Wednesday, November 8th, 2006

Is Converging Towards the Desktop Good?

Category: Editorial, Usability

Aza Raskin spoke at The Ajax Experience about the desktop being dead. The talk was entertaining, and he kindly posted his slides.

He has followed up the talk with a question: Is Converging Towards the Desktop Good? in which he takes the side of “No.”.

He comes out against recreating windowing toolkits in JavaScript, and instead embracing the web-way, and thinking outside of the desktop box to come up with a more humane interface:

In 2004, Google chose to use one nascent technology, Ajax, to create an e-mail service: since there didn’t exist any Ajax toolkits that allowed them to reduplicate the desktop on the web, they were constrained to think simply, “how can we work with Ajax and the web to make email humane?”

Their answer was something that was actually more humane than any desktop e-mail client already in existence. What’s even more interesting is that traditional desktop developers had long been able to create an email client as humane as Gmail–but they never did, because UI toolkits made it so easy to create something that was familiar, that was the same, that was inhumane.

You cannot be better without being different.

The desktop-like web toolkits being developed today endanger innovation by entrenching us in familiarity of the past. We need to remember that there is something better than the desktop in many of today’s web applications, and we need to carry this innovation with us as we move forward to create new tools and new interfaces.

Do you agree or disagree?

Posted by Dion Almaer at 7:43 am

3.7 rating from 26 votes


Comments feed TrackBack URI

Must agree with Mr. Raskin – the desktop is dead. Where I work all local install applications have been end-of-lifed – web apps only.

When the big Vista ( all six or seven variants of it ) disappointment washes over the masses next year, no just us geeks will say its dead.

Comment by Todd — November 8, 2006

I concur. Fully.

Comment by Dan — November 8, 2006

I’m puzzled by his use of the words “humane” and “inhumane”, he seems to be using them in a way I’m unfamiliar with. I thought “humane” meant compassionate and to be “inhumane” is to lack the proper compassion. Is he arguing that Gmail is more compassionate than Outlook?

Comment by Richard — November 8, 2006

“desktop vs browser” is irrelevant – I just need to be able to read my email from any computer I am at, access my files, IM ppl, write and compile code, etc etc. Caring too much about “it has to be done in the browser” might detract from focusing on solving the actual problems – helping me do stuff.

Additional thought:
Perhaps Javascript is the new Java and instead of Applets allowing Java to be run in the Browser, we need desktoplets to allow Javascript to run on the desktop (e.g. Adobe’s Apollo – I think that’s what its called).

Comment by Dr Nic — November 8, 2006

I don’t agree. If you think about it, you’ve got two types of developers. Desktop and web. Each comes from a completely different paradigm in terms of development.

“You cannot be better without being different.”

He’s right there, but the difference is the paradigm, not the platform. I really don’t understand what is so different from running a site in a browser, or in a VM-like system, as like with Adobe’s Apollo. Both platforms bring down client-based code and execute, no matter where the code resides (server or hard drive).

As for endangering innovation… it sure doesn’t seem to have stopped Google or Yahoo. The fact of the matter is that both kinds of developers might actually learn something from each other, and that certainly provides for innovation.

Comments like Aza’s will only work to create a division between developers and that’s just not going to cut it. We deal with it in politics, we don’t need it here.

Comment by Andrew Herron — November 8, 2006

I built a complex ‘desktop-like’ webapplication Without using windows it would be almost totally impossible to serve such functionality in an orderly way. If I had to, that would be inhumane! :)

Comment by Camiel — November 8, 2006

Having now watched the slideshow I’m now more confused about what the presenter is trying to say. If anything, he needs a new desktop model not to get rid of the desktop fullstop.

Comment by Dr Nic — November 8, 2006

I don’t see the tie between the two. Many of the toolkit widgets are quite innovative, and expand on how the web interface can be used in a way the desktop never was. For instance Google’s Gmail interface was better than other web mail interfaces for a while, but now I would say it is inferior to Yahoo’s – which they built on their toolkit. IMHO, the drive towards excellence continues.

Comment by Jeff — November 8, 2006

If desktop apps are not the way to go, I can’t wait to see the Web 2.0 version of Maya, After Effects, Pro-tools or some deeply complex desktop inhumane converged application.

Tell me, what is a “humane” and “inhumane” software again? For me, they’re all inhumane… Maybe they should use a different adjective for this concept, like: ergonomic.

Comment by Hélio Miranda — November 8, 2006

I don’t know if I totally agree with the writers view. I do agree on the idea of embracing the “web” way of thinking.

However, there are benefits to desktop apps that we must consider. Taking advantage of hardware is still is important. Unfortunately we can’t all be online 24/7. So, something like Google’s Writely is great for writing a report online, however what happens when you disconnect?

College Kid (In front of computer): No connection to my internet…. means no connection to my application, oh @#%t my reports due tomorrow! My internet company is having technical difficulties! :)

Apollo seems to be the best of both worlds. If you build a web app and you can use the same knowledge to leverage the technology to build useful desktop apps then that is awesome! Both application types can coexist.

Comment by Rahsun McAfee — November 8, 2006

If we get rid of the desktop how will you run your browser? I’d like to see someone write Doom 3 in JavaScript!

Comment by Justin — November 8, 2006

I disagree with such “Desktop is dead” opinion.

Desktop has lots of features that current web applications are lacking of, like dialog, tree, sortable and editable table, and others. People converging toward desktop are for rich features on purposes.

I am developing Java2Script (, which aiming developing RIA in desktop (Java) way, and converting SWT desktop application into RIA upon JavaScript. The reason why I started such a project was that I found that web applications were requiring more and more richer features, which was already implemented in desktop widgets, or could be implemented easily with existed desktop widget libraries, such as Eclipse SWT.

“You can not be better without being different.”

“Different” also means “new” and requires time to mature. So from pragmatic perspective, converging towards desktop will help people to improve their efficiency, as people are familiar with desktop enough. Experience should be less important than working efficiency in my opinion. Only upon efficiency is promised, and enough experience of old widgets is gained, should we develop new more humanized ways.

Current web browsing experience is good enough and but it never means those already existed also good and efficient desktop applications, like Outlook, should disappear. Good things alway last long. So why not mixing both web browser and desktop windows together and let it be?

Comment by Zhou Renjian — November 8, 2006

You people are daft, at best. No one said to remove the desktop – no one is that dumb. No one is advocating online applications over desktop ones. No one has suggested to create Doom or Maya or anything else in a browser. If you read the article it asks whether or not the design/functionality of web applications should be made in the image of existing desktop applications. Period. Does doing so (mimicking the desktop on the web) restrict innovation in a product? That is the point of the article. The short answer is that it doesn’t have to – anything you can do with a connected web application can be done with a connected desktop application. The long answer is that desktop developers likely have a defined set of tools based on their environment and their target audience, whereas a web developer may have more flexibility with both. Read the damn article – don’t just hit submit everytime you see the words “browser” and “desktop” in the same paragraph.

Comment by Dan — November 8, 2006

Not only should they not converge, there are areas where the web browser cannot replace the desktop – i.e. they are incompatible.

I also really dislike that he has hijacked two emotional-sounding words (humane and inhumane)… it’s the type of psycho-analytic mind-games that make me feel like I’m being sold something and give me the urge to simply walk away.

Comment by Matt Nuzum — November 8, 2006

I just want to point out that you don’t need apollo to run javascript on the desktop, look at the windows script host.

Comment by Snootz — November 8, 2006

I thought the quote from Marissa Mayer was great. I think people are fair more creative when they are constrained. However, I am not quite sure what that has to do with web applications behaving more like desktop applications.

My feeling is that web applications are converging to desktop-like applications whether we like them to be or not (i.e. gadgets/widgets from Google/Yahoo! Konfabulator, Flex 2.0, offline access in Firefox, etc.). However, I am not sure how that hinders creativity when going from the web to the desktop model.

Comment by Oliver Tse — November 8, 2006

I agree that Single-Page-Desktop-Style applications are probably not the best metaphor for creating application, especially in a browser. The problem is that there are currently billions of people using desktop style applications and moving the human race off of this is not only time consuming but expensive.

Ajax provides better use experience and allows users to replicate desktop application and deploy it over the web. Developers and application provides will not take advantage of all the new design capabilities when migrating their applications. You will see many applications make use Ajax or other RIA technologies to replicate Win32 or Swing applications down to the last detail. The reason is it is extremely expense to retrain large groups of people on new UI design metaphors.

For those who have the luxury of create an application from scratch, it may be an option to explorer new designs, but you need to look at the usage model. Are people heads down, key board only, what are the accessibility requirements and so on?

Take a pragmatic approach is always best when dealing with usability.

Comment by Robert Buffone — November 8, 2006

the right tool for the right job..
why force the users to learn new interactivity, new UI etc.?
now the line between web and desktop applications started to fade, so in my opinion it’s better be coherent at maximum for the UI design..
for instance :
sorry it’s in bulgarian, but I think the design speak quite much for itself

just my 2 cents

Comment by Stoyan Pedev — November 8, 2006

I don’t agree.

There are well documented methods/conventions relating to the process of creating any human-computer interface. A good interface is transparent (doesn’t get in the way of a user accomplishing their task(s)) (and it makes no difference whether it runs in a BRE or any other “runtime”).

For instance, a Java Swing app (with a windows L&F) will look predominatly same as another Swing app (with the same L&F) it doesn’t mean that the Swing Toolkit is bad – what it means is that particular software house may not have done any user testing (and used the Toolkit “straight out of the box”) or that what they did do, resulted in their “target audience”, “average user” (or whatever you want to call “them”) wanting something that looked “straight out of the box”.

In a similar vein, Flash was not 99% bad – 99% what developers/designers produced using Abobe Flash was bad.

If user testing shows that an Accordian Component used to display a certain type of information is more usable (faster, easier to remember, etc, etc) than a Tabbed Component, then so be it or that navigating a file system using a “Component” that looks like you’re travelling down a worm hole to a different part of the galaxy is more usable thats great too.

Desktop-like web toolkits that are being developed today DO NOT endanger innovation – bad or incorrect software development practises do.

Comment by Wolfman — November 9, 2006

Unfortunately, most of the commenters are totally and utterly missing the point. Hint: You can’t figure it out by simply reading the presentation. You would at least have to *see* the presentation. Aza is talking about concepts which were refined during (literally) decades of research by Jef Raskin. If you want to find out what Aza is talking about and what he means by “humane,” read Jef Raskin’s book “The Humane Interface.”

Comment by LKM — November 9, 2006


Aza’s point (or the point of the extract that’s being commented on) was that “The desktop-like web toolkits being developed today endanger innovation by entrenching us in familiarity of the past. ” I stand by my view “Desktop-like web toolkits that are being developed today DO NOT endanger innovation – bad or incorrect software development practises do.”

Comment by Wolfman — November 10, 2006

I’m all for having to take a test before being allowed to post.

Comment by Dan — November 10, 2006

Wolfman: They both do.

Comment by LKM — November 10, 2006

LKM: Nope. They don’t.

Whether a brower-based application is developed using, Echo2, Dojo, YUI, GWT, etc, etc, does not mean that it isn’t possible introduce “new things” or “methods”. It is the mindset/processes of the developer(s)/software house(s) that can stop innovation – this doesn’t make gui toolkits bad – just like Flash isn’t 99% bad. Toolkits are there to speed up development (where possible) – but it is the choice of the developer(s) to sacrifice “innovation” (with respect to HCI) e.g. “I’ll not bother using any focus groups, paper prototyping, user testing, etc, etc I’ll just copy V 1.0 source and add a new button, we need to get V 2.0 out by January…” – the toolkit doesn’t force the developer to do anything.

Comment by Wolfman — November 10, 2006

It depends. Doesn’t it always?

I like to have my data locally–that way I know I can get to it. If my data is on somebody else’s server, they may be down for maintenance or due to, gasp, some attack/occurance.

Sure, the mail server could be down too, but one less potential problem is a good place to start in my book. The Software as a service model is good marketing right now, but my data on Basecamp helps me not if a hurricane hits my area and I can’t connect to it…

I’ll just download my e-mail (in text format please) so I can deal with it on my machine. I’ll keep my other data local. However, if you don’t have to have the data (personal e-mail maybe?), then network storage is fine.

It is all relative..

Comment by Counsel — November 10, 2006

This is not about desktop vs. web. It is about tailoring web functionality to mimic desktop functionality, and whether or not that is a good thing. Am I in the same room as you? Are we reading the same words. Check your browser settings, you might have the “Upside-down and off topic” language setting enabled.

Comment by Dan — November 10, 2006

Wolfman: Dunno about you, but most devs have a budget and limited time. If you can do something the easy way (i.e. with an existing framework), you do. This has got nothing to do with mindset. It’s just the realities of the IT market.

Comment by LKM — November 13, 2006

LKM: Precisely! Nothing to do with the Toolkit of choice!

The software company I work for is exactly the same – budget/time over “proper” proceedures/methodologies.

If the developers here stopped using Echo2/GWT/YUI and said “We’re going to create from scratch, but it’ll take us a year to learn JavaScript, DOM, CSS and all the browser differences, plus another year of development doing market research, usability studies, focus groups, paper prototypes…” they’d be out on their ears as manangement think that:
1. because of the longer dev time our software will be behind that of the competition
2. all that touchy-feely guff is only there to give strategic product/marketing companies something to do
3. they’ve been developing software this way from year dot and no customer has ever complained our that software isn’t “humain, “usable”, “user-freindly” or whatever the latest fad is
– but that’s not the fault of the Toolkit(s) of choice it’s the mindset of the software company…

Comment by Wolfman — November 13, 2006

free craps online

In need of free craps online

Trackback by craps online — November 21, 2006

I agree that software as a web service is big, I really like having my google spreadsheets available where ever I am.

Comment by Justin — May 3, 2007

Leave a comment

You must be logged in to post a comment.