RE: Apple's Dashboard (Was: The End of Usability Culture?)

19 Nov 2004 - 4:19pm
10 years ago
13 replies
1101 reads
Lyle_Kantrovich...
2004

<Ziya>
Once we get "canvas" and other goodies with Dashboard, across platforms,
I
think the notion that web's dead will look pretty silly.

<http://www.apple.com/macosx/tiger/dashboard.html>
</Ziya>

I don't see anything on the dashboard page you reference that indicates
anything related to web application development being "dead" or "alive"
in the future.

<Apple>
"The Dashboard is a semi-transparent layer that zooms across your
Desktop with a single button click, similar to the way Exposé works. To
put the Dashboard away, simply press the function key again and you’re
back to where you started without messing with your Desktop or dock. Use
the Dashboard to get fast access to information and application controls
you use most frequently."
</Apple>

I also can't help but think that "dashboard" is what "desktop" was
intended to be...it's just been hijacked by every icon-dropping install
program to come along in the last 13 or more years. Users multi-task
much more now than they ever have, and every app wants some attention.
Collectively, the applications users run today create a window
management nightmare...so now instead of
moving/closing/minimizing/sizing windows to get back down to the desktop
(has anyone else become and avid <windows key>-M shortcut user?), Apple
is creating a new "desktop" (aka dashboard) that you can popup on top of
all the other crud that's running.

<Apple>
"The Dashboard is home to a new kind of application called Widgets.
Widgets are mini-applications written in JavaScript that are designed
for fun as well as function."
</Apple>

Yawn. Please see "portlets" from "portals" hype last 4+ years.

<Apple>
"Leave No Widget Unturned
Elegantly designed Dashboard accessories animate in cool and interesting
ways..."
</Apple>

Yes. The animation will be much cooler and more interesting than the
content or functionality in the accessories, at least if the past
predicts the future. A clock, calculator or sticky note that animates
is not any better than one that doesn't animate...and animation loses
it's "cool" quickly.

<Apple>
"Complex transitions, movement and visual effects are just a few of the
amazing visual feats that make Widgets as great looking as they are
useful. Developers can build their own Widgets using the JavaScript
language to take advantage of Core Image capabilities."
</Apple>

Amazing! This won't significantly improve anything other than the stock
prices of CPU and memory makers. I don't think it's been the lack of
animation and drawing tools that have held web apps back. Designers can
do amazing things with Flash already...yet Rich Internet Apps
(Flash-based or otherwise) aren't very common today.

Let's face it, web apps are hampered by a puny UI toolset and limited
client capabilities. I think RIA's will eventually take off, but we
need more robust and mature technology standards to work with before
that will happen.

Meanwhile, GUI's are adopting aspects of web apps...

Regards,

Lyle

----
Lyle Kantrovich
User Experience Architect
Cargill
http://www.cargill.com/

Croc O' Lyle - Personal Commentary on usability, information
architecture and design.
http://crocolyle.blogspot.com/

"Simplicity is the ultimate sophistication."
- Leonardo da Vinci

Comments

19 Nov 2004 - 5:30pm
Listera
2004

Lyle_Kantrovich at cargill.com:

> I don't see anything on the dashboard page you reference that indicates
> anything related to web application development being "dead" or "alive"
> in the future.

I don't know why it would.

If you had been following the thread here *and* understood the strategic
significance of not just Dashboard but also the *xplatform* agreement among
all non-IE browser/platform vendors with which it's associated, you'd notice
that this gives non-programmers (those who do markup coding in HTML/XML,
scripting, etc) an easy way to create dynamic online applications that
bridges the web beyond the browser with the desktop. Canvas and other visual
goodies are just the icing on the cake.

(Your recitation of Dashboard details betray a fundamental lack of
understanding of the underlying technologies involved and the possibilities
they augur, so I won't deal with the rest. You'll of course be free to not
use any of this technology.)

Ziya
Nullius in Verba

19 Nov 2004 - 5:48pm
Andrei Herasimchuk
2004

On Nov 19, 2004, at 3:30 PM, Listera wrote:

> If you had been following the thread here *and* understood the
> strategic
> significance of not just Dashboard but also the *xplatform* agreement
> among
> all non-IE browser/platform vendors with which it's associated, you'd
> notice
> that this gives non-programmers (those who do markup coding in
> HTML/XML,
> scripting, etc) an easy way to create dynamic online applications that
> bridges the web beyond the browser with the desktop. Canvas and other
> visual
> goodies are just the icing on the cake.

That might actually mean something if Apple's share of the market was
significantly greater than 3%.

> (Your recitation of Dashboard details betray a fundamental lack of
> understanding of the underlying technologies involved and the
> possibilities
> they augur, so I won't deal with the rest. You'll of course be free to
> not
> use any of this technology.)

Even with a full understanding of Dashboard, it's not going to have
that much of an impact in the grand scheme of things. The past 20 years
are littered with Apple technologies that promised massive change, none
of which ever did. At least not by Apple.

Andrei

19 Nov 2004 - 6:18pm
Listera
2004

Andrei Herasimchuk:

> That might actually mean something if Apple's share of the market was
> significantly greater than 3%.

Sigh. I will mention this one last time, as I have already provided relevant
URLs for those actually following the thread earlier.

This IS NOT about Apple. The stuff underneath Dashboard is/will be
xplatform, everybody-but-MS signed on to it, it's not proprietary. It's
about enabling a large class of markup developers and non-programmers to
create (essentially) RIAs, without much of the pain.

You may think that's not important. Just like GMail showed millions what can
be done with HTML/JavaScript, so will these.

I have no idea what Apple's market share have anything to do with this. They
will run on Windows too.

> Even with a full understanding of Dashboard, it's not going to have
> that much of an impact in the grand scheme of things.

I'm surprised you reached that conclusion months before it is even released.

Ziya
Nullius in Verba

19 Nov 2004 - 6:36pm
Andrei Herasimchuk
2004

On Nov 19, 2004, at 4:18 PM, Listera wrote:

> Sigh. I will mention this one last time, as I have already provided
> relevant
> URLs for those actually following the thread earlier.

I missed those URLs.

> This IS NOT about Apple. The stuff underneath Dashboard is/will be
> xplatform, everybody-but-MS signed on to it, it's not proprietary. It's
> about enabling a large class of markup developers and non-programmers
> to
> create (essentially) RIAs, without much of the pain.

That would be news to me. The last time I checked (about two months
ago), while the stuff to enable Dashboard is not proprietary per se, it
was Mac OS specific in terms of execution, and required browser makers
to agree on the way in which Apple and the Safari team were
implementing new tags. Where are the details on this bit of news?

To say "everybody-but-MS" is signed on board is a nice way to rally the
troops, but it doesn't change the reality of how ***much*** of an
impact these new widgets will have if 90% of the market won't even know
they exist. You're talking about some large impact that Dashboard
widgets wil have. I'm not discounting they are cool, I'm discounting
how relevant they will be to most users in the larger picture of the
computing market.

> You may think that's not important. Just like GMail showed millions
> what can
> be done with HTML/JavaScript, so will these.

You're reaching. GMail is popular because the account is free and you
get a very large amount of free space to use for mail. It has little to
nothing to do with the UI and the DHTML-based browser of the
functionality. If Google had given away the same email offer but you
had to download a small client app, I'm willing to gamble a large sum
of money they'd have gotten 95% of the current user base.

> I have no idea what Apple's market share have anything to do with
> this. They
> will run on Windows too.

That would be news to me. Again, I wasn't aware that the Mozilla was
going to take the time or effort to support Dashboard tags and
technology. Further, how will the user experience be compromised on
Windows given it won't be as high level a feature as it will be on
MacOS? Will users have to launch URL's and shortcuts to load up
widgets, after they've launched a browser? Clumsy at best. And finally,
considering IE usage on Windows, even at 10% Firefox use, just how
significant will Dashboard widgets be on Windows?

I think you are vastly overstating your case, Ziya. I mean "vastly"
like some marketing people don't know what Design is kind of vast.

>> Even with a full understanding of Dashboard, it's not going to have
>> that much of an impact in the grand scheme of things.
>
> I'm surprised you reached that conclusion months before it is even
> released.

You're talking to guy a who used to go out of his way for nearly 20
years developing cool Apple technologies that were either 10 years
ahead of time, or so tied to the MacOS that in the end, they didn't
amount to much compared to what other companies were doing. (Shall we
begin to list all the technologies that are log since dead for
posterity's sake here?) I'm all for the innovation Apple does and
leads, but I've given up on being a blind cheerleader for them.

Their record speaks for itself.

Andrei

19 Nov 2004 - 9:27pm
Listera
2004

Andrei Herasimchuk:

> That would be news to me.

Well, I'm surprised you didn't connect the dots.
(There should be some posts from me in the archives for this.)

> To say "everybody-but-MS" is signed on board is a nice way to rally the
> troops, but it doesn't change the reality of how ***much*** of an
> impact these new widgets will have if 90% of the market won't even know
> they exist.

A Dashboard widget is just a web page. Let me repeat that to sink in: A
DASHBOARD WIDGET IS JUST A WEB PAGE! With the full power of HTML, XML,
JavaScript, CSS2, DOM2, Flash, QuickTime, Java, etc. I'm sure you'll grasp
the full impact of that.

> You're talking about some large impact that Dashboard widgets wil have.

The underlying xplatform technology will have the impact, not just
Dashboard.

> I'm not discounting they are cool, I'm discounting how relevant they will be
> to most users in the larger picture of the computing market.

It's months from shipping and you've already declared them dead? I just
don't get the hostility here at all.

>> You may think that's not important. Just like GMail showed millions what can
>> be done with HTML/JavaScript, so will these.
>
> You're reaching. GMail is popular because the account is free and you
> get a very large amount of free space to use for mail.

Wrong lesson! People are loving GMail not because it's JavaScript UI
wizardry. That's not the point. The point is that they have a very
functional, fast email app that does NOT require a fat client, plugin or
some other encumbrance. It simply validates the functionality and usefulness
of nonproprietary web technologies, by the millions.

> I think you are vastly overstating your case, Ziya.

I don't think you've understood my case at all yet.

> You're talking to guy a who used to go out of his way for nearly 20
> years developing cool Apple technologies that were either 10 years
> ahead of time, or so tied to the MacOS that in the end, they didn't
> amount to much compared to what other companies were doing.

Andrei, this is 2004. Do you want me to say that Nagel, Scully, Spindler,
Amelio were pinheads? That some of the moves made by Apple in the 90s were
criminally stupid? This is old news.

Just how many other tech companies of significant size do we have that have
design and user experience indelibly tattooed to every aspect of their
corporate lives? Let's have some perspective here!

Ziya
Nullius in Verba

19 Nov 2004 - 10:12pm
Andrei Herasimchuk
2004

On Nov 19, 2004, at 7:27 PM, Listera wrote:

> A Dashboard widget is just a web page. Let me repeat that to sink in: A
> DASHBOARD WIDGET IS JUST A WEB PAGE! With the full power of HTML, XML,
> JavaScript, CSS2, DOM2, Flash, QuickTime, Java, etc. I'm sure you'll
> grasp
> the full impact of that.

I know what a Dashboard widget is... calm down.

A dashboard widget, as implemented by Apple, requires the OS to allow
it to work outside the web browser to be useful. (thus, the widget
layer that comes into play to layout all these widgets on.) To get all
that cool transparency compositing and widget-like feel and to have
them appear side by side, and other effects needed to fake the fact the
widget is largely just a web page. Without that extra OS support to
make it happen, Dashboard widgets lose a lot of the usefulness factor.

> The underlying xplatform technology will have the impact, not just
> Dashboard.

What underlying technology will have an impact that already hasn't? You
just got through saying Dashboard widgets were just web pages, and now
you're claiming there's something there that will have a a large
impact, but yet there are just web pages. You're not making much
logical sense.

It's a web page with a few extra non-spec'd tags and displayed on a
special layer to make them work last time I checked.

> It's months from shipping and you've already declared them dead? I just
> don't get the hostility here at all.

I never said they were dead. I said they won't have whatever "coming to
Jesus" scale impact you seem to be claiming. You're hedging very close
into zealotry Ziya, something I'm not used to seeing you do.

> Wrong lesson! People are loving GMail not because it's JavaScript UI
> wizardry. That's not the point. The point is that they have a very
> functional, fast email app that does NOT require a fat client, plugin
> or
> some other encumbrance.

Except that fat browser client they already have their machines. I
could also point you to the popularity of poker gaming sites, which do
require a client download, and are used by some of the most
non-technical people in the computing user base. Neither proves much in
the context of this discussion.

> I don't think you've understood my case at all yet.

Then spell it out for a small minded man like myself.

> Andrei, this is 2004. Do you want me to say that Nagel, Scully,
> Spindler,
> Amelio were pinheads? That some of the moves made by Apple in the 90s
> were
> criminally stupid? This is old news.

And exactly what has changed since? Developing technology specifically
for Apple is doomed to 3% market share. I have yet to be convinced
Dashboard widgets are all that, and that they'll have a big impact in
their current form and with current support in Windows, based on my
last assessment of what they were all about this past summer. (If
things have changed, I'm open to seeing what that might mean.) Remember
when the "active" desktop on Windows was all the rage?

There have been many attempts at making dynamic widgets like this and
thei rimpact has been marginal. Further, Longhorn has its own brand of
Dashboard widgets which are arguably more powerful since the technology
to build the entire UI using a tag-based language that can be
repurposed permeates the entire OS. But that of course requires waiting
for Longhorn.

> Just how many other tech companies of significant size do we have that
> have
> design and user experience indelibly tattooed to every aspect of their
> corporate lives? Let's have some perspective here!

How is that relevant to the discussion?

Andrei

19 Nov 2004 - 11:14pm
Listera
2004

Andrei Herasimchuk:

> You're hedging very close into zealotry Ziya, something I'm not used to seeing
> you do.

I believe in the web, I'm not ashamed to say. I've been royally pissed off
at attempts by some companies to kill off the ecumenical nature of the web.
I view Microsoft's insistent move to privatize the web experience into its
proprietary platform with alarm. I see return to fat, platform-specific,
proprietary clients as evil. I see deliberate reduction of choice as
immoral.

So I feel passionate about a collection of technologies that will allow
potentially millions of nonprogrammers to create cool and project-specific
apps that are xplatform. Ongoing negotiations willing, they may have all the
polish you see on Dashboard widgets, and more, on all platforms.

> Except that fat browser client they already have their machines.

Yes, that's the whole point: they *already* have the browser! How many
internet users do you know without a browser?

>> Andrei, this is 2004. Do you want me to say that Nagel, Scully, Spindler,
>> Amelio were pinheads? That some of the moves made by Apple in the 90s were
>> criminally stupid? This is old news.
>
> And exactly what has changed since?

Are you really serious?

(With that, I think we have overstayed our welcome on this list on Apple,
etc., so I won't return to this subject, unless it directly and specifically
relates to design.)

Ziya
Nullius in Verba

20 Nov 2004 - 12:09am
Andrei Herasimchuk
2004

On Nov 19, 2004, at 9:14 PM, Listera wrote:

> I believe in the web, I'm not ashamed to say. I've been royally pissed
> off
> at attempts by some companies to kill off the ecumenical nature of the
> web.
> I view Microsoft's insistent move to privatize the web experience into
> its
> proprietary platform with alarm.

But not when Apple does it? That's what Dashboard is (or at least
started out as)... and the notion there is after-the-fact compliance by
others to support Apple's extension of HTML doesn't change the fact
that there's more being added to the spec without much quality control
over what the current crop of browsers do to support a seven year-old
spec. When Apple wants to add more stuff, is everyone going to
constantly jump and say, "Oh please! More! Make more!" and then rush
around to backfill their specs. Where is the consensus in that
approach?

When Microsoft does it it's evil, but when Apple does it, it's for the
good of all of us?

BTW, I believe in "networked" applications and all that can be done
with them. But I have yet to hear a good argument that makes "web =
browser" or "web = HTML".

> I see return to fat, platform-specific, proprietary clients as evil. I
> see deliberate reduction of choice as
> immoral.

As long as there are different operating systems, no amount of
hand-waving or spec writing will make "platform-specific" not exist.
And this coming from a standards cheerleader like myself.

Further, a browser IS a client, required to adapt to the platform.
Rather than writing clients specific to just an OS, of which you'd have
three main flavors (MacOS, Win, Linux), by writing specifically for a
browser you have created an exponential problem: Firefox, Mozilla,
Opera, Safari and Internet Explorer running on TOP of the different
OSes.

It has been my experience that writing DHTML and browser based
applications gains you very little in the xplatform delivery goal due
to xbrowser delivery problems and costs you drastically in the user
experience delivery goals by limiting yourself to only what a DHTML
browser page paradigm limits you to. The fact that IT departments all
across the globe have convinced people to believe that browser based
apps are somehow required in order for IT to do its job just shows how
little people understand how to run computer systems. It doesn't mean
that browser based apps are inherently the best way to build software.

What's the difference between a "link" saved as a favorite in a browser
somewhere that goes to a web site versus a shortcut icon in the Dock or
Start menu that launches a client app?

Answer: Little. Sure, technically, there's a large number of
differences, but at the end user level, none of those technical
difference mean anything.

Where does this perception that the "fat" client is somehow evil
because well... it's a fat client come from? (Whatever "fat" means.) Oh
wait, what about Apple's entire innovative offering of iLife apps?
Maybe I should dump iPhoto, iMovie, iDVD, Mail, GarageBand and iTunes
for browser based equivalents because fat clients are evil. and
especially because they are MacOS specific. Maybe I must return my Mac
G5 and demand Apple stop that sort of development because why on earth
would I want to use GarageBand in a fat client when obviously I should
do that sort of things as browser paged interface.

> So I feel passionate about a collection of technologies that will allow
> potentially millions of nonprogrammers to create cool and
> project-specific
> apps that are xplatform. Ongoing negotiations willing, they may have
> all the
> polish you see on Dashboard widgets, and more, on all platforms.

Glad you feel passionate about that. 99% of people I know would prefer
to just use their computer, not spend all their time tinkering with yet
another DHTML clock/calendar combo that they created. And that includes
many designers I know.

Further, how will this Dashboard stuff work on Windows and with IE
specifically? While Dashboard widgets are cool, they require
programmers to do the really cool things to make them useful. Outside
of hat, Dashboard widgets require that display layer to make them
useful, otherwise, you'd have a ton of active windows, which translate
into a lot of taskbar buttons to keep all those "pages" on all the
time. Not elegant at all.

>>> Andrei, this is 2004. Do you want me to say that Nagel, Scully,
>>> Spindler,
>>> Amelio were pinheads? That some of the moves made by Apple in the
>>> 90s were
>>> criminally stupid? This is old news.
>>
>> And exactly what has changed since?
>
> Are you really serious?

Uh... yes. That's why I asked. And by the way, a lot of decisions made
by Jobs are pinheaded as well. It just happens Jobs is good at covering
those sorts of things with new products every two years instead of
making the ones he starts evolve gracefully into true usefulness.

> (With that, I think we have overstayed our welcome on this list on
> Apple,
> etc., so I won't return to this subject, unless it directly and
> specifically
> relates to design.)

So now you're avoiding the point to be clarify how the old Apple of the
90s is no longer there with the new Apple of the last five years when
asked to explain. Color me not impressed, but fine, we can move on to
other more pressing topics.

Andrei

20 Nov 2004 - 1:08am
Listera
2004

Andrei Herasimchuk:

> That's what Dashboard is (or at least started out as)...

No, it's not.

> and the notion there is after-the-fact compliance by others to support Apple's
> extension of HTML...

Are you aware of all the collaboration by various parties involved in
developing these, some of it has even been conducted and written up on the
web several months ago, or http://www.whatwg.org?

> BTW, I believe in "networked" applications and all that can be done
> with them.

So do I. I'm building one right now.

> But I have yet to hear a good argument that makes "web =
> browser" or "web = HTML".

You won't hear that from me, unless you put words in my mouth.

Incidentally, web browsers render not just HTML, but XML, Flash, QuickTime,
Real, Windows Media, PDF, MP3, Word, Java, SVG and gazillion other formats
and doc types.

> Further, a browser IS a client, required to adapt to the platform.
> Rather than writing clients specific to just an OS, of which you'd have
> three main flavors (MacOS, Win, Linux), by writing specifically for a
> browser you have created an exponential problem: Firefox, Mozilla,
> Opera, Safari and Internet Explorer running on TOP of the different
> OSes.

This is the reason why common standards are important and something like
Firefox is welcomed.

> What's the difference between a "link" saved as a favorite in a browser
> somewhere that goes to a web site versus a shortcut icon in the Dock or
> Start menu that launches a client app?

Download time. Access from different computers. Familiarity. Perception or
reality of security. UI conventions. Integration with the rest of web
experience. Access to browser plugins. Etc.

> Where does this perception that the "fat" client is somehow evil
> because well... it's a fat client come from?

The historical experience we've had with fat clients was such that they were
silos. Their data and their state were often hidden behind a binary format
wall. They were black boxes.

> what about Apple's entire innovative offering of iLife apps? Maybe I should
> dump iPhoto, iMovie, iDVD, Mail, GarageBand and iTunes for browser based
> equivalents because fat clients are evil. and especially because they are
> MacOS specific.

iPhoto, iMovie, iDVD, and GarageBand are content production desktop apps.
They are neither networked, nor (fat or thin) clients to anything. Mail.app
can easily be supplanted by GMail. iTunes is a more interesting hybrid case
whose full functionally and fluidity because of what it does and the DRM
issues involved would be extremely difficult to cover in some other way as
effectively. It's interesting to note that iTunes, a networked app, is NOT
Mac OS-specific, though.

> Glad you feel passionate about that. 99% of people I know would prefer
> to just use their computer, not spend all their time tinkering with yet
> another DHTML clock/calendar combo that they created. And that includes
> many designers I know.

Huh? Who says (non-designer/developer) users have to do anything?

> So now you're avoiding the point to be clarify how the old Apple of the
> 90s is no longer there with the new Apple of the last five years when
> asked to explain.

I have no problem explaining it, this just isn't the appropriate forum.
Whenever you're in NYC next, let me know what your favorite beverage is and
I'll buy it.

Ziya
Nullius in Verba

20 Nov 2004 - 3:41am
Andrei Herasimchuk
2004

On Nov 19, 2004, at 11:08 PM, Listera wrote:

>> That's what Dashboard is (or at least started out as)...
>
> No, it's not.

Yes it did. Dashboard started out as the Safari team creating some cool
things with extension to HTML. Dashboard was never intended to be
anything outside of Apple until people got upset Apple was even
thinking about adding proprietary tags to the soup that has become web
technologies and mark-up.

>> and the notion there is after-the-fact compliance by others to
>> support Apple's
>> extension of HTML...
>
> Are you aware of all the collaboration by various parties involved in
> developing these, some of it has even been conducted and written up on
> the
> web several months ago, or http://www.whatwg.org?

Yes. but it STARTED as a Safari specific thing. It's now being
addressed after STARTING as a Safari only technology, then a lot of
people complained and the Safari backed up and got some feedback on the
thing. But if Microsoft did the same thing, would you have the same
reaction?

>> But I have yet to hear a good argument that makes "web =
>> browser" or "web = HTML".
>
> You won't hear that from me, unless you put words in my mouth.

Sorry. You said HTTP+HTML.

> Incidentally, web browsers render not just HTML, but XML, Flash,
> QuickTime,
> Real, Windows Media, PDF, MP3, Word, Java, SVG and gazillion other
> formats
> and doc types.

Bowsers DO NOT render Flash, QuickTime, Real, MediaPlayer, PDF, MP3,
Word or SVG. Plug-ins do the rendering work. To claim browsers do all
that is just ludicrous, Ziya. The plug-ins do all the work. All
browsers do is pass a rectangle port to the plug-in and say "render
whatever this stuff is here please." The browser have nothing to do
with rendering that content, and to imply so is to take away a lot of
hard work on the part of engineers who wrote those plug-ins by
attributing their work to the browser guys.

> This is the reason why common standards are important and something
> like
> Firefox is welcomed.

No disagreement from me on that issue. But reality is not that way
currently, and is a long ways away from that.

> Download time. Access from different computers. Familiarity.
> Perception or
> reality of security. UI conventions. Integration with the rest of web
> experience. Access to browser plugins. Etc.

Download time? A well designed client would take as much download time
as logging in. It really depends on well it coded. Further, even if the
initial hit is one minute to download an install, that saves a lot of
back-end time hitting pages for constant refreshes.

Access from different computers? That's a lesser argument, as most
people still access things from no more than 2 computers at most. A
laptop being amoung them, which they take with them. There are times
you're on a different computer, but those times are overstated in the
cost v. benefits of this discussion.

Familiarity? On what grounds? How is a browser based app inherently
more familiar than a client app? Both require good design to be
familiar.

As for security... you have to be kidding. Since when do people trust
web sites over client apps? In fact, I'm sure you could make the
argument people feel more secure inside client apps. (I know I do, at
least with something like iTunes workflow and approach.) And the poker
apps out there prove just how much people are willing to trust client
apps since billions are now being made a year with online poker.

UI Conventions? Again, the web world of browser apps is chaos in actin
with regard to following coherent UI conventions and client apps have
the upper hand here.

Integration with the web experience? I'm not sure what that means.
Especially since I use four or more other client app with my browser at
the same time. Heck, I even prefer reading RSS feeds via NetNewsWire
and its new HTML rendering tabs than dealing with plug-ins inside
something like Firefox.

Access to browser plug-ins? Not really. All the major plug-ins were
created from client apps, and libraries are readily available for all
that stuff if you didn't want to use the web OS services to gain access
to them.

> The historical experience we've had with fat clients was such that
> they were
> silos. Their data and their state were often hidden behind a binary
> format
> wall. They were black boxes.

And this is different on the web how? Last time I checked, I had
something on the order of 20 different accounts for the 20 different
web things I do, like buy tickets, books, music, read Salon, etc.
Further, data has been shared with many client apps for years, where
word could read Excel files, XPress could read Photoshop TIFs, CAD
programs used DXF... etc. One could make the argument my client apps do
a better job of sharing data and workflow than any of my web apps do.

> iPhoto, iMovie, iDVD, and GarageBand are content production desktop
> apps.
> They are neither networked, nor (fat or thin) clients to anything.

Networking iPhoto, iMovie, iDVD and GarageBand are fairly straight
forward. The question is to what degree they should be connected and
networked. As in, what sorts of things would someone want to do with
others in that regard. Many apps are already heading down that road, so
we'll soon see more networked features in these apps.

> Mail.app can easily be supplanted by GMail.

Maybe it could, but I would never tolerate GMail over Mail. There's no
point, the experience of using a browsed based email client is crippled
compared to either Outlook or Mail.

> iTunes is a more interesting hybrid case
> whose full functionally and fluidity because of what it does and the
> DRM
> issues involved would be extremely difficult to cover in some other
> way as
> effectively. It's interesting to note that iTunes, a networked app, is
> NOT
> Mac OS-specific, though.

Except that Apple tries and push its Apple theme onto Windows, making
it look like some bastard child on the Windows system. But yes, iTunes
is a classic case of a great client app that would have a very
difficult time being a browser based app. iTunes style apps are exactly
the wave of the future, imho. Not browser based apps.

> I have no problem explaining it, this just isn't the appropriate forum.
> Whenever you're in NYC next, let me know what your favorite beverage
> is and
> I'll buy it.

Wine. Very expensive wine.

8^)

Andrei

20 Nov 2004 - 4:57am
Listera
2004

Andrei Herasimchuk:
> Yes it did. Dashboard started out as the Safari team creating some cool
> things with extension to HTML. Dashboard was never intended to be
> anything outside of Apple until people got upset Apple was even
> thinking about adding proprietary tags to the soup that has become web
> technologies and mark-up.

Again, as I said, the point here is not Apple or even Dashboard. It's the
ability of nonprogrammers and scripters to create cool xplatform web apps.
Apple's iteration of this notion, Dashboard, is months from shipping.
Dashboard developers have been reaching out to the rest of the community and
have joined with others to coordinate the involved technologies for months
now. The other minutiae doesn't frankly interest me all that much.

> Plug-ins do the rendering work. To claim browsers do all
> that is just ludicrous, Ziya.

Geez, Andrei, give me some credit, unless you think I'm a complete moron.
I'm referring to the users' perception of what a web browser does for them,
not how plugins technically work. For all practical purposes it is the
Internet for them. Everything plays/renders/shows inside the browser. That's
the beauty of the browser, the ubiquitous client, for rendering, OK
displaying, everything.

> Download time? A well designed client would take as much download time
> as logging in.

Please, some clients will have to load an entire VM, runtime, shared
libraries, components, etc., even before they get to your app-specific,
optimized code. That'll be a bit longer than simple log in or revocation of
a stored state.

> It really depends on well it coded. Further, even if the initial hit is one
> minute to download an install, that saves a lot of back-end time hitting pages
> for constant refreshes.

Again, well-coded web apps like GMail, to use your ruse, do NOT require
constant page refreshes and are as fast as most other apps.

> Familiarity? On what grounds? How is a browser based app inherently
> more familiar than a client app? Both require good design to be
> familiar.

Half a billion people have experienced the web browser.

> Since when do people trust web sites over client apps?

If you're not using IE this has not been a problem, but that's another
thread. Are you saying that people will suddenly feel comfortable about
actually downloading executable code on their machines an order of magnitude
more times than now, when that becomes fashionable?

> And the poker apps...

Enough already, I am not playing those darn things yet.:-)

> UI Conventions? Again, the web world of browser apps is chaos in actin
> with regard to following coherent UI conventions and client apps have
> the upper hand here.

I doubt that very much. People grow familiar with their browser environment:
the back button, address bar, toolbars, favorites bar, bookmarks panel, etc.
Even when the contained web apps have varying UIs conventions, there's at
least that familiar container environment surrounding it.

> Integration with the web experience?

Something as simple as book marking a place from within a browser based app
or GMailing a Google search URL to a friend.

> Heck, I even prefer reading RSS feeds via NetNewsWire and its new HTML
> rendering tabs than dealing with plug-ins inside something like Firefox.

Me, too. But you know what RSS is and what a newsreader does. There are 1/2
billion people online.

> Access to browser plug-ins? Not really.

Unless you incorporate a browser rendering framework like IE or WebKit,
integrating all the plugin functionality is nontrivial work. And if you do,
well, you're back to the browser again.

I actually designed and coded two such apps several months ago. I don't
remember now on what mail list I described what I did (it may have been
here). One of them talked to two web services, parsed XML, stored some data
locally, some remotely, had a datagrid, a WebKit area, and a user
configurable XSLT transformation engine for output. So you are preaching to
the converted.

>> They were black boxes.
>
> And this is different on the web how?

For starters, you can do View Page Source and actually SEE the incoming raw
data and copy it out.

> Networking iPhoto, iMovie, iDVD and GarageBand are fairly straight
> forward.

But not done.

> iTunes style apps are exactly the wave of the future, imho. Not browser based
> apps.

Why on earth is this an either or situation?

This whole thread started in 1952, I think it was, when I blasted off on the
notion of web's dead. Web's not dead. It's not dying. Stuff like Dashboard
will make it even more useful and usable.

That doesn't mean that networked apps aren't a good idea. Some people here
know me as an advocate of that notion for years now. Since I had been
designing and developing all kinds of apps way before the web, I have no
problems with that. But web's not dead or dying, which, for the weary, has
been my principal and hopefully concluding point.

> Wine. Very expensive wine.

What do you expect from a guy who can't live with simple HTML. :-)

Ziya
Nullius in Verba

20 Nov 2004 - 2:24pm
david gee
2004

Andrei Herasimchuk wrote:

> Further, how will this Dashboard stuff work on Windows and with IE
> specifically? While Dashboard widgets are cool, they require programmers
> to do the really cool things to make them useful. Outside of hat,
> Dashboard widgets require that display layer to make them useful,
> otherwise, you'd have a ton of active windows, which translate into a
> lot of taskbar buttons to keep all those "pages" on all the time. Not
> elegant at all.

See: www.konfabulator.com

--
david gee david at mode3.com http://www.mode3.com/david

21 Nov 2004 - 7:41pm
Pradyot Rai
2004

This has some good opinions about Dashboard, and Apple in general --

http://www.igeek.com/articles/Interface/Dashboard.txt

If some opening lines throws you off, go start from the bottom - from
disclaimers.

Prady

Syndicate content Get the feed