News from Adobe
The presentation was around two hours long and a number of topics were covered. The things I felt were important were:
- The Automation section of the SDK will be donated to Apache as previously promised. There had been concerns that this wouldn’t happen due to legal issues. The caveat is that going through all of the automation classes and clearing them with legal folk in Adobe will take time. So we are not to hold our breath on this one. No time scale was given.
One thing that really disapointed me was the odd explanation Adobe gave for why they aren’t working on the MXML part of the Falcon compiler. Apparently, the feedback they got from the community was that the latter wanted to do the work. This is partially true, but it misses an important point: the community wanted to start work on the MXML part now, not in a year to 18 months time. I couldn’t decide if Adobe had simply missed the point or were being somewhat disingenuous over this matter.
Either way, it confirms my view that the community should turn its back on the Falcon compiler as the time scales before we can start doing worthwhile work with it are just too long. I feel therefore that I made the right decision in pushing for the community to develop their own Goshawk compiler instead.
- Adobe seem unsure of how they and the community will develop the Falcon compiler. They plan to use it in future versions of Flash Pro. When asked how they would handle the fact that they didn’t own the compiler when it came to wanting to make updates, they seemed confused by the question. They appeared to assume that any changes they would want to make would automatically be accepted by the community. Further, when asked how they planned to handle the community adding language features to the compiler and thus changing the AS3 specification, they seemed even more confused. They appeared to genuinely believe that no such change would occur. They could be in for a nasty shock in future therefore.
- I couldn’t work out whether there would be a Flash Builder 5. The slides talked only of 4.x updates, yet there was talk of a major update appearing at some time. There still appear to be no plans for 64bit support within Flash Builder sadly.
- It looks like Adobe will be releasing their Future of Flash white paper on Wednesday. One thing that it will likely contain are details of when worker threads will be coming to the flash player. The other is apparently some radical and positive news on the future of the Flash Player on Linux. My wild guess at this point is that they’ll be announcing that a third party will be taking over its maintenance. Canonical maybe? As I say though, this is just a wild guess on my part and I could be wide of the mark.
Peter Elst briefly presented details of Apache Flex. For me, this was by far the most disappointing part of the evening. Sadly Peter came across as very negative toward the community. He did little more than complain that the email list is full of topics that do not interest him and ridicule our efforts to develop our own compiler. Also, he seemed unaware of some of the great news we have had lately such as the fact that FlexUnit is to be donated to Apache, so we will have our own testing framework too. As he is touring around Europe with Adobe, giving these presentations to various user groups, I hope he’ll take the opportunity to make his future presentations more upbeat.
Tink gave a good upbeat presentation on his navigator components and how they have been committed to a branch of the Apache Flex SDK. There is a good chance that they will become part of a future Flex release.
At the end there was an open discussion with the audience. What I took away from that is that the community is less angry now and more just disillusioned. Adobe are making plenty of noises about how Flash has a future, but people just don’t feel they can trust them any more. Apache Flex is developing into a great community, but still people are nervous over whether it has a future. It’s a position I can completely understand. I’m trying to put a fair bit of effort into Apache Flex at the moment, yet at the back of my mind is a nagging doubt: am I wasting my time doing so?
Frustratingly I do not have a crystal ball that can show me the future. Flex remains the best by far RIA framework. It could have a great future, long after the Flash player is a technology memory. Its future doesn’t lie in the hands of people who choose based on best technologies though, so it’s future remains uncertain.