incubator-flex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Left Right <olegsivo...@gmail.com>
Subject Re: [OT] Flash Platform roadmap
Date Sat, 25 Feb 2012 20:17:13 GMT
Presentation interface of Flash and Flex interest precisely the prospects,
it's not interesting for a wide range of developers, who write code that
leverages client-server communication, data processing, physics, AI and
many many more. I, for once, being an AS3 programmer at my last job had to
deal purely with only the code that manages some video-peer-to-peer-chat
kind of thing, and the visual side of project never was in my real of
responsibility. However, the must-have visual side of thing (which in my
case was an ideal white screen 600x800 pixels) hindered the process a lot -
Flash player is creepily slow and unproductive when communicating outside.
On tests, you would often want to run 100 copies of the program, but maybe
even more... running so many flash players would be very unhealthy for the
system :)

And, yeah, like Omar Gonzalez says - testing. You may think about testing
as unimportant, but the larger the project is, the higher % of time it
spends in testing. In fact, a project with the development cycle of about a
year may spend 1-2 months in actual development, and the rest is testing.
Not being able to automatically register or send clicks to player is
maddening.

If, as I mentioned it before, the player could have been combined of two
parts, one being the runtime for the code, that expect callbacks from
another part for doing the rendering and piping in the input events - that
would be priceless, as someone (well, someone like me even!) could've
written a mock for the interaction API to send automatically generated
signals back and forth.
This is also where HTML wins on almost all positions - you can automate the
testing, the continuous integration, and make it all run automatically,
saving a lot of work hours for QA personnel...

Best.

Oleg

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message