views:

567

answers:

10

Although somewhat related to this question, I have what I think is a different take on it.

Is a desktop app that has no connections to the "cloud" dead? I believe that some things are going to continue to be on the machine (operating systems obviously, browsers, some light-weight applications), but more and more things are moving to network-based applications (see Google Docs for office suites, GMail and other web-email clients for email, flickr for photo management, and more).

So other than the lightweight applications, is there anything that, in 5 to 10 years, will continue to be (either out of necessity or just demand) remain on the desktop and off the cloud?

+3  A: 

IDE's will probably be "off the cloud" for a long time, if ever... powerful customizable editors like Emacs will also probably stay "off the cloud" for a while.

Mike Stone
if a long time is 10-15 years, I agree. But it will happen.
Tony BenBrahim
IDEs are one of the first things I'd want on the net. NetBeans, for instance, already has some collaboration features. It'd great to be WebStartable. Just launch the app from any (safe) machine and get to work with colleagues.
Tom Hawtin - tackline
+2  A: 

There's no reason that many corporations will move to an online system simply because of security concerns.

For example, One of the greatest assets of Outlook is to go offline and continue working. Sure Google Gears has similar functionality, but then you're trusting Google with your corporate security.

enigmatic
+4  A: 

Video editing and other resource intensive tasks will probably stay off the cloud for a long time.

Geoff
I thought about that. Then thought about PhotoSynth. That's got to be some heavy image processing, although static images...
Thomas Owens
Resource intensive tasks are perfect for the cloud as you can distribute the processing across hundreds or thousands of nodes
John Channing
You can also do CPU intensive tasks in runtimes such as the JRE. Not everything needs be ECMAScript.
Tom Hawtin - tackline
@John, yes you can distribute the processing but you also have to distribute the data.
MarkJ
+3  A: 

If I look at the application that we've selling and at the applications I've written as a consultant, I must very much agree with you. Most of them are useless if there is no internet connection. Some do work in disconnected mode, some don't, but all of them are pretty useless if you cannot connect to the big supporting system hidden far far away.

On the other hand, I wouldn't want to say that everything will move into the cloud in 5 years. Too much work with porting. There will be desktop applications that will function as a thin and offline-able client (just like, for example, Google Reader does if you install Gears) and there will be fully "clouded" :) applications.

I have no idea what will happen in 10 years. If I put myself 10 years back (and that is very easy to do as I was writing a lot for a local computer magazine in that time), I totally couldn't predict how the computing will become internet-dependant in 2008.

gabr
+1 Don't believe the conventional wisdom and the conventional predictions. Slightly off topic. 10 years ago everyone was predicting that "content would be king" and you'd only make a profit from the Internet if you owned content. Then came Google, YouTube, Facebook, and this place.
MarkJ
+2  A: 

Such applications are dead since 15 years, ever since Sun took market leadership with their JavaStation.

No, wait. They did not. And things are not "more and more" moving to network-based applications. Sure, there is Webmail, but even GMail is FAR away from the comfort of modern Outlook or Thunderbird Clients. Same for office. Google Docs is a nice toy for ocasional use, but it's vastly inferior to conventional Office suites.

The Desktop is not dead and it will not die anytime soon. Internet Applications are alternatives in some situations, but be are just starting getting proper functionality and performance. Let's face it: JavaScript performance is still a Joke, the IDE Support is not there yet and Browsers are too unstable at the moment.

Google Chrome, IE8 and Firefox 3.1 start to go in a better direction, but it will take years for them to be mature enough to create JavaScript applications that actually can fully replace desktop apps. But that would require some proper standardization accross browsers, and we all know that this will not happen before the next millennium or so.

Michael Stum
Not sure I agree with "FAR away from the comfort of modern Outlook" or "vastly inferior to conventional Office suites".I gave up MS Office a couple years ago for OpenOffice, and I've now given that up for Google docs. I've never looked back. I've not once wished for MS Office
Ben Collins
And I still use Word to write my Blog Posts :-) MS Office 2007 is actually quite good, but it's overkill for small lists and documents, yet in my corporate environment it's lightyears ahead of Google Docs. OpenOffice is also a good alternative, that then depends.
Michael Stum
+2  A: 

About 1% of users actually use Google Docs&Spreadsheets full-time. Almost all of the rest use Microsoft Office. So, no, off-the-cloud applications are not dead simply because a Google office suite exists. And those are, really, the only high-profile true web applications out there that are meant as desktop app replacements.

Webmails are a special case though. It actually makes sense to use those rather than a desktop app, since your email is next-to-useless without a connection anyway. But most applications don't NEED a full-time Internet connection. A word processor certainly doesn't.

What will definitely remain on the desktop:

  1. Games
  2. Small apps (calculator, notepad type of stuff)
  3. Anything that generates data that needs to be secure (I don't imagine tons of people or companies want to trust their accounting details to Google, for example)
  4. Web browsers (obviously)
  5. IDEs (Visual Studio via Ajax? Come on...)
  6. Auxiliary development tools (SVN, etc), since good security policy would forbid their use through a web browser
  7. Anything that needs high enough performance that network latency would be an impediment

What will probably remain primarily on the desktop, at least for the next 5 years:

  1. Office tools (unless web-based limitations can be lifted... which would require much better-performing web browsers than we have now)
  2. Photoshop and such tools
  3. Chat clients (web-based equivalents are disappointing so far)

That's not to say that any of the above cannot have an Internet-based component, of course.

DannySmurf
+3  A: 

Gosh, I hope not as that's my job.

The main piece of software I write controls electronic hardware (PXI boards and the like) for testing. Without "real" hardware, there's nothing to test. Even the very nature of the tests themselves prevent simultaneous access (once you set the state of a switch, you don't want someone else moving it).

So as long as you interact with any hardware, you're off-the-cloud.

Oh, and some companies have security issues with being on the Internet; I'd say security would also drive desktop apps with no connections.

Richard Morgan
A: 

Well, the obvious examples would be graphics intensive games.

Note that all that ajax-y applications aren't that "lightweight" to you computer as it looks/sounds like so, either. Compare Microsoft Office Words and Excel with Google Docs and Spreadsheets. With my average T7300 laptop, it takes 1-2 sec to load up the application while taking 7-10MB of the memory. Not possible with web browsers, especially when considering the way it's been heading to (trying to make web browsers rich client). IMHO, web apps can never match well-polished desktop apps in terms of usability and etc. Web apps are "web" apps, they are stuck in the square (the browser).

wbkang
+7  A: 

10 years or more ago this would have been, "Are non-internet applications dead?"

There's things the cloud does better than desktop applications, and in those places I'm sure non-cloud applications will become increasingly rare. But there's plenty of applications where you might not want to use the cloud, the benefits don't outweigh the costs, or the complexity just isn't worth it.

It's a new tool, and it's a better tool than desktop applications for many things. However, you don't throw away a hammer when you buy a screwdriver, you simply reserve it for when a nail needs to be driven.

OwenP
A: 

I personally will never leave my stuff on the web under someone else's control. All of my photos and e-mails I keep on local hard drives that I control.

I prefer to make my own stuff available to me through the web on my own hardware. The only way to have reasonable performance and be productive when offline is to use local apps.

To me the future will be local, but remotely accessible and synchronized. At least for the next 20 years or so.

Not only do I think it's not dead, I think it's the way everyone will want to go once we have a few disastrous failures (ie, websites disappearing with users content that isn't backed up anywhere or severe privacy breeches as some large company loses control of access to the data they are protecting).

bruceatk