There is a YouTube application on Apple’s new phone, and getting it there required Google to re-encode its entire video library into a different format.
Sit with that for a moment. Millions of clips, transcoded, at Google’s expense, on Google’s hardware, because one handset would not run the plugin everything else uses.
Nearly all web video today arrives inside Flash, and Flash won that position by being installed everywhere before video was even the question. The phone Apple shipped in June has a serious browser and no Flash at all, and Apple has given no indication it intends to permit one. So Google had a choice between being absent from the device or doing the work, and it did the work.
What makes this remarkable is the scale mismatch. This is a device with a fraction of a percent of the handsets in the world, sold in one country, on one network, at a price most people will not pay. And the largest video library on earth reorganised its encoding pipeline for it.
You do not do that for the current install base. You do it because you have looked at where browsing is going and concluded that being absent from that class of device is not survivable, even if the class is currently tiny.
When Google paid a billion and a half for a site with no revenue, the defence was that it was buying an audience, and audiences have to be reachable wherever they end up. That logic holds here. It is also expensive in a way that keeps compounding, because the cost of serving video scales with success, and now there are two encodings of everything to store.
The part I find genuinely interesting is what it reveals about leverage. Apple did not negotiate for this. It shipped a browser without a plugin and let the consequences propagate outward, and the largest video service in the world adjusted.
If that pattern repeats, if enough people reach the web through devices whose owners decide what runs on them, then the plugin that carries the web’s video is standing on ground that belongs to somebody else. Adobe currently has no answer to this beyond hoping to be invited, and hoping is not a strategy when the invitation is issued by a company with no reason to send one.