Ten years from now, let’s see if I’m right: the browser is where our apps will live, not on the desktop. The browser will be extended to give us offline connectivity and access to desktop resources. The browser and its context provides what we need for mobile access. The browser is where it’s at, not the desktop. Who cares about iTunes, man (and woman)? Just play some music through my speakers; you can do that from my browser, right?
Consider:
- The most exciting innovations are taking place in the browser. Do you feel more power and productivity when you’re in your browser or in a desktop app? My desktop apps are satellites to my real work. I hate working outside the browser because I don’t have my favorite tools like the Performancing blogging extension or Zotero research manager or Delicious bookmarklet at the ready. I always come back to the browser. It lives on my main screen. I live on it.
- The only good way to develop cross-platform non-enterprise apps right now is with the browser. With browser-based technologies: HTML and JavaScript. Because face it, Java’s dead outside of big enterprise projects. Except maybe as a tool to generate HTML/JavaScript apps.
- The browser is an information integration platform. Not data integration–that’s for databases and data cleansing tools. Think widgets, extensions, bookmarklets, RSS. I can integrate all the information I want, in the way I want, using a browser.
- The browser experience can be made consistent even across Windows, Mac, and Linux. Once I have Firefox configured on a given machine, the experience of doing my work is largely the same across computers, unless I have to leave the browser, of course.
- The browser’s about me. I set the text size. I control the tabs and the number of windows. I turn JavaScript on or off. I add or subtract plugins and extensions. I decide what gets downloaded or not. It revolves around me, while desktop apps seem to want me to adjust to their view of the world and their view of my data.
I’ve been saying ever since I went to Adobe Max that I’m not sure what I think of Apollo. I finally experienced an epiphany.* Apollo’s going in the wrong direction, trying to bring browser assets onto the desktop. No, no, no. That’s all wrong. Take desktop assets and capabilities to the browser. Let’s extend the browser–like Microsoft and Mozilla are already doing. Let’s keep building better and better apps for the browser. Let’s produce tools that make it easier for developers who know Java or C# or PHP or even just HTML to create interactive and informational experiences. Please don’t make us go back to GUI development circa 1995. When an Adobe engineer talked about putting chrome onto Apollo apps that would look just like all the other desktop apps for a particular operating system, I felt like puking. I, for one, never want to have to think about how my application’s chrome looks. The browser chrome is fine.
Here’s a specific example of how browser is better than desktop, if you are yearning for something concrete. There’s this thing called internet relay chat. To access it you have to download a separate client app, configure it, and figure out its interface along with all the etiquette of IRC. What a pain. I don’t want any more applications on my computer. What is different about chat that it needs something other than the browser to work? Don’t talk to me about protocols; I don’t care; I just want to chat. As we find more and more ways of connecting and working, it’s all the more important that we be able to do it easily and immediately without having to install or learn new desktop apps.
*Seems relevant to my blogger relations thoughts that an email conversation I’ve been having with someone from Morfik led to this epiphany. That’s the kind of human-to-human interaction I’ve been arguing for.
15 Comments
I’ve tried using several Web 2.0 apps, and I keep going back to their desktop counterparts. I used Gmail for almost a year before switching back to Outlook (2007). I also used Yahoo! Mail beta for awhile, but in both cases I didn’t like not being able to maximize the app. Plus Yahoo! Mail tabs inside Firefox tabs is a bit much to manage. I had to pause to think about which tab I wanted to close.
As an author, I can’t ever imagine using anything other than a desktop app. I have to create a master doc, generate TOCs, and reference figures. I use WordPress for blogging, and I always have to write posts on the desktop and then copy and paste them into the editor. Otherwise, I lose my content if I try to write within the WordPress editor. I don’t have that same problem with Amazon’s blog editor so part of the problem is software maturity.
At the moment, I have the following apps open: XML Notepad, Visual Studio 2005, Word, Outlook, Remote Desktop Manager, Context, and IE 7. I have three displays, seven tabs open in IE, and five desktops that I access using Remote Desktop. I don’t want to do all that stuff in a single application.
I’m going to have to disagree with your first commenter, Anne. I think you are totally right. I live in the browser — I’m a writer too, and I use Writely more than anything else now — with email, blogs, calendar and everything else I need. Desktop apps are clunky and bring me nothing I can’t get in the browser, even when it comes to photo-editing and other functions.
me too with the browser…. probably the only regular ‘app’ that I use is iTunes… although since i’ve had to work on a couple of different laptops I’ve had to ditch that for Pandora a lot of the time (a far from ideal solution). All my email, calendar, work processing, blogging, spreadsheets, and project management are all done with web based apps now. I have even *almost* replaced photoshop with a web app (not that my demands of PhotoShop were particularly extensive!)
I love web apps because they’re so easily accessible - either by me on my various machines, meaning I don’t have to carry a laptop around anymore, and also so sharable with others which is great.
Web apps are also more ’shiny and new’. that’s not a technical term, i know, but it’s still part of the reason I love them
what I’d love now is for the browser to kind of disappear now, so that my desktop IS a browser and my browser is not an app that I have to launch all the time.
Anne,
I have to agree 100% (and not just because I’m a webapp developer).
I’m betting big on the browser.
Vanessa, there are certainly apps right now that don’t have good web-based replacements. You mention power-user word processing. I was also thinking about programming with an IDE, like Eclipse. However, that doesn’t mean that down the line such apps won’t be replaced by their web equivalents. It is hard to imagine though building something like Word using HTML and JavaScript. Blech.
I also use Remote Desktop on occasion and I find it really painful. I’d love if I could instead access all the apps and data I needed without having to go that route. But we’re not there yet.
I tried Yahoo Mail beta for a while–hated it. Didn’t go back to Outlook though, I went back to Yahoo mail classic. But I only use that mail address minimally. My main mail app is gmail, seems fine. But I’m not a power email user.
Thanks for your comment, Vanessa, I’m glad to get an alternative viewpoint.
I haven’t tried Writely much, but I’d love to get away from MS Word. It seems like an intruder on my computer when I launch it, sitting there separately and haughtily in its own little world.
What photo editing do you guys use? I finally broke down and bought Fireworks for my Mac (I switched from a PC about six months ago and postponed buying any graphics editing software until I couldn’t do without it).
Leisa, I could see the browser just morphing into the main system we interact with… I’m sure that’s what Microsoft would like, so long as its IE and not something else.
Great. stuff. anne.
felt like puking? why dont you tell us how you really felt?…
this is good though, some ongoing analysis of MAX related issues.
Hi Anne,
Dinner at MAX was nice, I hope you didn’t feel like puking then -;)
You write: “Apollo
Hi David, thanks for stopping by my blog and sharing your thoughts. No nausea at dinner–that sea trout was really delicious! And the company was excellent too
I’d like to distinguish between the short and medium run versus the long-run. There’s no question that in the short and medium term we still need desktop apps and that we need a better way to build them. Apollo looks like a better way to build them.
Long term, I’d really like to see the browser paradigm win and I think it’s going to. That doesn’t mean, however, that I’m thinking of browsers as they work right now. There’s got to be more interaction with desktop resources and transparent data synchronization for offline work while meeting security and privacy requirements. Performance has to be excellent (Tamarin should help with that) and developers need tools like Flash or Google’s Web Toolkit to shield them from the pain of browser incompatibilities and give them more interaction. What works especially in Flash’s favor, obviously, is the richness of what you can build and how you can integrate video, audio, pictures, UI controls, etc. into a seamless experience.
iTunes is a good example from which to consider what to do about desktop vs. browser apps. True that there is no easy way to build such a cross-platform app right now. I could see it working well as more of a browser app, though. I wish it stored my data online and then sync’ed it with all my computers without my doing anything (okay, browser apps don’t do that yet, maybe in a few years). I wish I had more ways of integrating together my various music listening–last.fm with iTunes for example–in the same way that I can easily integrate and play with text data from RSS feeds and web sites. I wish iTunes didn’t bug me regularly about downloading a new version.
From a short and medium run perspective, Apollo looks useful and there’s no question that it’s an impressive piece of engineering. What I dislike about it–and what made me feel a little queasy–is the idea of putting more attention and effort into the desktop paradigm instead of starting from the browser paradigm and figuring out how to make that work better. I spent years developing GUI apps for X11/Unix and Windows; moving to the Web development paradigm was like coming out of a cave. I can’t quite articulate why.
The Tamarin project is really good news and I’m excited to watch what Adobe does next. Clearly your capabilities with video give you a lever into next-generation web development that other organizations don’t have. Plus I saw lots of other cool stuff at Max, like how Spry brings Ajaxy goodness to DreamWeaver. Apollo looks good as a medium-term solution, but I’d rather see momentum move away from the desktop not towards it.
Virtual Server Admin has a web-based remote control feature. I thought it was kinda cool when I first started using it, but then it seemed like a pain. You can’t maximize your remote session. Now I use Remote Desktops on a 19 in monitor. I have ample display space for my remote desktops plus I can easily click from one remote desktop to another using the navigation list on the left.
I admit that I have some desktop needs that are a little different. I have to admit too that I have a hard time fathoming a no desktop app world. It’s a metaphor that I’m accustomed to. It’s probably the same thing that happened when dumb terminals were replaced with PCs. Many workers were overwhelmed by the transition because it was so different. I certainly don’t want to be in that boat!
Vanessa, that’s interesting that you couldn’t maximize a remote session on the web version. There must be a way to take over the screen with a web app, because I was using Web Ex for a meeting yesterday and the meeting organizer was able to take over our desktops completely by some mechanism. Not sure what was going on or how Web Ex works.
You may be right, about a no desktop world. There are so many things that can’t yet be replaced by a browser app: serious document publishing, high-end graphics, software development. I could very well be wrong. But I love the ease of getting and managing apps via my browser. In contrast, dealing with desktop apps is such a pain.
Hi Anne,
Read your response to my comment and I think we are actually on the same page, but suffering from an absence of a common vocabulary to describe the next generation of applications. You don’t want to see the pendulum swing backward from the web app to the desktop app. Neither do I. I don’t think of Apollo as representing the world of the “desktop app” as we knew it, but rather a “web app” that can have many of the qualities we miss from era of desktop apps. It is still web deployed (no need to manage the install of .exes or updates) and built with web technologies (de jure and de facto standards) and in fact, your description here of where the world of web apps needs to evolve to is pretty much what I think we are doing:
“There
David, thanks for the detailed reply. I agree that what we want to see sounds very much the same: a unification of the best of browser and desktop apps. In many respects, Apollo seems like something that approximates that.
However, the direction of how Apollo gets there makes me concerned. It seems to be going against the broader currents happening in software:
1. compared to browser apps, it adds more friction to the process of installing and using an application, by requiring the presence of a player and because Apollo apps won’t transparently install and upgrade themselves (at least according to the session I attended… they described installations that would look like desktop app installs). Browser applications, by contrast, do not need to be installed and usually do not need any additional software on the desktop other than the browser itself. Upgrades happen transparently too.
2. Apollo moves away from the software as a service paradigm back towards the local desktop app paradigm when all around we see how much easier it is for people and companies to deal with SaaS.
3. it moves towards more proprietary and less open–even though it uses accepted and mostly open standards for content, it packages them up in a non-open way.
4. it puts more of our software and data on our local machine at the same time that people are wanting to use many different devices (mobile phone, PDAs, multiple computers, media centers) for application usage and data access.
We both agree that the desktop and browser experiences need to be bridged. But I think that there are right and wrong ways to get there. The right way, in my mind, is to swim with the current of the technology industry, not against it.
Two examples of companies swimming with the current rather than against: Zimbra’s announcement of offline access for their collaboration suite and Parakey’s moving the desktop closer to the browser/web.
My main concern is not this idea of “forking the web” but rather that the way Apollo addresses the browser-desktop gap is backwards looking not forwards. The problem with that is it may work in the short and medium term (for how long, who knows, could be many years). However, long run trends will make what Apollo accomplishes obsolete when browsers start offering almost everything that Apollo does, more openly and without introducing new friction into computing.
Hi Anne,
I not try to convince you further now–I think Apollo is closer to the open web app paradigm than to the historical desktop paradigm, but lets reconnect on this again in a bit when the Apollo developer release is out and see what you think then.
BTW, the Zimbra approach does require a local install as well (Apache Derby). I’m biased to the more seamless approach is that taken by iScrybe (http://iscrybe.com/main/index.php) leveraging the (almost 100%) already installed Flash Player :).
-David
David, I hope you’re right, as we need better ways of integrating our desktop experience with the web. I thought Apache Derby was just downloaded as Java jar files, which is a bit more straightforward than getting an extra player plus downloaded application files. No question that Flash brings some important capabilities to the browser–I will check out Scrybe too.
5 Trackbacks
I Resemble That Remark
Anne Zelenka posted a good commentary on her site yesterday regarding her preference for everything to be in the browser.
[…] I believe that browser-based apps are the way forward. They remove so much friction from computing. And even more important, they put the emphasis on information and data and navigability rather than on the application itself. Google Reader is a good example: the interface packs tons of information about unread items into your browser window. Then it gives you keyboard access to move through those items quickly. […]
[…] I believe that browser-based apps are the way forward. They remove so much friction from computing. And even more important, they put the emphasis on information and data and navigability rather than on the application itself. Google Reader is a good example: the interface packs tons of information about unread items into your browser window. Then it gives you keyboard access to move through those items quickly. At any time, you can move externally to seemingly infinite sources of information via the links in the articles. […]
[…] Observations on the long running “browser as a platform” debate. […]
[…] When I first confronted Apollo, I wanted to categorize it in ways I already understood: browser or desktop. And since I’m a confessed browser bigot, I didn’t like it. It seemed too much like starting at the desktop and mixing in a little web, leaving the browser entirely out of it. What might we lose without the browser? Zero install access to a whole range of applications and information, browser navigation capabilities like back and forward and search, data storage out in the cloud, potential for information integration, and browser tabs. […]