assign people to the high ground as well there are buffs tomorrow the only extra ones that I remember off the top of my head is no we wanted to discuss to settle our tasks our schedule for tomorrow I think okay I think the world has to enjoy a very overworked video team now hello by us be be it would be nice tour has been disenfranchised so Natti you don't have your list here all of you bastards so then we can decide to talk about this list during dinner or before dinner preferably before dinner that was when Neil was looking at this computer there isn't nothing here is it no wireless internet anyway boom have you recording that have you seen what they're doing ah streaming porn she was naked again should we stop the recording now and discuss