Re: Multi-platform discussion

[Sam]

Hello BFUG,

I wanted to ask the collective here for any advice that could be
offered up relating to making highly-interactive multi-platform
content. The majority of my work is related to generating training
software, games, interactions and simulations. In the company I mainly
work for we have built up a huge custom framework over the years to
handle this task. It is, unsurprisingly, all in Flash, which has been
the perfect tool for the job. It has also pretty much always been
targeted at the desktop. Some clients are starting to need their e-
learning courses to work across different platforms, in particular
tablets. Given the shrinking rate of desktop use online this comes as
no surprise. It also seems that budgets aren't necessarily scaling
with the increase in complexity of the technical requirements. I
wanted to ask your thoughts on two things, one short term and one long
term:-

1) Short term - I have one client who is asking for a course, already
built by us in flash, to be made available for a variety of tablets,
including the iPad. We priced out the cost of a remake in HTML5 (and
it's accompanying technologies) but this was rejected as being too
high (and given their original investment, I agree with them). So my
next port of call would be to use Air for Android/iOS. I wanted to
know if people had any input on the following:-

- The course is currently built to a fixed pixel size. What is a good
strategy for dealing with difference in resolution sizes between
devices?
- What about mouse events? I am assuming ROLL_OVER/OUT are no longer
any use when converting content to touch screen?
- Deployment.... Our courses are usually deployed to a Learning
Management System which tracks the user's progress and retains their
history and whether they have completed the course or not. How would
this work for an AIR app? Are there alternative appstore type portals
that corporates use to deploy this kind of non-commercial content?
- Is using AIR the best idea for this situation or am I missing
something?
- What pitfalls should I be looking for when figuring out what is
actually involved in remaking flash desktop content for tablets?

2) Long term - What is a solid technical strategy to cater for the
changing multi-platform landscape? Right now, you could argue Flash is
still the perfect tool for the job. It does everything we need for our
games and interactions, it behaves the same cross-browser (for the
most part), it's performance fairs well across a range of specced
machines and there are lots of already built, solid frameworks,
including our own custom one, to help get the job done. However, it
seems like this won't be the case for ever and I wanted to know what
people thought were strong alternatives:-

- What are people's thoughts on HAXE?
- What (non-flash) html5/js frameworks are worth considering for
generating highly-interactive content (containing multiple games,
complex animations, lots of external assets such as video, audio and
image) for multi-platform delivery?
- What other technologies/languages might be worth considering?

All input is much appreciated and well-recieved.

Much thanks,

Sam

--
Sam McFadden | Interactive Developer | w: http://sambrick.wordpress.com
| twitter:@SamBrick

[s'unya]

Hi Sam,

There are loads of contradictory answers to your questions and it seems that the best answers are largely to do with your particular situation. I would say that, clearly flash is no longer the best tool for the job if you are not reaching your full audience... and perhaps we should learn from the reliance on plugins?

There are plenty of HTML5 frameworks out there, each more or less suited to different types of job. I've been using limejs for a while now and I would definitely recommend it for general game/scene based animation. It takes some getting used to if you are not used to javascript inheritance and the docs are less than perfect, but it is a very well built system (and you can change it).

Apologies for skirting around your specific questions.

--
You received this message because you are subscribed to the Google Groups "Bristol Flash User Group" group.
To post to this group, send email to bristol-flas...@googlegroups.com.
To unsubscribe from this group, send email to bristol-flash-user...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/bristol-flash-user-group?hl=en.

--
                      o
            o                 o
                O   O   O
      o  o   O s´unya O   o  o

[Nick Forsberg]

Pint? Possibly the best way of breaking this down into manageable points.

[David James]

I second that

Sent from my iPad

[Sam]

Hi S'unya,

Thanks for your input, much appreciated. I'll take a look at limejs.
Have you heard of SproutCore (http://sproutcore.com/)?

In the process of my research so far I have decided that using AIR is
a non-starter for iPad. AIR for iOS is not an AVM and you can't load
external swfs with any useable code in them. This would require a
total rewrite of our courseware which is based around modular external
swfs. Also, deployment of "managed apps"/custom apps is another
complication as they will not easily interface into a lot of the LMS's
we use. And there is also a cost overhead in MDM server/IT setup and/
or a third party system such as AppCentral.

I am actually starting to think HAXE is possibly a great solution and
am going to investigate further down this route. It's great in terms
of refactoring our existing AS3 courseware, and also the flexibility
it allows you in targeting deployment platforms (Android, iOS, webOS,
HTML5 and Flash).

As for a pint...... sounds good, not sure I can make one this year but
January would be great :-)

Cheers,

Sam

On Dec 15, 9:29 am, "s'unya" <sunya.dick...@gmail.com> wrote:
> Hi Sam,
>
> There are loads of contradictory answers to your questions and it seems
> that the best answers are largely to do with your particular situation. I
> would say that, clearly flash is no longer the best tool for the job if you
> are not reaching your full audience... and perhaps we should learn from the
> reliance on plugins?
>
> There are plenty of HTML5 frameworks out there, each more or less suited to
> different types of job. I've been using limejs for a while now and I would
> definitely recommend it for general game/scene based animation. It takes
> some getting used to if you are not used to javascript inheritance and the
> docs are less than perfect, but it is a very well built system (and you can
> change it).
>
> Apologies for skirting around your specific questions.
>

[David james]

Have you looked at mdm zinc? I notice that now compiles to mobile platforms and you can include external swfs in the final app

[s'unya]

Hi Sam,

Thanks for sproutcore, that looks good as a framework... at leas the docs are clearer than limejs.

I agree about AIR, it just isn't going to work in a complex application... On the subject of HaXe, I remember looking at it and wondering whether it would be sufficiently efficient. It is one language that compiles to any of several languages, which means that it must automatically convert to different languages. I'm always suspicious of 'automatic' translation, I usually read 'bad'.

Basically, if you are making an application for html5 + associated techs, I'd recommend doing that specifically so that you can optimise to that platform.

Cheers,