User:VGrigas (WMF)/How to make Video using Wikimedia

THIS IS A DRAFT AND WORK IN PROGRESS

How to make video for the Wikimedia Movement (5 parts)

Introduction

Do you want to record a video for Wikimedia? Here’s how.

Every year, the ability for ordinary people to make and distribute high-quality video becomes less expensive and easier to do. Over a billion people now have tools for making video in their pocket. If you have a smartphone, you have a movie studio. This is incredible if you consider that just 20 years ago, consumer digital video was new and expensive. Feature films have been shot using phones, you just have to learn how to use it. I’m optimistic that if you read this blog series and are bold with experimentation that you can make the video that you want to make, as long as you realize the limits that you have and are patient with your production.

This guide is by no means comprehensive. It’s a start. You can also find lots of great advice on making video on creativecow.net, philipbloom.net (whom I’ve had the pleasure of working with in the past), Kenrockwell.com, B&H tutorials, Rode University, Vimeo Video School and Vice Creators Project, this post will also be forked here on meta, so YOU can add what you know to keep this updated.

This document is split into 5 parts, (generally how it’s taught in film school): Idea Development, Pre-Production, Production, Post-Production and Distribution.

TLDR:

• 48k 24bit 24fps

Idea Development

Idea development is the phase when you figure out what exactly you want to make, draft an outline, and estimate what your budget might be. Sometimes you may start thinking that you want to make a video documentary, but then it turns out that you actually want a slideshow or an animation. It’s also normal for ideas to sometimes go unproduced, or shelved.

The first thing I’d recommend for the idea development phase is to figure out your message. What IDEA do you want to give people? What do they need to understand and how should they feel about it? Your video suggests things to think about - so what do you want people thinking about? I’ve found that it helps to find your message if you identify major and minor THEMES. For Wikimedia I’d say that themes of international, open, transparent, free, knowledge and education all work. You can have multiple themes for your project. Themes can be shared with any artist and work across any creative endeavor (music, writing, dance, photography). Once you have your themes, you can start to get an idea of where to go with your work, but more importantly where NOT to go with your work. Themes are your first guide.

This video series was produced with teachers in mind as the audience, because it’s about how to use Wikipedia in educational settings.

Figure out your audience for your video. Don’t think that ‘everybody’ is your audience - if you can narrow who you want to see this, then you can really think about what might resonate with them. You can imagine what values your audience may or may not have, and how you can appeal to, inform or entertain them. If you are making a video to play for a conference, you can assume that attendees of this conference will be your audience. You could also assume that these conference attendees can’t press fast-forward on the video like they could if your audience was online. Audience will help you to determine the genre or style of your video. Are you making a video to illustrate a topic within a Wikipedia article? Then there are some general guidelines to follow to keep it ‘encyclopedic’. Are you making a ‘silent newsreel’ for social media? Then you’ll want to make sure your text is big enough to read on a phone. Try to narrow audience down as much as you can.

Wikimedia Commons has images that you can legally remix. You can SHOW this idea in a video with pictures and sounds and you can TELL this idea with words.

Organize your ideas: You can use index cards to organize your ideas. I use physical paper post-it notes. Write each individual idea you have on a post-it. I try to follow the rule to SHOW, not TELL when possible. This is usually a balancing act: when do I have to say something with words or text and when can I show it with pictures or sounds? When I have my ideas on post-it notes, I can break each idea up by all the possible ways I can think of that could show or tell that idea. For example, I might want to make something to make people aware that Wikimedia commons has free images that you can remix. So under my ‘remix-(theme)’ post-it note, I could write a sentence “Wikimedia commons has free images that you can remix.” and note that it could be read by a narrator on screen or off screen or it could be text-only on screen or both. These are all in the ‘tell’ category. Under ‘show’ I could make a note to see the Mona Lisa and Genghis Khan reading the 1912 Sears catalog under the Egyptian pyramids as an example of a remix. It might be goofy and confusing, but now is the time to draft ideas and decide what might or might not actually work. I like to start wide and then narrow it down. Basically, find the key ideas you want to communicate and then decide the order exactly that might be best to show or tell them.

Put your ideas in sequence in an outline (or into what’s known as a ‘’’film treatment’’’). Place your ideas in a row all next to each other, and ask yourself ‘does this make sense?’ The sequence of your ideas should accept the level of understanding that your audience has about your topic. You know a lot, but does your audience? Put fundamentals first. Make sure that your audience has the information they need at every stage of your video to fully absorb your message. On Wikimedia projects, you could put your outline/treatment on a page on Meta, and then let Wikimedians add to it or comment on it.

Make backup plans. Not every idea works when you actually edit your video, so keep other ideas as backups in case you need them later. Sometimes the bad ideas turn out to be good ones.

Guess at a budget. What do you think this might cost to make?

Below are a few examples of the kinds of videos that you might want to make. I include them here so that you can see and think about the differences in style, production quality, cost, format, production time, and so forth.

Stock Footage and B-roll

This newsreel was produced after we had found a true story about MediaWiki being used at NASA. I came across the public domain 4k video footage on Archive.org and edited it to fit music from Jamendo.com. The photo at the end was part of a photo set shot by a photographer that we hired for one day.

CreativeCommons.org estimates that there are over a billion creative commons-licensed works out there. Much of that is ‘’’Stock Footage’’’ that can be remixed into various productions without having to pay any royalties. Here are a few libraries you can look through to get started:

• Archive.org (NASA puts much of their footage there, Prelinger archives is awesome)
• YouTube.com - (firefox has an add-on that makes it easy to download videos)
• Vimeo.com
• Wikimedia Commons - You’ll likely have to convert it (more on that later), but there is a lot of footage, much of it migrated from the above sources. Just search ‘filetype:video’ press space and add your search term.
• You can search via Creative Commons as well: https://search.creativecommons.org/

In this case, imagery of the WMF offices were requested by the PBS NewsHour, so it the above footage was edited and released.

B-roll is extra or alternative footage of something that’s usually produced as part of a larger work (slightly different definition from stock footage). You may find that if you plan to shoot video of real locations or events to use to illustrate Wikipedia articles about those locations or events, B-roll might be a good container to use. As a volunteer Wikipedian, I’ve used it before. It’s fun to take shots with your phone and cut them together in this way.

The Silent Newsreel

This is a silent newsreel from 1918 made by the great early film director Dziga Vertov. By the way, does anyone want to add captions on Commons? I don’t speak Russian, but I’d love to know what the innertitles say.

Facebook revived the 100-year old silent newsreel when they enabled video that auto-plays on your Facebook wall with the sound off. Today if you make a video for social media, you have a high chance that people will be watching on their phone, and that the sound will be off, so intertitles, subtitles and captions are key to communicating with video on social media. There is a good analysis of this style of video here. What I’ve learned about this style: it helps to have FAT BOLD FONTS (I use OpenSans) for innertitles and when people talk, I use bold yellow subtitles with a blurred dropshadow. if there is text in the background, italicize the titles you use.

Below are a few examples of silent newsreels I’ve made for the Wikimedia Foundation:

This is a newsreel that uses only text. This type of video can be cheap and fast to produce. The music was reused from a previous production.

This newreel uses text and a few images with a few of video filters to make the point that images can be remixed. Note that the text and images are paced to the beat of the music, which must be done carefully so that everything appears in sync. If a few frames are off, it can feel off.

This is an example of a ‘silent newsreel’ that uses live footage.

This was made using footage sent in by volunteers. Each shot needed written approval from the shooter for a joint copyright agreement with WMF so that the video could then be shared on the Wikipedia social media accounts.

This newsreel was shot by a contracted filmmaker who was also hired to shoot interviews for five days at the Wikimania 2016 conference. Upon completion of photography, she sent the appropriate shots via Google drive to me to be edited. This was produced within a few days of the end of Wikimania, and with significant planning ahead of time to clear a schedule to be able to turn the footage into a final video while Wikimania was still fresh. The video incorporates footage shot by a volunteer and by videographers at Wikimania. That footage was downloaded from YouTube and Vimeo and credited appropriately. Versions were produced with burned-in subtitles to play on social media.

Esto es un editatón de Wikipedia - 'This is a Wikipedia edit-a-thon' - a newsreel made in Spanish. A video shooter was hired to document the event for one day, and was instructed to only photograph art that is in the public domain. The video shooter used a Canon c100 and a DJI drone to photograph the event. Music comes from Dub Terminator which was licensed CC BY-SA 3.0 on Jamendo.com.

This silent newsreel uses public domain images from Wikimedia Commons and advertises Wikipedia's Instagram account. It’s under 15 seconds per Instagram's 15 second video limit.

The WikiArabia tech meetup in Ramallah 2016. This required creating an Arabic transcript of all the dialogue recorded, and then translating that to English, picking the phrases that might work for the edit, and then having a native Arabic speaker time the phrases, then make the edit, then run the edit back by the Arabic speaker to make sure it's timed well and that words aren't cut off. Hired a local production company to shoot for one day.

The 2016 Wikimedia Hackathon in Jerusalem. The video shooter was given a short list of questions to ask everyone - Who are you? Where are you from? What is a Hackathon and what are you working on? Covering the who, what, when, where why and how of the event. They interviewed about 6-7 people and then the footage was edited for the best parts of the interviews.

Animation, Slideshows and Screenshots

You may want to create a tutorial or other video that uses screenshots from your computer.

'The Wikipedia Content Translation Tool reaches 100,000 translations' .This video was developed as both an announcement of a milestone achieved by using the content translation tool, and a tutorial for how get started using the tool. It took about 3 months of intermittent work to create from start to finish. We first drafted a script and then captured screenshots to illustrate the tool. It took several iterations for everything to make sense. We had to hire a 3D artist to animate the books that appear and a video editor to take screenshots.

This video was the centerpiece of a brand campaign in 2014. It took many dedicated weeks to develop the script, research the year-in-review format and find imagery. Had to hire a musician, motion graphics artist and a video editor. In the end, we only used a small portion of the music that we had commissioned, and instead used a public domain work from Wikimedia Commons for the majority of the work. We also unexpectedly found that we had to record typing sounds from a few different keyboards to simulate the diagetic sounds of different users around the world. More here.

This is a Product Video. Creative Commons works are not compatible with many (most?) social media terms of service. It is important to note that in order for a video to auto-play on social media services like Facebook in order to receive more views (as this video was designed to do) WMF must be a copyright holder of the media being shared. The video above uses music that was specifically composed for WMF and by contract the musician signed the copyright of the work to WMF.

Slideshows can be used to showcase still images in sequence, and making them into a video allows you to add music if you want and choose your pacing.

This was a way to show the top fifteen photos of Wiki Loves Monuments 2016. It was made in 4k and at 60p to better show the detail of the photos, and fade between the images more smoothly. Note that since each image is a different rectangular shape, in order to fit the images into a fixed rectangle of the video, some space was left either on the top or bottom of each image.

This animation required booking Jimmy Wales' time in a London recording studio, and weeks of iteration to turn each line of script into a corresponding animation. Music was reused from a previous production.

This video uses videos and photos found on Wikimedia Commons along with music originally from Free Music Archive to show the breadth of visual resources on commons while advertising a Wikimedia strategy site.

The Talking Head

Any time you interview someone on camera, you probably want to see their face say the words that you are recording as audio. Below are a range of ways you can use that kind of footage, either by itself, or mixed in with other media.

This is video 1 of a 12-part video series about how to use Wikipedia in education. We hired a video producer in Stockholm Sweden to interview and record attendees of a 2-day meetup of people using Wikipedia in education from all around the world. We ended up with about 10 hours of footage which was edited down to about 30 minutes of content across several topics. Over the talking head footage, we placed still images and other footage and for concepts for which there were gaps in, we used intertitles.

This talking head video did not use a teleprompter, and aimed for authentic, unscripted (albeit prompted) responses. This was recorded at Wikimania 2012, and was the result of 100 interviews conducted over the course of 5 days. It took many weeks of outreach to screen and schedule the interviews, and prepare logistics for recording them. The interviews were transcribed, and 25 interviews were edited into individual testimonials, which can be found in this campaign from 2012. After those testimonials were finished, the video above was edited. Shoots such as the one above often require renting lights, backgrounds and other production equipment.

An individual testimonial from the series above. Once the equipment is set up, it often can save money to shoot several interviews in a row: most shooters (videographers) charge a flat day rate for anywhere between 8-10 hours, so it makes sense to get as much out of them as you can if you hire them.

Another video from the series above in Spanish. Creating videos in other languages can increase costs of interpreters, transcribers, interviewers, captioners and sometimes video editors who would need to be able to speak the language needed (and sometimes understand cultural relevancy) in order to create such a video.

This is an Extended Interview with Ward Cunningham (inventor of the Wiki). His story deserved an in-depth on-the-record account, so this was used as a centerpiece of a blog post for his 65th birthday.

This is a Skype Call recording. A program called Call Recorder allows Mac users to record video and audio calls over Skype on their computer. In the settings, you can configure the size and quality of the image you record, as well as whether you record your video, the callers video, or both in the same image. In this case, the interviewer couldn't figure out that setting, and recorded both the interviewer and the himself, and then the final image had to be cropped in post-production. It is important to note that image and sound quality is often low for these types of recordings, due to webcam and bandwidth issues.

This is a Talking Head video. This took time to light correctly and requires a teleprompter, plus a script for the subject to read. This type of recording can save time and travel costs of attending events. A high-quality lapel microphone is recommended.

You can also shoot a webcam recording of a Talking Head video on your own. You probably need to buy a cheap lapel microphone. For reading a script from a laptop like a teleprompter, I use this site.

This Talking Head video also uses b-roll and screenshots to augment the message. A high-quality lapel microphone is recommended.

Documentary

Documentaries can be expensive, usually because they are time-consuming. Below are two documentaries I’ve produced, one with a low budget and the other with a higher budget.

This is a Short Documentary. It involved months of planning, location permits, airplane tickets, hotel lodging, contractors (music, editing, motion graphics, shooting, color correction, audio mixing), and took over a year to complete. The campaign petition required development of mw:Extension:Petition a petition extension. We produced four versions of the short, and released a version (designed to stand alone) of the students reading their open letter.

This is a lower-budget example of a short documentary. No contractors were hired. It involved one week of travel and lodging and myself to shoot (while using footage shot by conference videographers). It took several weeks to edit.

Pre-Production

Pre-Production is the phase when you prepare for your shoot. It includes everything you do to prepare up until when you press record. That means finishing your scripts, scouting locations, finding crew (for Wikimedians, this may mean hiring a professional photographer or editor), finding your budget and so forth.

Often in the production phase (that’s when you are recording video and usually when most of the money is spent) it’s good to produce more than you need, so that when you edit in post-production you can decide what footage is most effective, or eliminate anything unnecessarily redundant or superfluous.

This video was the backup plan. The original concept was to mimic the style of the Art Institute scene in Ferris Beuller’s Day Off, but when it was edited, it didn’t feel right. So fortunately we had captured extra footage of the museum and the event to edit in the way you see above, and had planned for it in pre-production.

Turn your ideas into a script. It’s important to try to learn and follow the script style and format because the script is the main document you can share with anyone else who might become involved in your project. (show script style - script of pine how to edit series, and translation tool script) If you draft a script on meta, others can add their thoughts. I often think that Wikipedia articles could have scripts written collaboratively on the talk page of an article to better illustrate the article. That way, anyone who likes writing but might not want to produce video could draft a script, and anyone else could check and see if there is already a script and (since half the thinking is done) produce it. I’d personally love to find people who might be into this and brainstorm a way to make a production queue for a system like this.

This talking head video was made by Wikimedians at Wikimania 2016. If it was read as a shot list, the sentences spoken by each person would correspond with each shot. The shotlist could read: Shot 1 -- line of dialogue, Shot 2 -- Line of dialogue, and so forth.

Turn your script into a shotlist. Take each line from your script and turn it into a shot or shots that the viewer will see, make this a list. You may find that you need storyboards too. Your shotlist can be a the number and description of your storyboards.

Now you can make a proper budget. Look at each shot in your shotlist and estimate the cost of making each shot. Put a price on each shot and add it all up. Estimate a range of what it might cost on the low end and what it might cost on the high end. You might have one shot that you envision as a 10 second animation and figure you have a friend who could do that if you bought them lunch, but it might take them 6 months to finish. You also might know a professional who could do it for $1000 in a week. By the way, if you plan to see Superman reading a manual on Wikibooks about how to blow up the Deathstar, you might need a redraft of your script. Copyright applies, and you have to be realistic about what you can make. Know that since you are making media for Wikimedia, grants may be available to you. If you have followed all the steps above, and your topic is considered something that advances Wikimedia goals (like getting knowledge to everyone everywhere for free) you could get funding. Buying hardware might be tough but renting it and hiring professionals as-needed can be considered.

Develop your production schedule. Look at your shotlist. Try to book times to shoot things in a sequence that make the most efficient use of everyone’s time. You don’t have to produce each shot in order. Have a backup plan for all the things that could reasonably go wrong. It could be a rainy day, but there probably won’t be an asteroid hit.

Feel free to skip anything above if you feel you need to. If you have a script or an idea in your head, and a camera in your pocket and want to go shoot it yourself for $0, please go for it! I find it’s best to push and see if I can do something badly the first time and then slightly less badly the second. All these written tools aid in having ideas and scenarios for you to use to make your final work. They help when you are putting the final product together, but only assist in making the final product. If you can skip steps and still get the job done, then do it.

Crew

Sometimes you need help for your shoot. Make sure when you hire anyone that you specify the licensing you expect to publish their work under. There was one time I wanted to hire a very talented professional musician who loved Wikipedia and the project I was working on, but he just couldn’t bear the idea that anyone could then re-use his work for other projects that he hadn’t approved first. We couldn’t hire him. Draft a contract that specifies everything you want in it. I always ask for 24p as a framerate and at least 1920x1080 HD video. (I expect in the future that everything will move to 4k, but we’re not there yet and 4K can be overkill for some projects). I usually have contractors paid half up front, and half when all the footage is in my hand or on my computer.

After you hire someone, make a list of questions to ask the people you’ve hired about how they do what they do, and let them know that you will have questions. You’ve hired them! You can ask all kinds of questions about how to produce and use equipment. If you’ve hired the right person, they’ll answer all your questions. Some video-shooters feel that knowledge about how to make stuff should be secret, and that they will get more work for it. Don’t hire any of these people. Tell them they are crazy and misguided and might actually get a lot more work if they make themselves a hub of free ideas.

To find people, make a list of anyone you already know who might be able to do what you need. If you still need more help, here are a few ideas:

• Google search terms like ‘Video production house’ or ‘video shooter’ plus your location work to find local help.

• Sometimes film students can do good work. Sometimes not. They are in school and have access to gear and are knowledgeable to a level, and may have a level of experience. They all use Wikipedia and usually are on the cheaper end, but not always. Students usually make more mistakes than professionals.

• Look on Craigslist under gigs/crew (If you are in the USA). You can find professionals and volunteers there. Don’t rely on it however, you don’t know who or what you might get.

• Productionhub.com is a fantastic rolodex of professionals all around the world.

• In my opinion, Vimeo.com is the best forum for video producers. They have every type of creative commons license available to publish under and design their service for producers first, users/viewers second. Lots of producers publish their best work there, and you can contact them there.

Generally, professional video-shooters and crew charge half or full dayrates (8-12 hours) for their work, and have hourly overtime rates around $50-$100 an hour. If they are in a union, there can be more rules too. Prices can vary widely, but generally expect that a video shooter may cost between $300-$1000 USD for a full day anywhere in the world, and you may need to pay for equipment rental, food and transportation. In my experience, the more prepared and knowledgeable you are, prices are lower because it’s less work for the video-shooter - they have more things already decided for them and planned out ahead of time. It’s best to have a script to give them before they are hired. They’ll need to know what specifically you need them to shoot, what interviews you may want and what questions to ask and b-roll might be needed. They are used to creating enough footage to edit, and if you can let them know in advance of what will be happening, they can record it.

This was recorded with one lapel microphone and it should have been two. You can hear how the speaker on the left sounds more distant, while the speaker on the right is more present. There was also a loud fan in the ventilation that should have been turned off.

If you are going to record people talking, you many want to consider hiring a location sound recordist, especially if you are going to be in a noisy location. Monitoring both video and audio simultaneously while it’s being recorded can take a great deal of focus, and when the focus waivers, it’s usually the audio that suffers. Most video cameras only have 2 inputs to record audio, so you are limited to a total of 2 microphones or audio sources (like a mixing board output during a live music performance). Often this can be enough, but you might want someone focused on ONLY the audio that you plan to record live, so that you don’t record any bad audio. For example, maybe you want to record a debate, where there might be several people talking or maybe you want to hear a bird chirping and see it too, in which case you may need a telephoto lens to see it and a parabolic microphone to record it, each of which would require it’s own human-being to hold and aim at the bird. Location sound recordists can run you $300 a day and up, and have the same day rate considerations as video-shooters do.

Equipment

Every major city has at least one audio or video equipment rental house in it or just outside it. Here is a list of rental houses from productionhub.com, and for a general idea of prices of equipment, check out Gassers in San Francisco. Usually, you can rent a wireless lav mic from gear rental houses (in any city) for one day for around $25 plus a deposit for insurance. This is a bargain. Lenses are priced according to their cost, and you usually can play with the full range of lenses they have available to rent in the shop. Always test the gear in the rental house with your camera. You don't want to rent defective gear. Rental houses are the best places to ask questions because they make money if you rent gear and you'll only rent gear that you know how to use, so they have an economic interest in teaching you (this is why Rode makes fantastic tutorials about how to use their equipment). You’ll need a valid credit card and ID. Sometimes if you tell them what you are working on and they think it’s awesome they can be flexible about return times.

More about equipment below in ‘’’Production’’’.

Permissions

Sometimes you need to get permission to film somewhere. Just be polite, very honest and clear about what you are doing and ask those in charge what the proper procedure is to get permissions to film. Sometimes this can come from one agency or organization, or two or three. Sometimes no one will don’t stop you if you just use a cellphone, but the moment you have a tripod, suddenly you need a permit. This is stupid because the quality of video is extremely similar but it’s the reality, and you’re better off accepting it. When you get permission, make sure to have it in writing and have contact information handy in case officer-know-it-all wants to stop your filming. Sometimes getting permission can get you extra benefits. You may get exclusive rights to record in a location for a period of time.

Sometimes you need consent from people to record their likeness or obtain their interview. I get the subject to sign a waiver that also tells me their name, email, telephone and other contact info. It’s good to have this stuff handy. These are waivers that I use. It’s basically the same waiver and has been professionally translated by a legal translation service. I’m not advising that you use these, just showing you what I use.

(LINK TO CONSENT WAIVERS) ADD LOGISTICS GO THROUGH ALL YOUR GEAR BEFORE YOUR SHOOT

Production

Production is the time when everything gets recorded. It’s normally the most expensive part of any shoot.

Audio

You probably figured that I’d start out talking about cameras first, but half of your video is audio, and if the audio that you record into your video is bad then you will make your audience hate what they are watching and they won’t know why -- it will be because it takes them extra work to understand what people are saying.

Ninety per-cent of what is intended to be recorded during production is people talking (dialogue). The rest is a mix of things like presence (the sound of the room or space you are in), sound effects (actions of people and things), music and noise (things you don’t want to record).

When you record audio, you don’t record an object -- you record a space. It’s up to you to decide what you want more or less of from that space in your recording. It’s also up to you to use the tools you have to record appropriately. If you are recording a talking-head interview you’ll want to keep all the noise to a minimum. If it’s a documentary, you may want to let it all be part of the scene.

Before you record, you may want to identify all the things making sound around you and try to control them. Take a listen to the space you are in. Is there an air conditioner or ventilation system on? Turn it off, or see if you can adjust the fan speed. Are there other people there making noise you can get to be quiet? Is someone using a lawnmower? Ask them to take a break or bribe them with a free lunch. Airplanes flying overhead? Anti-aircraft missiles aren’t really standard audio engineering tools -- there are some sounds you just have to live with. Sound blankets can keep uncontrollable noise to a minimum and help to reduce echo in rooms with hard, reflective surfaces like concrete, tile or glass. Any soft dense cloth material can function as a sound blanket.

Aside from the room you are in, there’s also the interference that can mess with your recording equipment:

• Wind - Have you ever blown wind from your mouth into the mouthpiece on your phone? That’s wind noise. It happens when too much air pressure hits the capsule on your microphone, and makes an unnatural sound in your recording. There are a whole range of screens to use to cover your microphones to prevent this. They sometimes look like tribbles.

On the left is a blimp with a dead cat windscreen, and the right is a foam windscreen, both exist to prevent wind noise to different degrees.

• Handling/clothing noise - This is when you handle the microphone or the recording device and the sound of your hands or clothes travel through to the mic and end up in the recording. I sometimes wear bicycle gloves to soften this effect when I’m holding a microphone and try to mount lav mics away from extra clothing if possible.

In this video, at 0:07 you hear the speaker briefly from the onboard camera mic (which sounds off-axis), and then at 0:09 from the lapel mic. The speaker brushed her hand against the lapel mic, creating ‘handling noise’. Later, in post production, the audio briefly went from the onboard mic back to the lapel mic. It’s best to record interviews from two sources to be able to patch situations like this, otherwise you have to pick between noise or silence.

• Electromagnetic interference - Sometimes radio waves in the air get into the wires of your device and change the signals going through them. Usually this is prevented by avoiding using wireless transmitters when possible, using only as much cabling as you need and not crossing audio cables with power lines when possible.

Like frame rates for video, audio is recorded in portions of a second (called a sampling rate), but many more portions. An old CD plays 44,100 little pieces of audio in one second. Standard audio for video is 48,000 samples or 48k. Each one of those portions records a number of ‘bits’, usually 16 or 24 bits. Generally, when recording audio for video, it’s best to record 48 kilobytes-per-second (kbps) at 24 bits. If you plan to slow down your audio for some reason, you may want a higher sampling rate.

Microphones

This talking head video was made by Wikimedians at Wikimania 2016. I like the video, but there are a few details I would have done differently. The audio now sounds noisy, probably because it was being recorded from a microphone mounted on top of the camera or built into the camera. The video could have benefitted from a lapel microphone that would have pinned to the shirts of those on camera and plugged into the camera directly. The lighting seems a bit harsh in some situations, and a tablet (or phone) teleprompter might have helped to keep the subject looking directly into the camera. Otherwise, I think it’s well-targeted at an audience and it’s an efficient use of being at an event.

Every microphone has this invisible bubble that comes out of it called a polar pattern. Generally, sounds that are recorded inside of this bubble sound good (or on-axis) and outside the bubble they sound bad (off-axis). It’s important to pick the right kind of microphone for what you record, and to place it properly. This all depends on the polar patterns of the microphone. Most cameras have microphones built into the camera, but these aren’t usually the best place to record the sound that might be happening, they are just there to record some audio, which is better than no audio.

Here are a few types of microphones typical for location recording:

Lavalier - also called bug, lav or lapel microphone. These have small polar patterns and are designed to capture vocalizations within half an arms’ length near the person speaking. They typically pin to a shirt or can be hidden just under a shirt. You can get wired or wireless versions.

Boom - This type of microphone is for capturing dialogue or sound effects straight ahead from the source of sound, at a distance of a few arms lengths. Better and more expensive mics tend to record less off-axis sound

Parabolic - You see these at ballgames, they are like a sniper rifle for capturing audio. You probably won’t use these unless you are capturing nature sounds. You can be far away from a subject, and still hear them clearly.

Podium -

Sm57-type - these are usually used on stage, PHOTO OF mic on speaker from wikimania 2014

Phone-mics

XY stereo mics -

If you have a DSLR or a camcorder, there are powered microphones that you can attach to the cold or hot shoe on top of your camera

Cellphones and many camcorders have microphone inputs, use them! There are many companies that make

How to record

When you set up your recording, set your levels. The VU meter (looks like sound volume) will be in green, yellow and red. Red = bad. Avoid red, because the audio will clip, but aim to get it to

You can set it to autolevel, but know that that will boost up quiet sounds and lower loud sounds.

How to record audio:

If someone is speaking into a microphone on stage, try to see if you can record from the mixing board directly to your equipment. If recording an interview with someone, try to record using a lapel/lav/bug microphone, and also any shotgun or onboard microphone that might be on camera. Mic placement - every microphone has an invisible bubble that comes out of it called a polar pattern, inside this bubble dialogue sounds good, outside it’s bad or ‘off-axis’.

If you want something more permanent, you could buy or borrow a shotgun mic for dslr, they mount to the top of the camera and cost roughly $2-300. - not the best for interviews however.

You could also get a zoom h4n, mount it to the top of the camera, and get a cable that sends the headphone output to the camera and to your headphones. That way, you would also have an audio recording device that you could use for audio-interviews, podcasts, etc. it costs roughly $250. It will record stereo audio on top of the camera and has 2 inputs available for any other source (like 2 lav mics for a discussion).

There's another workaround where you get a Rode Smartlav- it's a $90 wired lav mic that uses a cellphone to record. You'll have to sync the audio with the video in post production though and you have no way of monitoring the audio as it records. I don't recommend it as your only source because it's too risky.

Live Music

Music is difficult to record and my guide won’t be comprehensive, so this section will be short.

If you have a situation where there is a person mixing sound that is being amplified live to an audience, you can often ‘patch-in’ to their mixing board or amplifier if you have a cable or a transmitter to do it with. If you have 2 microphone inputs, I’d recommend using a wireless transmitter, and then using your on-board mic too. This way, you are free to walk around the space with the camera, you won’t be tethered to the mixing board.

If you record without a mixing board, or unamplified sound, try to find a stationary place where the music sounds good without too much noise, and I’d recommend keeping the microphone (probably the one you have attached to the camera) there, and don’t move it. You don’t want handling noise.

If I have a wireless lavalier or an external recording device handy, you could mount it anywhere that the music sounds good to you and then I’d leave it alone until the session is finished. Make sure you have your batteries fully charged.

You may or may not have the right to distribute the music you record. Get the contact info of any musicians, recording engineers or venue owners you meet in the process of recording. You will need to clear the music with the copyright holder before you publish anything.

Video

It’s hard (impossible?) to be open-source with video: I don’t know of any camera that actually natively records video into an open source format. This means that in order to share your video onto Wikimedia Commons you HAVE TO CONVERT your original media into webm or ogv, both formats that no professional video editors use when editing, and when you convert, you’ll lose fidelity of the original video - color and other details will be slightly different. You also may find that your audience is on YouKu, YouTube, Vimeo, Facebook, Twitter and you have no need to convert to webm. All this said, I do advise that you convert all videos that you distribute to .webm or .ogv and upload them to commons so that there is an open format (with no paywall stopping anyone from viewing) that can have open-captions added (more on captions in post-production).

Since about ten years ago, there have been an explosion of settings for audio and video equipment that you can choose from. Frame rates, aspect ratios, bit depth, sampling rates and so on are all designed to give you fine control over your video for different purposes. When you pick your settings, it’s important to stick with them for your whole production because when you start editing, you have to choose one playback setting, and conversion from one type of footage to another can mess up your footage in all kinds of ugly ways that you may not be able to fix.

LINK TO BAD CONVERSIONS

Frame rates

Your camera shoots ‘frames-per-second’, which give the viewer the impression that things are moving when they are actually just a sequence of still images. It’s the same principle that makes animation work. I shoot everything in 24p which means that its 24 frames per second. About 100 years ago, the early motion picture industry settled on 24 frames a second as a standard and that standard became the ‘film look’. For Wikimedians, there can be all sorts of frame rates and standards you may want to use. I’d recommend for video that you shoot in 24p because nearly all professional equipment and much consumer equipment can shoot in this framerate and it also converts well to and from many other framerates when editing. This is important because Wikimedia is a remix culture, and it’s important that video for Wikimedia is as easy to remix as possible.

Aspect Ratios

PICK https://en.wikipedia.org/wiki/Display_resolution

Aspect ratio is the shape of the rectangle of your image. Sometimes it’s square (1:1) but usually it’s wide. Aspect ratios are measured in pixels. You can shoot in an array of aspect ratios, but I usually choose HD (1920 x 1080) which is mostly standard today. As screens improve, I expect that 4K (4096 x 2160) will become normal, but most devices Wikimedians will be making video for (cellphone, desktop or laptop computer, social media) probably don’t need movie theater-quality resolution, which is what 4K was designed for. If you shoot in 4K, you’ll also have four times as much hard drive space to buy, and buffering with streaming services might not work as well.

Also view on YouTube. This video was shot by astronauts on the International Space Station using a 4K camera. When you play this video do you have buffering issues when you set it to 1080 versus 4k?

Lighting

• I've used 1/4' foam core boards and wrapped christmas lights around it to get a nice soft light effect in the past. - foam core boards also work to bounce light (reflector).

Good tutorial on lighting here: https://www.youtube.com/watch?v=MLlMl2KuZi0 And the cheap DIY version: https://www.youtube.com/watch?v=zuLaW53GRBk

• clip lamps, tuff spun for light diffusion, black wrap (to control light), clamps

I'd also recommend that if you take a day to make a test setup at home or in your office, it will teach you a lot and give you great confidence when it comes time to produce your video.

Smartphones

These days, smartphones all record good video. There are many options

Filmic pro - This app is incredible. Once you turn off the auto-white balance and auto-focus you can see how powerful and film-like your cellphone really is.

See audio above

Phone attachements

For around $20 USD you can have a whole lens kit for your smartphone. This allows you to capture closeups, wide shots, and fisheye shots.

Cellphone tripods

DSLR / CAMCORDER GUIDE

Camcorders Drone photography

• EXTRA BATTERIES FOR YOUR DSLR (i'd recommend at least 3 batteries total)

Focus on bridge of the nose

Talking heads, interviews

How to do an interview: Put all your quesitons together for the interviewee, and research appropriately. Use a key light, a fill light and a backlight or a hairlight Use a lav/lapel mic on your subject and if you can use a boom mic too. If you plan to hear the questions from the interviewer int he final piece, put a lav mic on that person too. You can shoot with one camera, focused on the bridge of the nose Be mindful of the background, make sure it’s not distracting

Video: Manual

Walk and talk

Video: A/V mode

Voice-overs

Note that during any production phase, you may also want to capture still photos of your subject or the production itself. I mention this because it’s easy to forget.

Scenery or B-Roll

I shot this as a volunteer visiting a museum. It's basically a montage of b-roll of a mill. It's the kind of thing that a news report could show while they talk about this place. I would consider this somewhat incomplete, because it only shows part of a process, not the whole process of what the grist mill does. That said, this video is better than NO video on a Wikipedia article for the mill.

Note - When recording audio or video you should be very careful that you don’t invade anyones privacy. You could easily violate wiretapping or privacy law if you accidentally record someone who didn’t know they were being recorded.

Post-Production

Editors: Generally editors can charge anywhere between $25 and $50 an hour and up. You can do this yourself too.

So now you just wrapped production. Whoot! Don’t mess everything up now. It’s easy to get excited and rush your edit and upload everything right away. Now is when it’s time to shift gears. Unless your production is quite small and consists of just a few shots, you should evaluate everything you have and thoughtfully assemble it.

This video covered a Wiki Arabia conference. To me, it looks like the footage captured was good, but the editing appears to me to be rushed. It’s easy to get excited after shooting footage, but it’s important to ‘shift gears’ and make the best edit that you can with the footage that you have.

Back up all your footage. I can’t stress this enough. Watch your footage. It’s important that you know what you have to work with. Log your footage as you watch it. This is where you take notes and put your shots into buckets - categories for what fits together. You can do this on paper or in your editing software. (VICTOR- BE PHILOSOPHICAL HERE ABOUT EDITING) Assemble your shots in your video editing timeline. Try to fit it all into the story you created in idea development. Instead of words, you now have audio and video to illustrate the ideas you wanted to communicate. See if when you assemble them together if they flow like you imagined they would. If everything flows perfectly then you know that you’ve died and gone to heaven. That never happens. Something breaks. You missed an important shot, some audio turned out bad, the shooter shot in the wrong framerate, you ran out of money, the student shooter you found who said that your shoot was a piece of cake flaked out half way through, etc. (SHOW Wikimaina 2016 video - note the speaker couldnt memorize the script and spoke in the future tense, which required an intertitle at the beginning explaing in that his event happened in the past)

Frame rates - So the thing that makes film and video work is this idea called ‘Persistence of Vision’. Film and video is actually 24 or 25 or 60 STILL IMAGES of FRAMES all recorded and then played back one after another so quickly that your brain doesn’t have time to see each individual frame, so it sees ‘motion’ - your brain invents motion where there isn’t any. So - each video is set to a frame-rate: how many frames per second play on-screen. Most cinema plays at 24 frames per second. TV can be 30 frames per second, or 25 fps and video games play at 60. This means that when you shoot and edit your video, you need to pick what playback rate you want your audience to see. I don’t know of any player that can switch between framerates, so you have to CONVERT any video that you may have to fit the framerate in your editing timeline. This all can get very complicated very quickly. A good guide can be found here: (FIND GUIDE)

Syncing dialogue / location audio

Bumpers: (provide bumpers and legal disclaimer language) Here is a logo/bumper that Wikimedia uses at the beginning and end of some of it’s videos:

Music

Music is interesting, it can keep your attention or turn people off. Try to find music that doesn’t compete with any dialogue in your video.

The WMF has music that it is the copyright holder of. Wha!?!? copyright?!? -- Yes, copyright. This is because in order for us to be able to share anything on modern social media like Facebook, WMF needs to be the copyright holder of the work or the work has to be in the public domain. The music will be published cc-by-sa 3.0 onto commons as soon as it is possible, so that anyone can use it as well.

I’ve had music composed for specific wikimedia videos, and had music composed to be used for more general-purpose uses. Andy R jordan (current and knowledge for everyone)

Dub Terminator

Luis Aguliera https://commons.wikimedia.org/wiki/File:MOS-ART_-_Luis_Gabriel_Aguilera.wav https://commons.wikimedia.org/wiki/File:Prep_for_Joy_-_Luis_Gabriel_Aguilera.wav https://commons.wikimedia.org/wiki/File:Moving_-_Luis_Gabriel_Aguilera.wav https://commons.wikimedia.org/wiki/File:You_Are_Here_-_Luis_Gabriel_Aguilera.wav https://commons.wikimedia.org/wiki/File:Seh_leh_nah_-_Luis_Gabriel_Aguilera.wav

Classical - all music from edit 2014&15, other classical tracks Jamendo Freesound Archive.org FMA Soundcloud.com

https://soundcloud.com/anoophariharan/lightning-creative-commons

Color correction

If you can spend some time on color correction, do it. I wouldnt recommend it for anything that’s strictly documentary, but Usually what comes out of the camera is good enough

Facial smoothing - digital anarchy

SOFTWARE:

Kden live Blender https://opensource.com/life/15/1/current-state-linux-video-editing openshot ADD archive.org editor

Conversion:

• Firefogg works well to convert video to .webm http://firefogg.org/ (you need to use the firefox browser) Guide for HD/ 4k video? settings?

• Miro works, but seems to be designed to make video as compressed as possible, so you lose the quality of audio and video. It does allow for a conversion queue.

Screen capturing

• Screenflow

• online video editors (youtube)

Titles, fonts Show free fonts

Captions and subtitles

• The titles and subtitles in this video are burned-into the video, and cannot be removed. You can export a cut without titles, to allow Wikimedians to make their own titles.
• Version without titles, designed for YOU to add titles in YOUR language using video editing software (instructions in file description)
• forked Japanese version
• forked Chinese version
• forked Polish version

Captions and subtitles are two different things that have similar purposes. Captions are metadata that are added to a video and can be turned on or off by the user. Subtitles are part of the video itself, they are burned into the video and cannot be removed. Use Amara.org to make your captions. What’s important is to create GOOD timecoded captions, because anyone can fork that timecode into another language, and it can be a mess to try to clean it up if the timecode changes for each language version. I’ve also used 3playmedia.com to create transcripts and captions. They charge by the minute. The captions on Wikimedia Commons are superior to every other platform that I know about because they are open to edit and have fantas tic language support. Also if you use double brackets on a topic in the timecode you create a hyperlink. It’s a little hard to read but it’s a neat tool. Translations (show process of copying captions with code)

Distribution

Add advertising -

COPYRIGHTS & social media So assuming that social media is how you’ll be sharing your video, you have to respect the licensing. Most social media outlets have a clause in their terms and conditions that state that you assign them joint copyright to the media that you upload to their service. So that means that (and this is crazy) you can’t legally upload creative-commons content that you don’t have the copyright to to social media. You can use works that you create, works that others have given you license to, and public domain or cc0 material (fortunately there is a ton of public domain material on Wikimedia Commons LINK).

• Code to embed in media wiki

• Links to rules for video on wikipedia

Final thoughts

Final thoughts

I dream of an app that you can shoot video on and upload that video directly to Wikimedia Commons. I also dream of a way that I could browse through b-roll and footage treatments that our community has written collaboratively, and ranked in order of importance to be produced. I’d love to be able to record bird sounds or musicians on my phone and upload them to commons directly. But right now we have to work with what we have, and if you start with an idea, stay patient and determined, you can produce the video you imagine.