More on video recordings

No idea why the anonymous poster did not approve my long comment for his post, having allowed another three comments after mine, so I copy my reply here.

http://yapc.tv is another great start :-)

Seriously (why did you post anonymously?), every small step in recording conference video is easy. All together it is a pain in ass for the organises.

1. Cameras are cheap indeed. In Riga we had a few HD camera that cost less that 200 €. Transporting cameras with tripods is a hassle (OK, I admit that most organisers really live in the city where the conference takes place, and this is not a problem).

2. You need a dedicated cameramen for each camera (thus, three of four additional persons in the stuff). One of the cameras in Riga was stolen. You'd better pay these cameramen and do not ask the attendees as it is difficult for them to sit the whole day in one room.

3. Post processing is another pain in ass. It's not simply cut the talk from the recording and upload it on YouTube. You also need to find the talk title, speaker's name, maybe attach a link for the slides. Links/embedding to the conference site are desired too.

4. Sound. The only way you can cheat and record sound using camera's microphone is a recording in the room with sound amplifiers. Sounds is recorded OK-ish in this case (see http://yapceurope.lv/ye2011/talk/3531 for an example of such recording). Would you like to have better sound, especially in the room which has no amplifiers, be so kind to run around the speaker before he start talking. Barely that means that we need another stuff guy in each room (he/she may also serve a time keeper).

5. Screen capture. Extremely great technique to record video that is displayed on the big screen. Extremely difficult to implement at YAPCs with very few exceptions speakers demonstrate their talks from their own laptops.

6. Permissions from the speakers. Well, not that big problem, and you've got three categories here. a) those who ignore your requests, b) those who agree immediately, c) those who would only agree if they see the video recordings first (which means that you may do all the above with no guarantee that you will be able to publish that).

7. Money-wise. Vimeo PRO account, or a YouTube account, or hosting for the site are not expensive. Time to prepare all above is quite costly.

I have lots of video files archived in my disks at home since 2005, and I have no idea of when this will be (if ever) published. Also keep in mind that it is very easy to receive messages (both public and private) that like "you did a terrible mistake/fucked up/you are idiot", after which you feel that it is just insane to continue volunteering.

8. Together with that you may encounter a speaker who thinks he tells something secret and turn your camera off (here you are: http://blogs.perl.org/users/andrew_shitov/2011/11/lets-say-no-to-no.html). Upd. Ha-ha, there are more comments at that old topic added by the newsmaker a few months later.

Anyway, saying for this year's conference in Kiev, we either will not record at all, or will do the recordings with no guarantee to publish, or will invite a third party. In the latter case we need just to find another 5000 € for that to be done. So simple.

13 Comments

I posted anonymously because I don't maintain a blog. This is my first posting here. I also wasn't aware that blogging here automatically requires me to manually approve comments. If so can you point that out to me? I just wanted to share with the community, not learn MovableTYpe. My intent was to suggest a positive idea.

I took the time to compose my post with an intent on clarity. But I should accept why blogging is difficult specifically and communication in general is hard even when talking about about a linguist's computer language. I will try to flesh out more more in succinct statements:

-Central repository for videos. Googling is harder than it needs to be. Why make this a problem for Perlers and foreigners alike? Lower the barriers to finding them.

-Recording with recent hardware. Even smartphones of the last 2 years have optics with extremely performant relative apertures. Set the camera on a horizontal surface and record. Audio recording is performed by another device(either another smartphone or something like a Zoom brand device at close proximity of the speaker). What kind of knowledge is expected of a provider: is the camera level, is the camera on a stable surface, is the lighting adequate, is the audio recorder close enough to worry about not echoes but rather exhalations of the speaker.

-Post processing. Why does this fall to the idea of a single entity shouldering this burden? Crowdsource it to all the people who are interested in viewing this. Separate audio and video matching is not a task that requires esoteric skills.

-Perl was and still(unfortunately) somewhat is CGI. Hosting is a not a problem today is it? This is part of the reason for fundraising or separate cost-center within the Perl Foundation.

-Licensing.Permission? Since the recordings are performed by the YAPC organizers, permission can be given in the same fashion that talk ideas are submitted. If approved for a talk, do we have permission to record your speech for use? Yes or no. If no, then no recording is performed.

We are still a large community. Why do we think so many things are impossible?

Ok, you can call me Bob.

So digging around in the interface here at Movable Type...apparently local posters are expected to manage their spam comments.

I see you have the same comment posted twice and Movable Type marked this as spam.

The details:
SpamLookup - Link 0.0 Number of links exceed moderation limit (3)
TypePad AntiSpam -1.0

Is this the best we can do with blogging software? You're a longtime poster so why would MovableType flag someone who has been here a while? Any help from administrators on this?

I will say again regarding cost of processing...community outsourcing. We have a large community. Many who can't attend would surely volunteer to process.

Regarding equipment cost, please read my initial posting.


I asked...can we create a mechanism to do this specifically such as fundraising. TPF can then use these as grants for YAPC orgranizers. The issue of equipment cost is also the reason why it would be shared between a few of the larger YAPCs.

Did I ever say money was the only key to making all of this happen on a quality level?

Ok, so let's talk hypotheticals and worst case scenarios. Best to get doomsday out of the way so we can worry about Monday instead. The first YAPC to get a grant would be you! Enough to purchase 2-3 cameras and 2-3 handheld audio files. Everything is set up and recorded. Files are uploaded. Post-processing volunteers then get to work on what you've captured.

Everything is done and then reviewed by a few to see if the quality of the captures meet expectactions. If not, guess what...TPF has lost some money. Yes! This is terrible! But you also won't get a grant next time so the chance of this happening is greatly reduced.

Why do I feel like I'm butting my head against the idea that simple recording is such a difficult idea? If you are not familiar with contemporary recording devices then that is ok but don't keep responding that making a decent product these days requires a lunar orbit project. As an example, peruse kickstarter.org to see how great modern (and cheap) devices record. Many of these videos were not done by external media companies but a device recorded straight-on and then processed with commodity software.

Is this easy? No. Is this difficult? No. Would this help the Perl community? Yes. Can this be done? Yes. Now how do we do it as a community? Let's figure it out.

The process you describe makes me think you haven't done something like this before. I have neither.
But full HD (especially raw) video files are BIG. So "uploading" wouldn't be as simple as you write. Same about recording: smartphones can record full HD, but surely not an hour or more of it.
And postprocessing/editing takes time. Of course it can be done, but I follow a lot of YouTubers who do this for a living and they are really tied up with this.

I would love to see more recordings of these events, but don't think it's as simple and easily done as you seem to think.

>Volunteers are not the solution, btw.

I agree. Just imagine if a language was written like this.

>Anyway, I do not feel comfortable communicating with an anonymous, even if there's a lot to discuss here, sorry.

I also agree with this. I proffered an idea for the community to take hold of and manage or mangle to an acceptable end. Since it is a community of volunteers I ask nothing in return. Bikeshedding is best left to those who are not anonymous.

Hagen...Excellent points!

How could we overcome the upload bandwidth issue from the raw capture? What if the "suggested" capture device has removable media? At the end of the conference, the organizer ships the card(s) to yet another volunteer who does have high bandwidth such as a 100MB or Gig connection? This is not common but also not unheard of either.

Yes, the processing does take time. There is no mechanism to get around this other then to recognize those who are willing to "TUITs". Perhaps if we have a pool of volunteers then it is simple done as a means of first availability? The post-processed files are still large but some sort of contempoary encoding should solve that. Or another round of shipping a media card to a high-bandwidth uploader?

Anyway, I do not feel comfortable communicating with an anonymous, even if there's a lot to discuss here, sorry.

He is pseudonymous, not anonymous.

Would you refuse to communicate with chromatic? Or Abigail? Or Father Chrysostomos? If this person had chosen to sign up as Robert Sander or Richard Lark or Walter Pilgrim, would it even have occurred to you that it might not be their real name? And even if it’s not, what difference does it make here? Do you know whether my name is in fact Aristotle? Whether anyone else who call themselves whatever they may call themselves, is “actually” called that? (What does “actually” mean, anyway?)

When someone under a pseudonym responds to something you write, you know that it is the same person you wrote to. What meaningful gain beyond this is there in knowing someone’s “real” name rather than their pseudonym?

Unless you were simply reaching for any which reason to end the conversation, I am utterly baffled.

Leave a comment

About Andrew Shitov

user-pic I blog about Perl.