Interview with Google Chrome Team

6 Sep 2008 - 6:59pm
5 years ago
6 replies
654 reads
Jared M. Spool
2003

For anyone interested in trying to revive the IxDA dead horse called
Google Chrome, Steve Gillmor had an excellent interview with the
product manager and UI developer. Lots of things we discussed here
were talked about in the interview, confirming my thinking on where
they are going.

Gillmor's synopsis: http://tinyurl.com/5bv645
The audio of the interview: http://tinyurl.com/64bcbz

Jared

Jared M. Spool
User Interface Engineering
510 Turnpike St., Suite 102, North Andover, MA 01845
e: jspool at uie.com p: +1 978 327 5561
http://uie.com Blog: http://uie.com/brainsparks

Comments

6 Sep 2008 - 8:33pm
Andrei Herasimchuk
2004

On Sep 6, 2008, at 4:59 PM, Jared Spool wrote:

> For anyone interested in trying to revive the IxDA dead horse called
> Google Chrome, Steve Gillmor had an excellent interview with the
> product manager and UI developer. Lots of things we discussed here
> were talked about in the interview, confirming my thinking on where
> they are going.

They keep saying they want people to make cool web apps, but that they
are still going to operate fundamentally as Web Browser #4, they'll be
OpenSource and they'll follow web standards. I have to believe they
are being honest. Nothing tells me they are trying to be devious here.

So... The only way I think they can begin to believe that premise is
that they also believe that an SDI application model is sufficient for
everything that can be considered a "web application." That Google
Docs, Spreadsheet, Maps, and a whole host of certain kind of apps can
be sustained in single window interfaces and be completely self-
contained.

That's where the breakdown occurs for me. Web apps currently work in
an SDI mode, and a fairly limited SDI mode at that. You can't take
over the keyboard interaction, you can't make floating palettes or
slave windows that are aware of each other to pass data via a common
pipe, and you can't do other things like use OS alerts, OS dialogs,
etc. And web apps in that SDI model have to worry abut the address
bar, the back button, and other "browsing" interactions resident
inside a web browser that have nothing to do with more tool oriented
application interactions.

Given all of that, they are basically building Browser #4, and all
innovation will stop there, or at least innovation done there will be
done across the browsers, and nothing will be done for Chrome
specifically. Their route is certainly legit (even Photoshop Lightroom
works largely in an SDI conceptual model so that type of interface
approach can certainly do a lot if the task at hand is reasonably
specific), but in going this route, it will be clear that the RIA+
route of AIR will be very different, as the RIA+ route will head back
towards more fully fleshed out little desktop applications.

Who will win? Not sure I care. I don't pick sides in these sorts of
things. I just design what I have to for whatever I'm asked to do it
for.

But there is a big difference between Chrome being Browser #4 and
Chrome being a new application platform that happens to use OpenSource
web technologies but plans on making a clean break from being a
browser. As I'm sure you might have guessed... I wish they'd do the
latter.

If they did, we'd have a lot of choices going forward to make
software: traditional web browser for more service like applications,
richer web application platforms for more robust tool-like apps, RIA+
using proprietary tech for even more complex tool-like apps, and good
old traditional desktop application built right on the OS itself to do
whatever the heck you want.

But Chrome in its current trajectory is clearly not going to help me
with what we design anytime soon, as being Browser #4 will only mean
for me that things will largely be faster. It won't solve the problems
of trying to build multi-window, rich interaction based web
applications that just happen to use a lot of web technologies at its
base instead of a proprietary technology like Flex, et al. Ah well.

--
Andrei Herasimchuk

Principal, Involution Studios
innovating the digital world

e. andrei at involutionstudios.com
c. +1 408 306 6422

6 Sep 2008 - 8:59pm
Jared M. Spool
2003

On Sep 6, 2008, at 9:33 PM, Andrei Herasimchuk wrote:

> That's where the breakdown occurs for me. Web apps currently work in
> an SDI mode, and a fairly limited SDI mode at that. You can't take
> over the keyboard interaction, you can't make floating palettes or
> slave windows that are aware of each other to pass data via a common
> pipe, and you can't do other things like use OS alerts, OS dialogs,
> etc. And web apps in that SDI model have to worry abut the address
> bar, the back button, and other "browsing" interactions resident
> inside a web browser that have nothing to do with more tool oriented
> application interactions.

Yah, yah, yah.

However...

First: Gears could do those things and does some of them already. One
of the things from the interview I found interesting was that Chrome
is optimized for Gears, but Gears is also a middle-ware package for
the other browsers. I think that's a fascinating strategy to make
things work.

Second: I'm betting that somewhere between 65% and 80% of the
applications that are built today *could* operate in an SDI model.
Granted, it's nice to have floating windows, but apps with those types
of interaction modalities are fairly advanced. UPS, for example,
doesn't need that kind of interaction model for their WorldShip app,
which is sophisticated in functionality, but straight forward in terms
of the demand on UI modalities. Even something as sophisticated as
Salesforce can get away with an SDI model for 90% of what people try
to do with it. (And Flash/Flex/AIR can provide the rest.)

Not everyone builds a sophisticated tool for manipulating artwork.
Many just build tools for manipulating customer data.

> But there is a big difference between Chrome being Browser #4 and
> Chrome being a new application platform that happens to use
> OpenSource web technologies but plans on making a clean break from
> being a browser. As I'm sure you might have guessed... I wish they'd
> do the latter.

There's a third possibility, which is what I heard in the interview:
Chrome is a stimulus for a competitive response by the other big
browser producers. It came out that Sergei Brin/Google would consider
Chrome a success if MS IE9 adopted the core components from the Chrome
open source set.

I think that's really where I think this is heading and why I'm
excited about it.

Jared

Jared M. Spool
User Interface Engineering
510 Turnpike St., Suite 102, North Andover, MA 01845
e: jspool at uie.com p: +1 978 327 5561
http://uie.com Blog: http://uie.com/brainsparks

6 Sep 2008 - 9:13pm
Andrei Herasimchuk
2004

On Sep 6, 2008, at 6:59 PM, Jared Spool wrote:

> Not everyone builds a sophisticated tool for manipulating artwork.
> Many just build tools for manipulating customer data.

Agreed. Like I said... Photoshop Lightroom is largely an SDI model
type of application, and it's pretty intense with the rich interaction
model it follows inside the confines of SDI. If you haven't used it,
go grab the 30 day trial download to see what I mean.

But there are plenty of examples that aren't. Some light, some heavy.

Instant Messenger applications are relatively "light" apps that work a
ton better outside the SDI confines. The WebSketch product example
would as well. Many industrial strength enterprise apps would benefit
immensely from being web technology based but built outside the SDI
confines.

We also helped design an enterprise application for Agile Software
(who were acquired by Oracle last year), which is a prime example of
something that needed a multiple window environment. The application
was a document management and project workflow product where you had
to track thousands of parts that go into building physical products.
We ended up having to do a lot of custom work to make it happen so
windows could talk to each other).

There a ton of enterprise level applications that are quite simply
*begging* for such an development environment.

> There's a third possibility, which is what I heard in the interview:
> Chrome is a stimulus for a competitive response by the other big
> browser producers. It came out that Sergei Brin/Google would
> consider Chrome a success if MS IE9 adopted the core components from
> the Chrome open source set.

I heard that too... but that doesn't change the "browser" paradigm, it
only really makes it more robust at the tech level, which helps it
somewhat at the interaction level, but not in the core mode it
operates. In other words, an intense, interaction rich product like
Lightroom could someday be built in that new platform with that more
robust engine, but in the end, it's *still* SDI.

--
Andrei Herasimchuk

Principal, Involution Studios
innovating the digital world

e. andrei at involutionstudios.com
c. +1 408 306 6422

6 Sep 2008 - 10:41pm
dszuc
2005

What if Chrome was simply a stepping stone towards a Google OS?

What if some of the principles in the Google apps to date suggest not
having to rely on the need for deeper functions?

One trend is to be able to serve up applications that have a few key
simple functions that people use more regularly with the ability to
switch on more features as needed.

rgds,
Dan

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=32718

7 Sep 2008 - 12:43am
Jared M. Spool
2003

On Sep 6, 2008, at 8:41 PM, Daniel Szuc wrote:

> What if Chrome was simply a stepping stone towards a Google OS?
>
> What if some of the principles in the Google apps to date suggest not
> having to rely on the need for deeper functions?
>
> One trend is to be able to serve up applications that have a few key
> simple functions that people use more regularly with the ability to
> switch on more features as needed.

What's interesting about the interview is that the product manager
said that the reason there is no Mac or Linux version yet is that
they've optimized the existing beta heavily to Windows.

He also said that Android, which *is* a Google OS, isn't using Chrome
for the same reason. They are using the same rendering engine, WebKit,
but have different UIs and components because their OS environment is
very different.

I thought that was very interesting and telling about Google's view.

Jared

8 Sep 2008 - 8:05pm
dszuc
2005

This is a good read --
http://www.informationweek.com/news/internet/browsers/showArticle.jhtml?articleID=210500091

"Google is impatient to give people better access to its massive
data centers and believes it needs to build a browser from scratch
with the goal of shifting the focus to running applications, not just
displaying pages."

rgds,
Dan

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=32718

Syndicate content Get the feed