Sheep and Sisyphus – The Toils of a Data Wrangler

As we loaded the van in Bournemouth ready to head off to Chalke Valley, the guy behind the kit desk gave a knowing chuckle when I told him I was to be data wrangler at the festival this year. ‘It’s a fancy name for the poor sod who sits in a dark room transferring files’. Blinded by my hubris, I laughed him off, oblivious to the truth of his words.

It wasn’t entirely true of course, there was no dark room. Just a very hot tent. In a field. There were sheep.

The major challenge I faced was storing the footage in a way that allowed me to later find specific files which needed to be passed on to our editors. This was crucial because over the course of the week I ingested over 10,000 files into the system. The recipient of these files was a hulking external hard drive sat next to me, which I affectionately christened Tobias.

Tobias and the backup hard drive, his son Toby Jr.

Anyway, the system I developed revolved around the file names – I put enough information in the names of the files I was transferring that I could easily find them again. The format I used was day_project_equipment_name and it worked really well. Using equipment as a variable rather than merely camera/sound allowed for the fact that people would often use different cameras filming the same project. Adding their name at the end further helped avoid confusion as to who filmed what.

I set out a portion of the table where people could fill out the relevant information on sticky notes, which they could then attach to their SD Cards and leave for me to ingest. Another feature which saved my life was the ability to mass-rename files (Press F2 with the files selected), saving valuable time in the transfers.

If this is sounding a little boring to you, it’s because it was. Once I had the system figured out it was just a matter of clicking and dragging files across from one folder to another.

What made it interesting is that in amongst this constant stream of transfers, I also had to review all the footage coming in and cut the best of it together into daily roundups. Each day around 5pm the pressure set in to meet the 8 o’clock deadline and I would enter an almost zen-like focus, shaped by the stresses and pressure of the task at hand. Anyone foolish enough to disturb me was met with primeval grunts and a look I didn’t know my face was capable of making. If anyone from the team is reading this I can only apologise.

BUT, I got them done, in the later half of the week helped by the brilliant Naomi. Each day, when the dust had finally settled as I hit that export button, I would look over and see the pile of SD Cards waiting for me to ingest. My very own version of Sysiphus’s toil.

*sad violin music*

 

Chalke Valley 2018

After sleeping on something other than the stony ground of a field (which is a convoluted way of saying a bed [which is a needlessly long explanation for it {sorry}])  I finally feel human enough to write about my week at Chalke Valley History Festival 2018.

And what a week it was! Having taken on a more senior role this year as Assistant Project Manager/Data Wrangler, the experience was very different to that of last year. I’ll go into more details in a future post about my role,  but overall it was a challenge which I learnt a great deal from.

Overall though it was a fantastic week. The team this year was half the size (16 compared with 30) but were all really driven and keen to learn. We managed to put out as as much if not more content than last year and still at the quality Bournemouth Uni is renowned for and it was amazing seeing how quickly their skills developed over the week. It was also good getting to know them all as I’ll be joining them in September to complete my final year at Bournemouth.

I made a moth-friend

One highlight was seeing Bryan Begg’s wife Diana and close friend Paul again to go to his memorial talk along with James and Georgia who were my production managers last year. Bryan was a long-time patron of the festival and at CVHF 2017 I edited the last interview he ever did before passing away earlier this year. Paul orchestrated a surprise of flowers for Diana which we gave her and it was a special moment.

Another was going to see Rob Wilkins give a talk about Terry Pratchett. I owe my love of reading to Terry who has been my favourite author from a young age and it was touching to hear such personal accounts of his life from Rob. It was amazing to learn that Terry lived in Broad Chalke, just a few miles from the festival site.

It’s been a whirlwind week with so many memories, hopefully I’ll be able to make it to CVHF 2019 and make even more!

 

The Transcribinator

Transcribing is the kind of frustratingly labour-intensive task that makes me realise humanity still has a ways to go before being at risk of being enslaved by sentient AI overlords.

However, I’ve come to realise that is a necessary part of the editing process. For those of you who are unaware, transcribing is the process of converting speech into text. When faced with large chunks of footage, such as after shooting a series of interviews like I recently have, figuring out how to go about editing it all can be a daunting task.

Transcribing helps because it allows you to edit before you even import your footage into your timeline. Known as a paper edit, having the words written out in front of you makes it much easier to figure out how you’re going to structure your interview. This is because it’s faster to read two sentences next to each other than having to scrub back and forth in the timeline to hear if your cuts work.

On top of this, if you have the time and patience to do the transcribing yourself, the words will be seared into your brain via repetition-based monotony. This might not seem like too much of an advantage until you’re actually editing and your nice clear text become ambiguous clips in the timeline. Trust me, it helps.

Transcribing is a fairly time-consuming process (please see previous comment about lack of robot overlords). So even though there are a few tips and tricks to speed up the process, be aware of this when planning your project. The biggest tip I can give you to help reduce the workload is to run your interviews through a speech-to-text converter. This allows you to go through and make corrections rather than writing it out from scratch; the better your audio quality, the less changes you have to make.

There are loads out there, but my software of choice is VoiceBase. Although you have to pay for it, you get $60 free credit which I’ve only used about $10 of in the year I’ve been using it.

So, to recap, transcribing is a painstaking but necessary part of editing which, when done properly, will end up saving you time in the long run.

 

It’s All About Tone

Colour correction is a skill which I’ve been slowly getting to grips with. My biggest challenge yet has been matching the footage from my Blackmagic Pocket Cinema Camera (BMC) with my Canon camera. This is because the two cameras capture information in very different ways.

The Canon records in a .MOV file format, with H.264 encoding. Don’t worry if you don’t what that means, what’s important to know is that it allows the camera to record HD video with a relatively small file size. This comes at the cost of being less flexible when it comes to colour-correction in post-production. A smaller file size means less information captured, which means you can’t manipulate the colours as much before it affects the image quality.

The BMC uses ProRes. This captures much more information and therefore creates a much larger file size, so when it comes to colour correction you have a lot more range to work with.

Think of it like taking notes in a lecture. The Canon notes down the key points, so that by the end you’ve got a general but clear idea of what the topic was about. The BMC covers all the points of the lecture in detail, with cute little subheading’s and the occasional highlight.

More notes = More paper                   SO                More information = Larger file size

What this means visually is that the Canon footage looks much closer to a finished product, while the BMC is a blank slate, devoid of any striking tone or colour.

With the help of a course on the amazing Lynda.com (please sponsor me), I began the process of matching the BMC’s footage to the Canon’s. The first step was to set up a workflow where the two could be compared side by side, which I did by cropping both shots and placing them next to each other, making sure they had similar framing of the subject (which in this case is the dashing Jonathan).

Next was to match the tone, which is shown by the Luma waveforms below. As you can see, the waveform on the right (which is the BMC) is more compressed and a different shape to the one on the right.

To match the tones, you use the Basic Correction settings in Premiere to match the shape of  the BMC’s waveform with the Canon’s as shown below. The BMC footage on the far right now has more depth to it than it did before.

After matching the tones of the two cameras, I then moved on to colour. This followed a similar process of matching the waveforms, this time using a RGB Parade:

As you can see, the three pairs of waveform are similar in size and shape. This is done using the curves you see on the right hand side of Premiere (the green line in this screen grab). The two are now much more closely matched:

Jeff Sengstack, the author of the Lynda course I watched, said that there was both an art and a science to colour correction, with tone being the science and colour being the art. Understanding this was key to my understanding of the skill – it was easy to match the tones of the two shots, but matching the colours was a much more subtle and subjective process.

 

Interviews Done!

So, two weeks and 17 interviews later, I’ve finally finished interviewing the staff at NEJ.  Some people dropped out or didn’t sign up to begin with,  which was something I hadn’t had real experience with before – people are nervous about getting in front of a camera. Having always done university projects with other media students who were pretty used to it, I underestimated The Fear of the Lens (cheesy horror movie anyone?).

This also affected people who had done the interviews – it was a challenge to get them feeling comfortable enough to give natural answers. There was a really magic moment when one of the guys, who was extremely nervous about being on camera, just for a moment forgot that he was and delivered one of the best lines of the week. That was a great feeling.

You’re very aware that these people are placing trust in you to show them in a good light, overcoming their own fears about being in front of the camera. It’s a humbling responsibility.

My ingest and file management has also greatly improved. Having never handled content on this scale before, I made sure to set out folders and a workflow for getting the footage ingested into Premiere.

The scale of this also meant I’ve had to transcribe all the interviews and do a paper edit prior to actual editing, a skill which I gained from my time with Atlas of the Future doing Meaning Conference transcribings. Quickly seeing the benefits of being able to go through and highlight the best parts and more easily visualise how they can fit together. It is melting my brain though.

And because I love a gif, here’s some of me setting up and then packing up in the workshop:

 

Foiled by a spindle moulder

Sooo, some stuff went wrong and I had to change my plans.

I spent the whole of yesterday setting up for the workshop interviews, only to find out the machine right next to my set would be being used until Tuesday. How I only found this out after hours of prep I don’t know, but its been an excruciating lesson in Sod’s law – ‘If it can go wrong, it will’.

My contingency was to interview the office employees, which meant finding a new location and all the challenges that posed.

I’m going to say this now, and until experiencing it myself I wasn’t able to fully appreciate it:

SOLO SHOOTS ARE REALLY HARD

Like the animals on Noah’s ark, my kit came in pairs. Having to set up a  camera, mic and light by yourself is not ideal at the best of times, but when each is multiplied by two it pushes you to the limits of your skill set. Nevertheless, I’m really grateful for the opportunity it has provided me to push myself. It’s a challenge having your motivation coming from the same source as your frustrations and problems, the overcoming of which is invaluable in all aspects of life.

One of my favourite tasks of the day was having the sound-proof the cupboard behind my camera which, after getting the guys in the workshop to turn their music down, I discovered housed a large and rather noisy server.

To get an idea of the noise prior to my handywork, hum moderately loudly until someone nearby smacks you.

Eventually I managed to get Ed, one of the managers from upstairs, to come down and be the guinea pig for my first (ever) solo interview. It was a really good run, by which I mean I found out a lot of the problems I hadn’t known about before I hit record.

Watching the footage back, I’m really pleased with how it turned out. My next challenge is going to be matching up the colours of the Canon and Blackmagic footage (which I *think* involves something to do with LUT).

Watch this space!

Workshop interviews

Having found a good location in the workshop, today I did a test setup to figure out any issues before the actual run. The first, I discovered, is that its quite very difficult to frame an interview shot without a person sitting in the chair. Not wanting to disturb the employees, who all looked very busy and were often holding something sharp, I was forced to run back and forth from the camera to the chair to make adjustments. I got there in the end and the resulting GIF which you see before you outlines my process of tightening up the shot, adding props to fill out the background and bringing in some lighting. (I apologise for looking so bleak, probably just sad that all the bees are dying)

This will change slightly from person to person as I adjust the lighting and focal length to accentuate their features, but it’s been really useful to get a feel for the space. It’s also allowed me to block out tripod and lighting positions like so:

Here’s a pic of the setup minus your’s truly. Hoping to start the interviews tomorrow, so we’ll see how it goes!

The silence of the tools

So now I’ve been with N.E.J. Stevenson Ltd a few weeks, I’m ready to start prepping for interviews. Most of the work I’ve done prior to this has been action based – filming around the workshop and an installation in London.

I wanted to hold off conducting interviews with the employees because I feel it’s important to establish a rapport, allowing the interviewees to feel more comfortable with me behind the camera and so able to give more natural answers.

I also wanted to wait a few weeks because I was still getting to grips with my kit (tools don’t tend to laugh at you when you mess up, unlike their human counterparts) and the complexities of solo- shooting.

Last thing before I start is to consult the fountain of knowledge that is Lynda.com on how to conduct professional interviews. A good tip I picked up from the course is a simple list of things to tell your interviewee before turning on the camera:

  1. Don’t look into the camera, this is just a normal conversation between the two of us.
  2. Include the question in your answer using full sentences.
  3. It’s okay to restart questions or rephrase your answer.
  4. Can I get you anything before we start?

If you’re interested in learning more about interview techniques, you can view the Lynda course I’m using here.

Artisan GIF’s

I have spent the last 2 days making a GIF. That’s about 16 hours making 16 second’s of what is essentially a moving image. I’ve been staring at the timeline in Photoshop for so long the images are ingrained in my eyeballs. It’s probably not been the most efficient way to make this GIF, but with so many components it was the only way I knew how.

Yet I don’t feel bitter towards it, instead feeling strangely proud of my labours. Others will probably glance at it and then continue their scrolling, but I’ll remember the hours of staring at that flashing sequence of images and the subsequent kind of mild epilepsy I inflicted upon myself to get it done.

It makes me wonder how many GIF’s I’ve scrolled past, oblivious to the blood, sweat and tears which someone poured into those few seconds of content. It’s a humbling realisation that there is immense craftsmanship in even the simplest of things all around us. I shall try to be more aware of it, as should you. The world might seem a little bit brighter.

Compressed Cats

So whilst watching a course on DSLR lenses on Lynda.com like the studious media producer I (occasionally) am, I learnt something interesting about focal lengths and their relationship to image depth. The difference between taking a photo whilst standing physically close to an object and taking that same photo from a distance but zoomed in is remarkable.

Standing close, your focal length is smaller (or shorter? Forgive me if my technical jargon isn’t 100%) and so there is a greater sense of depth, allowing you capture a larger amount of the background. Standing further away and zoomed in, your focal length is larger/longer and so the image becomes compressed, making objects in the background appear closer.

To illustrate this, I enlisted the help of a small furry animal called Millie, who happens to live in my house:

CLOSE UP
ZOOMED IN

 

 

 

 

 

As you can see from this effortlessly photogenic feline, close up the photo captures more of the background, whilst zoomed in the image looks more compressed with the background much closer. Personally I prefer the first photo, which accentuates the shape of her head, whereas the second gives her a bit of a Garfield aesthetic.

You can clearly see, however, that even though the photos are framed in a virtually identical way, their overall look is noticeably different. This can work the other way around as well. I took two photo’s of Millie’s brother, Ollie, again close-up and then further away:

CLOSE UP
ZOOMED IN

 

 

 

 

 

 

 

 

This time the zoomed in picture looks better, that compression helping to clearly define his facial features. Granted, the first picture is from a slightly higher angle, but you get the point. This serves to show that there isn’t a set focal length which is best for everything, and that experimentation with different focal lengths can drastically affect the quality of your image.

Because I’m feeling particularly zealous today I’ve also included the comparison in GIF form. Anyone seen discussing the pronunciation in the comments will be blocked.