Like a Wal-Mart shirt with hand-sewn seams
Quick observation. I was tinkering with video sites, trying to figure out how best to pipe videos from my iPhone over to robinsloan.com, and was struck, again, by the surprise of YouTube.
Usually, we expect quantity to compete with quality. You know, like: YouTube’s all about volume; Vimeo’s all about quality. That’s the breakdown that we expect. Cheap and mass-produced vs. high-end and artisanal.
Except that YouTube is the quality leader, too. They’ll host your HD videos for free and play ‘em back as many times as you want, embedded or otherwise. They’re now bumping things up to 1080p. They encode videos in a flash. And it’s precisely because YouTube is so gigantic—because Google is so gigantic—that they’re able to do all this. Quantity is a prerequisite for quality.
Now, I’m just talking about technical quality, of course, and Vimeo has done a great job cultivating quality of a different sort. It’s full of music, art and wonderful stuff like this. Really, the “features” that Vimeo offers are social and psychological, not technical. That’s the right move, because they can’t keep up with YouTube on video quality. No one can.
Anyway, this isn’t a huge insight—just another example of the weird physics of digital media.
Finally, FYI, here’s a tip for video uploads from the iPhone 3GS: Don’t send the video directly from the Camera app. It compresses it severely, and there’s no way to tell it not to. Instead, copy the video from the Camera app, then open Mail, create a new message (to your YouTube upload-via-email address) and paste the video in. Voila. No compression. Of course, the video takes commensurately longer to upload, but it’s well worth it. The same trick works for photos.
Cheap and mass-produced vs. high-end and artisanal.
Wait, do we really expect that with anything but food, haute-couture and the occasional carved wooden toy? In terms of technical quality, I mean? With small number manufacturing justifying the set-up for customized parts and processes is much harder, so you’re much more likely to jerry-rig something from whatever is available. In cooking this is usually a good thing (since the ingredients were once alive and are much harder to control, trying to control them too much often causes bad things to happen, and there’s an innate flexibility in how the parts interact), but in manufacturing anything that has moving, interacting parts (i.e. it works or it doesn’t work) it’s usually a bad thing. Factory clothing usually lasts longer than equivalently complex handmade clothing in the everyday use regime; software that’s had enough of a user base has more features and justifies more QA and testing. I would argue that the edge case of a truly unique object which is labored over in a very particular way (a truly custom app, the mother of the bride making the wedding gown, a piece of art) is a) a different kind of beast entirely (one-off existing in a separate continuum from few to many) and b) usually the beneficiary of a large quantity of practice.
Both Google and IAC’s revenue incentives are mostly eye-ball based; the amount of cash they can bring to bear on the problem is essentially a function of the amount of advertising they get from solving it. Not to be a downer or anything, I’m just a little skeptical that this “weird physics” is really particularly different for digital media, once you translate the problem to the standard incentives basis of analysis.
I guess I’m responding partially to how Vimeo and others position themselves. The message is: “Go to YouTube if you want to be one of the fuzzy billions; come here if you want the most beautiful video, if you want real attention to detail.”
And actually, re software, I think there are a lot of people who would make an argument for “artisanal” software. An app like Scrivener or Tweetie — “hand-crafted!” — vs. an app like Microsoft Word.
Re clothes, I don’t think the distinction is between Wal-Mart cargo pants and home-made dungarees, but rather between Wal-Mart cargo pants and fancy pants from, say, Bonobos. Or between an H&M shirt and a Thomas Pink shirt. Or between Reeboks and Cole-Haans. No question: the higher-end stuff is higher-quality, and it lasts longer.
One more pitch: look at the Bold Italic vs. Demand Media. Both more or less in the same business. The former is small-scale, with hand-crafted articles and an emphasis on inventive design. The latter is huge—the single largest uploader to YouTube—with flat, churned-out, templated content and an emphasis on volume.
What if, somehow, Demand Media figured out how to harness its incredible scale to actually make all of its content better-designed—to make it absolutely beautiful? What if they beat the Bold Italic both on quantity and on quality? It would, I think, be a little bit shocking.
You know, I think here’s where you lost me (and have now re-found me)—a) I had forgotten that Vimeo has a pro-user paid pacage and b) because, at first glance YouTube & Vimeo have the same price (free = $0), I was trying to think of examples where price was fixed and scale was varying. If you allow price to vary as well, well then, obviously higher end usually means better quality. Given that I have $120 to spend on cargo pants, however, am I better off going to a mass producer or paying a tailor (even one based in the same country that Bonobos has its manufacturing in) to make them for me? (Let’s ignore the cost of travel.) If my pants need to have special features or an unusual fit, I’m probably better off getting them tailored, but otherwise–probably better off sticking to the optimized and tested factory stitching. (I say this as someone who has plenty of experience getting things custom tailored.) The higher the price, the more you stand to gain from customizing despite losing scale’s production-data, because the more you can purchase the time and experimentation and life-experience of the crafter. I also thought you were specifically ignoring beauty and style and going after features, functionality and durability.
Okay, reading and rereading your posts, I guess here’s the way I see it. There are several variables: price, custom-feature-#, scale-of-production (these are independent variables); durability/functional-quality; aesthetic quality (these are dependent variables). And what I actually think is that none of the dependent functions will have monotonic or even unimodular behavior along any axis for any kind of product, but that the various zeniths and nadirs will be industry specific, and reflect the underlying sociology specific to the kind of organization that develops when operating at that production regime for that industry of production. By guessing against any monotonic trends, I’m guessing for non-invertible functions–i.e. I am skeptical that any industry has any parameter the increase of which indefinitely benefits quality; by guessing against any unimodular trends, I’m guessing for multiple sweet spots: any industry should always be able to duplicate success in multiple combinations of circumstance.
Basically I’m guessing that since production is innately discreet (objects, people), it’s going to have many separate sweet spots, much like an atom has many separate energy levels, with lots of empty gaps in between.
And I’m going to keep saying that food is exempt from this; it just feels different than the rest of it.
You know what I’ve always wondered is what happens if Google either a) gets a designer and user experience agency to make YouTube look much more lux/aesthetic or b) if they buy Vimeo. If they buy Vimeo, that’s it — the monopoly of Google on internet video would be assured!