Ignorance and Confidence

Ignorance and Confidence

Kevin Murray September 26, 2017

Hi! I’m Kevin Murray, the principal engineer for the Core development team. I’ve been with Wazee Digital for over 13 years, since it was a wee little start-up that didn’t know exactly what it wanted to be when it grew up.

When I started with the company, our main focus was a digital asset management system for bulk-marketing postcards. Yes, postcards. You know, the ones you get in the mail with a picture of the Hamburglar trying to break into a house, advertising a home security system? The ones you just drop in the recycling bin on the way back into the house? Yeah, those. (Those of us of older than 30 will remember.) Bet you never really thought about where those postcards came from, did you?

Lots of talented people around the country create those little suckers and sell them to other people; those people then market them to local businesses and mail them to you. The part we cared about was the part between “lots of talented people around the country create” and “sell them to other people.” Somebody had to figure out how to effectively import, preview, display, search, and deliver thousands of QuarkXPress and Adobe InDesign documents … we were those somebodies.

Back then, we’d get shoeboxes full of CDs every month (literally shoeboxes, usually Nike), and then some poor schmuck (me) would spend dozens of hours just feeding hundreds of CDs into an Apple PowerBook, opening documents and generating thumbnails – by hand. We’d then import those documents and thumbnails into a database. During that time, I was slowly learning to fear the sight of a Nike shoebox: My wife would just bring me shoes already de-boxed so I wouldn’t break down at home. Just walking past the display window of a Famous Footwear would trigger cold sweats.

After the documents were imported into a database, someone would then type in keywords and descriptions for each postcard. I also did this job for a while, and by the time I had to type “burglar, thief, night, security, window, alarm, safety, peace of mind” for the 100th time, I was starting to welcome “shoebox day” and really beginning to fear for my sanity.

Once we tagged the postcards with keywords, users from marketing companies could then find the ones that matched their clients’ services. When they found what they were looking for, they would purchase the license to the postcard design by talking to a human-person on the phone, and then another human-person would email them the documents or mail them a physical CD containing the documents.

Over time, to preserve what was left of my hair, we started to automate as much as we could. Eventually, we built a system that could import documents from the CD, automatically open them and produce thumbnails, and ingest the documents into a database that was used to generate workflow pages. Those pages could track which postcards had been key-worded.

Postcards that were fully vetted could then be searched, browsed, and previewed online. Authorized clients could log in and find postcards they were interested in, then purchase and download them. At the time, purchasing was still done manually over the phone because we simply didn’t feel like we could ensure the security of an online transaction. This point just goes to show that security of our clients’ information has always been paramount. If we couldn’t guarantee security online, then we just didn’t do it that way. (Eventually, of course, we did develop an e-commerce package that we had confidence in, but that wasn’t until later.)

Around the time we started to get the postcard operation running smoothly, we started looking at the big picture and realized that there probably wasn’t much of a future in charging $2.40 to host a postcard. The market wasn’t very big. To be honest, it was pretty small and fading quickly. I made a point above about getting those postcards in the mail, and I’ll bet that some of you 20-somethings thought, “What’s this guy talking about? I’ve never seen one of those!” Yeah, I haven’t seen one in a few years either, but trust me, in the 80s and 90s, they used to arrive in cellophane-wrapped bundles by the dozens, depending on where you lived.

We looked briefly at hosting images, but we suspected video might be the more progressive path. Then our previous CEO traveled to Japan and came back with pictures of vending machines with video screens, and elevators with video screens, and gas pumps with video screens, and suddenly it seemed so obvious: Video was everywhere, but finding and purchasing video wasn’t easy. In many cases it meant sending an email to a researcher, who would send back a text listing of possible videos. You would select the interesting ones from the list, and two weeks later you would get a VHS tape with low-res versions of them. You could then make a final selection and buy it, and in another two weeks you would get a final copy on VHS, ¾-inch tape, DVD, or external hard drive (depending on how big it was and what quality you needed).


At the time, online stock photos were everywhere, but video was pretty scarce, so we looked around and said, “Waitaminute! That’s a media importer. We have automatic preview generation. We have a system for key-wording and metadata management. We have a team of key-worders. We can do document workflow. We know a bit about licensing. We have an online presence with search, browse, and preview capability. And we can digitally deliver documents! We’re 95 percent of the way to having a fully automated, commercial video asset management site!” That was an exciting time at Wazee Digital.

All you need in this life is ignorance and confidence, and then success is sure. - Mark Twain

Less exciting were the next couple of years, when we realized just how naïve we were, and that we were really only 20 percent there. We still had to get systems that specialized in high-speed video transcoding; we had to purchase and install a petabyte storage infrastructure (who would have guessed that video is 10,000 times bigger than a postcard?); we had to learn a lot about video rights and licensing; etc. Over the next couple of years, we ended up rewriting most of the systems to specialize in video, and we hired a lot of excellent people who knew more about video than we ever even knew existed, but it was the best decision we ever made.

We didn’t know the video world, so none of us knew what not to do. We designed everything as if our supplier was a wedding videographer with extra b-roll, and our client was a church-sound guy who just wanted something pretty on the wall. In other words, we didn’t have years of experience in video, so we assumed that nobody else did either. We tracked white models and chroma format, whether something was anamorphic or bottom-field interlaced, but we didn’t want our customers to have to know about all those things just to license some video.

We were the first online site to have roll-over playback of video in a search results page, which meant we had the fastest online preview there was. We were the first to offer “Ready Now” downloading of watermarked comps for review and mockup purposes. We were the first to partner with nonlinear editing systems so we could effortlessly replace low-resolution comps with full-resolution masters once a purchase had been made. We replaced the six- to eight-week video-purchasing cycle with something that takes less time than getting a pizza delivered.

There have been several years since then, and as you can imagine, we’ve seen all sorts of new stuff: Our typical video hasn’t been 10-second stock footage in years. We’ve seen everything from 6 FPS content shot in the 1800s to breathtaking 4K. We’ve added audio and image support because, after doing video, audio and images are easy (I guess the naïve beliefs haven’t changed). We’ve hired videographers to create custom content just for our site. We were part of the rise of the cloud and continue to leverage all the new technology we can. Our clients are some of the coolest household names in the world, and we’ve partnered with some pretty awesome companies over the years to do things like driving live-event video management for international sporting events. And we’ve hired even awesomer people to do it all.

Thirteen years doesn’t always seem like a long time. But when you start with postcards and end up with hundreds of hours of video immediately available during a live event that can be shared – globally – within seconds … thirteen years seems like a lifetime. I cannot wait to see what we can do by 2030!


Freemantle