Savior of the Universe

Following a link from Michael Tsai (who’s been my go-to blogger recently on the OS X sandboxing controversy), I read this post by Jeff LaMarche on Adobe’s announcement of the end of mobile Flash. The key passage is this:

I missed a huge factor in the demise of Flash. I assumed the performance issues they were having back then were simply technical hurdles that would be overcome by Adobe’s engineers before long. In the end, the lack of a monoculture was certainly a significant factor in the demise of mobile Flash, but the real nail in the coffin was that Adobe never even got mobile Flash working demonstrably well on a single model on a single platform, let alone working well on the “billions of mobile phones” they were shooting for with the Flash Consortium. I completely overestimated Adobe’s ability to deliver, technically.

Michael then comments “It would be interesting to know why they weren’t able to deliver.”

I think I know part of the answer.

From January of 2005 until December of 2010, my home computer was a 12″ iBook G4. In the early years, I spent many pleasant hours on YouTube, looking at old music videos and concert movies. The videos ran smoothly.

From about 2008 on, though, YouTube was worthless to me. It started with stuttering video playback as the audio went on normally. Then both audio and video were jumpy. Finally, it was rare that even jumpy playback was possible—even with videos I know I’d watched just a couple of years earlier.

Think about that: a video I’d watched comfortably in 2005 was stuttery and unwatchable in 2008 on the same hardware.

(There are two obvious jokes here: “How did I get here?” and [Not the] same as it ever was.” I can’t choose which to go with, and it’s probably a mistake to use a Talking Heads lyric in a post with a Queen-derived title anyway, so I’m just going to let it pass.)

Adobe was riding the Moore’s Law train, putting out less efficient (albeit probably more general) code with each release and trusting the ever-more-capable chips inside our computers to more than make up for the inefficiencies. Mobile derailed that strategy.

I’m no hardware expert, but I’d guess that smartphone processors are at least half a decade behind personal computer processors. So even if a smartphone is “equipped” with Flash, the user experience is going to be much like my experience on my iBook G4—no experience at all.

Adobe’s decision to stop mobile Flash development is a recognition that it can’t turn the clock back. It’s let too many—wait for it—days go by.