Adobe had a long lead time, talking about AIR. There was some talk of “a universal runtime” when Adobe’s acquisition of Macromedia was announced back in April 2005… a first Apollo public build arrived in March 2007… it wasn’t finalized into a usable 1.0 release until February 2008.
That’s about two-and-a-half years between announcement and delivery… a long time in this case, because “universal runtime” and “Apollo” and “AIR” was a new type of application, an old way of developing but a new way of delivering. Took a lot of conversation along the way to make sure we all understood this new thing. But Adobe was betting the company on this new publishing platform, and such a level of commitment requires a corresponding level of care. Adobe took 2.5 years between the first talk of AIR and its 1.0 delivery.
I’ve been hearing a lot lately about “desktop applets” and “Acme’s answer to Flash” and so on. There isn’t so much a need to talk about it in advance these days… lots of people already understand photo-editing in your current browser, or when it’s more useful to have a desktop app than browser. It’s real already. You don’t need a long lead time to prepare people anymore.
For something new, it’s good to talk about it beforehand, sound out people on the new ideas, reality-check the concepts. But if there’s an existing realworld model, like Flash or AIR, then such a long lead time isn’t necessary, and it’s usually better to just deliver.
An actual new choice can be useful, but a new set of speculative debates, not so much. Better to ship.