Adobe Air / MS Silverlight... what's next

9 Apr 2009 - 11:56am
5 years ago
13 replies
1945 reads
Greg Petroff
2004

If you are working with Air or Silverlight ... where do you think it's
going? What is the current state of the tools? There yet? Not? How do we
see these changing what we do?

Love a discussion on what changes with them? What's the same? Are they
harder to build and develop then other alternatives? What do they mean
for the programmers/developers we work with?

Personal interest here:

Looking for examples of AIR and Silverlight that mimic a singular
environment...i.e. almost like creating your own os with its own
behaviors, affordances etc. where the environment for work takes over
from the standard desktop. Anyone have any favorite examples?

-gp

Comments

9 Apr 2009 - 12:33pm
dferriman
2009

With cloud computing on the rise in popularity (even though I am sure
there are many talking about it that do not know what it is) I think
AIR has a bright future. The lines between desktop and the internet
are blurring and apps like Adobe AIR are the future. Silverlight is
just Microsoft's attempt to join the band wagon, like everything
they do -other than MS Office (for the most part). I am personally
hoping to see it fade out.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

9 Apr 2009 - 2:18pm
Dave Malouf
2005

Greg it is really hard to answer this, b/c it really depends on your
perspective, the culture of your team, and your existing relationship
of your team to any other tools.

I would say that both environments have a bit to go before they are
fully baked, but then again so does Java and other mature
alternatives.

Also, I think you are saying Silverlight when you mean WPF/.NET 3.5.
Silverlight is to Flash what AIR is to .NET 3.5 (so to speak) in so
far as Silverlight is really about being inside the browser and is a
limited subset of functionality than you get with full WPF/.NET 3.5.
>From the ground up there is so much in WPF that make it much better
as a technology for doing full screen, multi-layered complex
applications. But it WPF does create limitations that are not in .NET
3.5 by itself, but w/o WPF you really can't take advantage of the
other end of the spectrum that I think you want to consider which is
the designer environment:

AIR is a developer environment of Flex Builder and a designer
environment of either Flash or Flash Catalyst (coming to a theatre
near you).

.NET 3.5/WPF has Visual Studio for the developer and idealing
Expression Studio for the designer (but said designer better have
some solid coding chops to really make it useful). I.e. I still have
not seen a designer do with Expression what I've seen designers do
for a good decade with Flash.

That being said, I've seen parity in coding output from both Flex
Builder and Visual Studio.

If you want to build a desktop operating system that replaces the
existing one (man! who wouldn't) and it has to be cross platform,
you really have no choice but to use AIR if you want to be outside
the browser.

I forget, but I'm pretty sure that AIR has many of the same sandbox
protocols as Flash even though it is outside of the browser. This may
prevent you from easily doing things at the desktop level that you
might assume.

Java from what I've seen is the only cross-platform distribution
that allows you outside the browser give the user the option to allow
you the full robust feature set of an installed desktop application.
Of course, maybe you don't need that, but if you are trying to
emulate an OS you just might.

Oy! this just gets really complicated!! And I'm sure I'm butchering
it to death and others are going to save me from myself.

Good luck Greg!

-- dave

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

9 Apr 2009 - 2:33pm
ambroselittle
2008

On Thu, Apr 9, 2009 at 12:56 PM, Petroff, Greg <greg.petroff at sap.com> wrote:

> If you are working with Air or Silverlight ... where do you think it's
> going? What is the current state of the tools? There yet? Not? How do we
> see these changing what we do?
>

Well, I'm pretty familiar with Silverlight. That's what we built
Quince<http://sn.im/quince-intro>,
our UX design patterns explorer, on, and I'm the lead author on Wrox
Silverlight 3, a programmer's reference (not geared at designers, but we do
cover aspects of design--our director of visual design, Grant Hinkson, is a
co-author and writes on that)--due to be released early this summer, barring
unforeseen surprises. :)

Personally, I think the foreseeable future of software applications for
desktop/laptop and, eventually, mobile and other devices is Silverlight/AIR.
It'll take a while for us to get there, but they're just waay more solid
platforms to build on than HTML+CSS+JS, which were not designed for rich
interactive stuff (although I am impressed what we've [the software
industry] been able to do with those technologies).

As far as tooling for Silverlight goes, the foundations are certainly
there--VS for devs and Blend for technically-adept designers. I think VS is
pretty solid for the devs already, though I think most devs don't want to
have to go to another tool for an effective design experience (doubtless
that problem will be solved in the not-too-distant future).

Even though the tools for SL (and AIR) are still young--babies--I still
maintain that we could not have produced Quince or (more importantly)
maintain and enhance it as effectively had we chosen Ajax as our platform.
I can tell you from personal experience and from my knowledge of many devs
that the majority of devs will (or already do) vastly prefer a more reliable
platform on the client (like Silverlight). It's just ridiculous how much
head-banging-on-desk they have to do for HTML et al, and the tools are flaky
at best (even though they and browsers are light years better these days).

>From a designer perspective, you will be empowered to explore more
interaction possibilities than are available with Ajax, which sounds like
something you want to do, not just in terms of technology/platform
capabilities but just in the lower cost for making those kinds of
potentially richer interactions available in the end product. These (SL/AIR)
ameliorate the feasibility design constraint.

In terms of recommendations, there are many factors to consider. What is
your team's background--what technologies and tools are they most productive
with today? That's one, potentially big factor in choosing. Sadly, there
are also prejudices/bigotry that you may bump up against as well, both on
teams and with some users. Then of course there's the target audience,
which is the most critical factor--will they have or be willing to install
the necessary plug-ins?

As for being outside the browser, Microsoft announced at MIX09 that Silverlight
3 does support out of browser experiences. I can confirm this from
experience--you can get the beta today if you want. Blend 3 also has some
interesting improvements for designers, but I leave it to individuals to
judge if it is non-technical enough for their tastes/capabilities.

I can't speak so much for AIR, but I can say that IMO the future is very
bright for Silverlight, and I recommend it as a de facto platform to do new
work on and then move away from it as needed given the other considerations
for particular teams, target audiences, and problem domains.

Maybe the big bummer is the mobile story right now. I think it's the next
big battleground for software (if it isn't already!). It will be fun to see
how it all works out, but also a bit painful for those of us trying to build
great stuff in the meantime..

Hope this helps.

--Ambrose

9 Apr 2009 - 3:16pm
Angel Marquez
2008

infragistics=B*D *SS

On Thu, Apr 9, 2009 at 12:33 PM, J. Ambrose Little
<ambrose at aspalliance.com>wrote:

> On Thu, Apr 9, 2009 at 12:56 PM, Petroff, Greg <greg.petroff at sap.com>
> wrote:
>
> > If you are working with Air or Silverlight ... where do you think it's
> > going? What is the current state of the tools? There yet? Not? How do we
> > see these changing what we do?
> >
>
> Well, I'm pretty familiar with Silverlight. That's what we built
> Quince<http://sn.im/quince-intro>,
> our UX design patterns explorer, on, and I'm the lead author on Wrox
> Silverlight 3, a programmer's reference (not geared at designers, but we do
> cover aspects of design--our director of visual design, Grant Hinkson, is a
> co-author and writes on that)--due to be released early this summer,
> barring
> unforeseen surprises. :)
>
> Personally, I think the foreseeable future of software applications for
> desktop/laptop and, eventually, mobile and other devices is
> Silverlight/AIR.
> It'll take a while for us to get there, but they're just waay more solid
> platforms to build on than HTML+CSS+JS, which were not designed for rich
> interactive stuff (although I am impressed what we've [the software
> industry] been able to do with those technologies).
>
> As far as tooling for Silverlight goes, the foundations are certainly
> there--VS for devs and Blend for technically-adept designers. I think VS
> is
> pretty solid for the devs already, though I think most devs don't want to
> have to go to another tool for an effective design experience (doubtless
> that problem will be solved in the not-too-distant future).
>
> Even though the tools for SL (and AIR) are still young--babies--I still
> maintain that we could not have produced Quince or (more importantly)
> maintain and enhance it as effectively had we chosen Ajax as our platform.
> I can tell you from personal experience and from my knowledge of many devs
> that the majority of devs will (or already do) vastly prefer a more
> reliable
> platform on the client (like Silverlight). It's just ridiculous how much
> head-banging-on-desk they have to do for HTML et al, and the tools are
> flaky
> at best (even though they and browsers are light years better these days).
>
> >From a designer perspective, you will be empowered to explore more
> interaction possibilities than are available with Ajax, which sounds like
> something you want to do, not just in terms of technology/platform
> capabilities but just in the lower cost for making those kinds of
> potentially richer interactions available in the end product. These
> (SL/AIR)
> ameliorate the feasibility design constraint.
>
> In terms of recommendations, there are many factors to consider. What is
> your team's background--what technologies and tools are they most
> productive
> with today? That's one, potentially big factor in choosing. Sadly, there
> are also prejudices/bigotry that you may bump up against as well, both on
> teams and with some users. Then of course there's the target audience,
> which is the most critical factor--will they have or be willing to install
> the necessary plug-ins?
>
> As for being outside the browser, Microsoft announced at MIX09 that
> Silverlight
> 3 does support out of browser experiences. I can confirm this from
> experience--you can get the beta today if you want. Blend 3 also has some
> interesting improvements for designers, but I leave it to individuals to
> judge if it is non-technical enough for their tastes/capabilities.
>
> I can't speak so much for AIR, but I can say that IMO the future is very
> bright for Silverlight, and I recommend it as a de facto platform to do new
> work on and then move away from it as needed given the other considerations
> for particular teams, target audiences, and problem domains.
>
> Maybe the big bummer is the mobile story right now. I think it's the next
> big battleground for software (if it isn't already!). It will be fun to
> see
> how it all works out, but also a bit painful for those of us trying to
> build
> great stuff in the meantime..
>
> Hope this helps.
>
> --Ambrose
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

9 Apr 2009 - 6:17pm
Greg Petroff
2004

Thanks ...all.

Confirms most of what I know as of right now. One thing I am
wondering about is how people might work with Rich Clients that
become almost sovereign over the desktop. Will there be issues or
will it be a logical place for people to fall on.

The context switching between groups of apps on your iphone, bringing
up widgets in OsX, etc., having collections of work in different
"desktops", is becoming a more familiar set of patterns.

The promise of Air/Silverlight is to decouple the front end from the
underlying objects which as Ambrose suggests make it much easier to
build and modify.

A lot of this also "depends" on so many issues from what the user
has installed, their needs, the cost to develop and support, the
skills of your team etc.

Love to talk about and better discern from a designer's pov what
criteria one would use to evaluate looking making a choice between
these different paths from AJAX to Air etc.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

9 Apr 2009 - 6:56pm
Chad Vavra
2008

>From a new user perspective:
AIR = Application download and install
Silverlight = Plugin download and install
AJAX = works in a browser.

But from the content owners perspective:
AIR = fairly stable, stand alone. Cross-platform
Silverlight = not so stable, bloated IE framework
AJAX = we already have a website.

So AIR is winning for it's cross-platform, not a website, but losing
a little as a application install.

However, if I think back I know that a ground-breaking app. A must
have thing could change all that. In 1999 content was king, and it
still is. Platforms, frameworks, libraries are still just tools for
the artist. They are only worth the creations made of them.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

9 Apr 2009 - 7:59pm
Dave Malouf
2005

Greg, I would caution your application of a pattern that works in one
platform (user context, device type, operating system) to another.

We have found in recent research that changes in platforms change
expectations and that different conventions do not all transfer
between platforms especially those where the form factors and context
of use are so incredibly difficult.

-- dave

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

9 Apr 2009 - 8:04pm
Andy Edmonds
2004

It's important to note that AIR has two user interface toolkits: HTML
via the Webkit rendering engine and their Flex UI toolkit (declarative
grids, buttons, etc) with support for Flash in both cases. The rise of
webKit as a platform for rich networked applications is quite
interesting. IE has been used extensively as a UI toolkit for desktop
apps but is waning in that context due to antiquated technology and
the windows only issues.

Firefox has a thing called XUL Runner which allows creating
applications, and a really nice UI toolkit called XUL, but the Mozilla
organization is focused on the consumer over the application developer
and is strongly motivated to participate in standards based
technology, limiting their willingness to jump ahead. The XUL toolkit
is especially interesting given it's robust handling of platform UI
consistency. There's support for swapping the orientation of buttons
based upon platform standards for instance.

That's a big issue for "RIA"s -- consistency with the platform.
Still, at least among the twitter crowd, it doesn't seem to be slowing
down the adoption of custom desktop applications for twitter and, as
of late, services like friendfeed and facebook.

Webkit is clearly taking over as the embedded HTML rendering engine of
choice. With it's inclusion in the iPhone and Android and rapid
evolution of features, my bet is on Webkit as the most common
technology in apps which form hybrids between web and desktop. The
HTML 5 sqlite database support makes standalone applications, and
offline persistence of online data, easy and a re-usable skill. Apple
is pushing standards in interesting ways to optimize Webkit's success
-- like CSS transform animations (eg.
http://surfmind.com/lab/typhographic/). The Canvas tag, providing a
pixel based drawing palette, also means almost any visual effect is
attainable. Library's like RaphaelJS (http://raphaeljs.com/) provide a
more DOM like development experience with canvas. Webkit is a
delightful platform to work with for the developer use to being held
back by IE and hampered by variations across browsers while still
taking advantage of those skillsets.

Of all the features enabled by OS native code, notifications and
desktop overlays seem to be the most commonly deployed. Adobe AIR has
a strong sandbox limiting local code to a greater degree than say
Firefox extensions (unsure about Silverlight).

A new contender, Titanium (http://www.titaniumapp.com), offers a
webkit based UI environment with native installers for Win/Mac/Linux,
as well as full desktop access and integration of Ruby & Python, as
well as the standard notification and windowing APIs.

I don't see these technologies offering alternative OS experiences,
but I am excited about the ability to provide applications which are
more useful than web applications in a variety of ways, using the
technology I've spent over a decade mastering :)

Cheers,
Andy
http://delicious.com/andyed/js

On Thu, Apr 9, 2009 at 12:56 PM, Petroff, Greg <greg.petroff at sap.com> wrote:
> If you are working with Air or Silverlight ... where do you think it's
> going? What is the current state of the tools? There yet? Not? How do we
> see these changing what we do?
>
> Love a discussion on what changes with them? What's the same? Are they
> harder to build and develop then other alternatives? What do they mean
> for the programmers/developers we work with?
>
> Personal interest here:
>
> Looking for examples of AIR and Silverlight that mimic a singular
> environment...i.e. almost like creating your own os with its own
> behaviors, affordances etc. where the environment for work takes over
> from the standard desktop. Anyone have any favorite examples?
>
> -gp
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

10 Apr 2009 - 12:07am
Greg Petroff
2004

Dave I share your caution and pov on context, use, etc. I'm doing
some outside in right now, more looking for some attempts at unusual
ways of using RIA's as benchmarks.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

10 Apr 2009 - 12:59am
Rob Christensen
2009

Hi Greg, Andy, Dave, and Chadvavra,

My name is Rob Christensen and I am a product manager on Adobe AIR.
Adobe AIR is really centered on allowing developers to build
applications that need to run outside of the browser. Since it
includes WebKit and Flash, it's generally very easy to migrate your
existing UI code to AIR and also take advantage of any services you
have already built or tap into third party services.

A certain class of applications require greater access to the local
system than what is currently possible in a browser. For example, if
you want to have access to local file system, display notifications,
encrypt a local database, have full control over windows, etc., Adobe
AIR is a great option. We do much of the heavy lifting in the runtime
so that designers and developers can focus on building a great
application experience.

One application that I would point you to is Salesbuilder built by an
Adobe employee. It is a sample application that showcases what is
possible when you build a desktop application using Adobe AIR and
Flex.

Video:
http://vimeo.com/3882718?pg=embed&sec=

A blog post about the application including a download link to the
actual application:
http://coenraets.org/blog/2009/03/make-and-receive-phone-calls-in-the-new-version-of-the-salesbuilder-sample-flex-application/

The developer also made the source code available here:
http://coenraets.org/blog/2009/04/new-version-of-salesbuilder-with-ribbit-integration-source-code-available/

Feel free to contact me directly if you have additional questions.

Cheers,
-Rob
Product Manager, Adobe AIR
http://blogs.adobe.com/air/

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

10 Apr 2009 - 10:01am
ambroselittle
2008

On Thu, Apr 9, 2009 at 12:17 PM, greg <greg.petroff at sap.com> wrote:

> One thing I am
> wondering about is how people might work with Rich Clients that
> become almost sovereign over the desktop. Will there be issues or
> will it be a logical place for people to fall on.
>

I think if it is "the" environment that folks work with mostly during the
day, it could work--people shouldn't have to think "I need to load SAP and
then get to X app that I want" but if they're just sort of living in SAP and
hitting different apps unconsciously, it might be okay. I would be
concerned, though, if folks have to sort of take two conscious steps just to
get to the tool that helps them accomplish their goals. In any case,
certainly a richer, more reliable RIA platform would make this more
manageable, especially in the long term.

> The context switching between groups of apps on your iphone, bringing
> up widgets in OsX, etc., having collections of work in different
> "desktops", is becoming a more familiar set of patterns.
>

In Quince, we talk about the Hub and Spoke and One Window Drill Down
patterns. These are used a lot in mobile but not so much (yet) in desktop,
but if your context calls for it, so be it. You might check those patterns
out and see if they help at all in your thinking about this.

> Love to talk about and better discern from a designer's pov what
> criteria one would use to evaluate looking making a choice between
> these different paths from AJAX to Air etc.
>

I think the key thing, from a designer's point of view, is the reduction of
the feasibility constraint (comparing Ajax to the RIA platforms). There are
some things you just can't do in Ajax; there are some things that are just
too hard to do, and then there are many things that are just harder to do
and/or maintain in Ajax.

When choosing between RIA options strictly from the designer's perspective
(all other things being equal), I guess it would come down to tooling &
background--something you have to evaluate based on your team. But in
reality, I think the choice of technology is going to be decided by other,
more weighty factors already noted like target audience, perceived
maintainability, perceived future success (i.e., will I be able to continue
to recruit talent to work on it), and even cost, and the designer should
just work with what makes sense for the project as a whole.

HTH.

-ambrose

10 Apr 2009 - 11:06am
ambroselittle
2008

On Thu, Apr 9, 2009 at 12:56 PM, chadvavra <chad.vavra at imedstudios.com>wrote:

> Silverlight = not so stable, bloated IE framework

This just perpetuates prejudice and ignorance.

The Reality
Silverlight is not an "IE framework." It is based on the .NET framework,
which has been in production since 2002. The UI engine was based on WPF,
Microsoft's latest Windows platform UI technology, and has been in
production since 2006. Silverlight was originally called WPF/e, i.e., WPF
everywhere--Silverlight brings the core of the .NET framework and Visual
Studio (which MS devs *love* and non-MS devs envy) to cross-platform,
cross-browser.

It is not Windows-only, nor is it IE-only.

I suggest you check out the Silverlight 2 System
Reqs<http://www.microsoft.com/silverlight/overview/faq.aspx#sys-req>;
as for browsers, I refer you to this post on Silverlight Browser
Support<http://blogs.msdn.com/jstegman/archive/2008/12/21/silverlight-browser-support.aspx>;
We've found that Quince runs fair decent in IE, Safari 3 (and 4 Beta on
Mac), Chrome, and Firefox.

"bloated"? Nope. Considering how much they cram into those 4MB for Windows
and 7MB for Mac, it's not bloated at all. That's like one MP3 on Win or a
long one on Mac. And with Silverlight 3, they've added a bunch of good
stuff while reducing the download size (through optimizations).

"not so stable"? How do you corroborate this statement? Last year's
Olympics ran on Silverlight 2 *beta*. It went so well, they're renewing for
the next Olympics. Netflix is giddy about Silverlight for their player. I
can speak from experience that Quince has been a dream of stability as well.
Sure there are issues, but overall, it's been great. On the other hand,
consider this<http://www.pcworld.com/businesscenter/article/162805/mlb_in_social_media_stumble_with_mlbtv_blog.html?tk=loom_art>
.

Don't Let Prejudice Impede Your Professional Growth
I don't want to start a war by any means, but I am just repeatedly amazed at
the level of ignorance and prejudice some people express about Microsoft and
Silverlight, particularly. It bears correcting. We should be
professionals, not tribesmen.

Look, I'm sure folks could look at me and blow me off as a Microsoft fanboy.
If so, let me say I'm just a fan because my long, in-depth experience with
them has made me that. Microsoft has been very good to me, as a
professional. While I've never worked for them, they've empowered me--and
the businesses I've worked for--to be more successful; they build and
support great professional communities, and for devs especially, they build
great tools.

That said, I don't let that blind me to see what is good in other solution
platforms or what could be improved in theirs. In fact, I've long been an
"insider" of sorts precisely so that I could provide positive feedback to
improve the tools I worked on daily. I don't hate Adobe by any means; I've
been learning Fw cuz it enables me to do my job. I don't hate Apple--on the
contrary, I really love my iPhone and enjoy my MacBook!

Why would I spend so much (negative) emotional energy on a technology
provider? That's just silly, IMO. Especially as a designer--the choice
between technologies that are more or less equal in terms of interface
capabilities (design constraints) should really have little impact. Just
use the tools that make you most effective as a designer.

If my situation--the project I were on--called for some other technology
platform, I'd think of it gleefully. An opportunity to greatly expand my
horizons, my knowledge, and my skills. I wouldn't let prejudice stop me
from improving myself, and it'd be one more tickmark on the ol' résumé
that'd make me that much more valuable.

Even if you don't care about your own professional development, at a
minimum, I suggest folks stop parroting ignorant talking points and just
actually evaluate things for what they are. The information is out
there--get it from the horses' mouths or reputable, dispassionate 3rd
parties; download trials and try for yourself. If Silverlight or AIR or
Flash or <insert tool here> doesn't make sense for your context, that's
fine, but dismissing it out of hand due to some prejudice is just plain
silly and unprofessional.

--Ambrose

12 Apr 2009 - 9:54pm
ehaagwlke
2008

I am using an AIR twitter client - twhirl, and it works soundly.

In my opinion, at present, AIR is the best of the three. I tried it
in Ubuntu/Mac OSX/Windows, the app behaved exactly the same. Though
Silverlight has its linux version - Mono, but its not the same as the
windows version.

@Greg: What about JavaFX?

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=41126

Syndicate content Get the feed