R.I.P. Browsers

I'm not really the product vision person at CA. I very rarely make any sort of prediction of the future of anything. I once tried to and was smacked down by Steve Gillmor in a Berkley parking lot (only figuratively, of course). I think he was saying something about desktop apps being dead and I can't remember what I said, but it was not too visionary.
But, after spending some intensive time with Chris, our clients and really starting to think about it, I feel more and more that someday - and it may be sooner than we think - there won't be browsers at all. Not that they will look different...we just won't use them. And I know Chris has been saying that for a while, but I didn't really know how to interpret his metaphors until just recently...and I had to come to my own conclusion.
I used to think that our desktops would disappear and browsers would replace them. I mean, we do so much online now. But the more I think about it, the more reasons I can come up with to backup this theory:
- We are really constrained by browsers. Think about it. People saw AJAX and went, "oooooo, ahhhhhh!" but we've been able to drag and drop on our desktops forever.
- Browsers suck up a great deal of virtual memory. They are an OS on an OS, basically now.
- We have to type in addresses to get to applications. Websites run applications. But instead of just clicking naturally, we type in an address, then fill in login information, then click.
- All of our information is stored in the ether. If we want to store it on our own machines, we have to take an extra step. Shouldn't it be the other way around?
- And speaking of information in the ether. Wouldn't it be good to work offline whenever we wanted to and have it update when we are re-connected?
- Have your browser crash recently? Any backup? Restoration of where you were? Heh. I'd think that would be pretty basic. No?
- The webtop gives us a much smaller working space.
- There is this whole having to design for browser compatibility thing that would be done with - just design for OS compatibility.
- I, personally, want to be able to read stuff offline. I want that to be solved.
- I know, I know...your web pages...your blogs, etc. Well, that is the only part that needs to be solved...although RSS solves much of that.
- Just think! No more stupid search! Search is sooo broken. I want find. There has to be something smarter...
eMail already runs that way and so do many of our calendars. I've actually totally effed myself by commiting to Google calendar, because when I can't get online, I can't see my information. [UPDATE: Thanks to Adam Kalsey for rectifying this for me...awesome instructions on how to feed your Google Calendar into iCal] I've already run into this issue countless times. Google spreadsheet? Ha. Imagine the horror.
I know Windows Vista is working on this angle. If they can get their act together, Microsoft will, once again, lead people into a new paradigm altogether. It won't be us poor Web 2.0 saps that are tickled to see a webpage that is movable (albeit laggingly slow and crashing our browser to leave us in the predicament of having no backup in the browser).
That is my prediction. No more browsers. Just connected desktop apps. So...what's next?
Tags: browsers, web2.0, predictions



31 Comments:
Interesting post Tara, when I first start reading it I thought you were nuts for even thinking of the possibility of no browsers. I must admit though it makes sense.
I have a question for you. Why do you think search is broken and how would you suggest it becomes unbroken?
These things you say about browsers and desktop application are some of the big reasons I am a fan of things like Pyro for Campfire
http://karppinen.fi/pyro/
It is still not perfect, as it does not allow you to see any Campfire logs or anything while you are offline, but it does allow you to make Campfire into an application on your OSX desktop. Only if other applications would do the same, and maybe even extend the idea and allow desktop synchronization.
RSS does get data, however it has one enormous problem: polling. RSS polling is a bandwidth problem that needs to be solved before everyone and their bloggin' uncle depend on it for their stuff.
Hi,
One of the most creative out of the line thought indeed.
Completely agree with the 'no browsers' line. I was getting some very confused looks back in 2000, when I first came to that conclusion (as not many had even started to use them then!).
Browsers are very anti-social, like sitting in your own little booth at a library.
I like the idea of being able to create a virtual environment where proximity = relevance (I'm not going to explain what I mean by that - I think you'll understand).
The whole trauma of not being connected to the web is something that will have to be solved, nothing short of ubiquitous, unfettered access will be acceptable.
So to buy something at Amazon, you would have to install the Amazon program for your operating system?
But how do you get that program if you can't download it from some download site? From a friend on CD or DVD? That sounds like going back to the pre-internet days to me.
And what if the Amazon program isn't available for your operating system? Guess you are screwed, then :)
Read stuff offline? Ever seen the File->Save menu in your browser?
And how are you going to read blogs if everyone has to download a separate program for each blog? Not to mention the fact that all modern browsers have RSS readers built right in.
I agree with most of your points, but you seem a bit hand-wavy on some...
(gripe: Blogger really needs to support the blockquote tag...)
"All of our information is stored in the ether. If we want to store it on our own machines, we have to take an extra step. Shouldn't it be the other way around?"
There is "you" information and there is "we" information. Things that only you will modify ("you" information), if you want to have machine-first/Internet-later, that's reasonable. Things that you expect to collaboratively work on ("we" information), that gets dicey -- just look at software development and all the rigamarole with version control systems. It is easier to get more reliable results if you can assume everyone is online... which just makes being offline more of a problem.
"Have your browser crash recently? Any backup? Restoration of where you were? Heh. I'd think that would be pretty basic. No?"
And which OS gives you that? Heck, with various browser plug-ins, you can get closer to "restoration of where you were" with a browser than you can with most OSes -- if Windows blue-screens, for example, restarting it isn't going reopen your Word document for you. The fact that, say, Firefox hangs more frequently than does Windows (let alone Linux) is more a statement of maturity of the application than anything else.
"There is this whole having to design for browser compatibility thing that would be done with - just design for OS compatibility."
Ummmm...that's rather difficult, even in languages like Java that are supposed to make it easy. Web browsers were part of the solution, actually. Moreover, "design for browser compatibility" is a progressive problem: you choose how much of it you want to have to deal with. It's perfectly possible to create plain HTML-based Web apps that work just fine in all browsers, but you can't have graphic designers trying to pixel-position things and you can't do Ajax.
What I suspect you really want is a different sort of frequently-connected client application to replace the Web browser. Web browsers are best for displaying documents, not rich user interfaces, and all the Ajax in the world ain't gonna help that -- that's a lipstick/pig situation. What Ajax will do is start people down the REST road, which is a step in the right direction towards decoupling Web browsers from the Internet application experience.
All I know is that IE has become horrible and I'll never use it again.
Tara:
You are 100% correct. I think this flies in the face of many of the "Web 2.0" evangelists, but I think this will be the next step in really using the web to its full potential
It's hard for us to remember that not everyone is as "tech savvy" as some of us and the Internet is not necessarily the friend to them as it is to us. I think that a desktop/web combination will be a big step to removing some of the barrier to adoption of things like RSS for the "common person."
And, personally this would be a big step to me. We recently moved to an area that only gets...wait for it...DIAL UP! So, the ability to work offline and sync with online would be a big thing for me.
Your description of not being able to access information online when you need it (ie Google calendar) has caused some of my reluctance to completely adopt some of the same tools.
Very insightful, as usual!
wow. If you bring this 'master of the obvious' quality to other apsects of your agency you'll surely be a success.
As a developer I'm tethered to apps, the ones I use most of the day. I can't live in the browser, so I'm not about to declare that desktop apps are dead.
But the most apps I use are on the Web, they're just easier to reach and use. They also benefit from being on the server. Because the data is huge (think Google Maps, Flickr), because servers can do more interesting things (GMail search, Wiki), because they're just cheaper/faster to deliver (BaseCamp, del.icio.us). Because I can read my e-mails and calendar from a cell phone.
Browsers suck up memory and slow down the computer, but so did Windows early on, when it added drag & drop to DOS. And people complained about it. And they were right. And they moved on.
Browsers will get slightly better, but memory will get much cheaper must faster. In fact, less apps means less disk space, so trade the hard drive for a flash card and you can get much more battery life, less weight. At airports, WiFi is easier to find than power cords :-)
I have a monster backup script because I run so many applications. One line to backup my browser. (And paying for space, backing up just the browser is cheaper)
I'm with you on the screen real estate, I wasn't impressed by Zoho Writer because they only use 80% of the screen, and I can't get productive without full screen view. I use OpenOffice instead. But I do get full view in some apps (F11 in Firefox), and open GMail/GCal in separate window that has nothing but border and title.
Designing around browser quirks is hard stuff, it sucks up a lot of development time. Designing desktop apps is even harder. DLL hell, broken registries, viruses, locked files, reboots. When GMail broke a few days ago (the dreaded 700 error), I had to restart the browser. When my IE broke two weeks ago, I had to reinstall Windows.
How many people do have the offline problem? Most of the people I know are always connected to the network, they're at home or the office. And if you're really on the move, there's EVDO. Yes, it costs a lot, but so did online back when most people didn't know the Internet existed (hundreds of $ a month). Adoption came, prices went down, it's one of those problems that fix themselves.
And yes, search is broken, because crawling HTML over a lot of URLs is a hard problem to solve. But crawling weird data formats that are spread in different directories, sometimes in locked files, is an even harder problem to solve.
I'm not about to declare the "browser is the new OS" like many people do, but my gut tells me it will be the 80% app.
Tara, I'm impressed - you ALMOST get it, and I don't mean that negatively. With technologies like Flash, you can free yourself from the browser, but still keep the "web application" that has become so valuable.
Adobe's Flex, Microsoft's WPF, and OpenLaszlo enable the Rich Internet Application right now that is slowly eroding the browser. Once people realize there is a better alternative to Ajax, they'll embrace it.
It makes perfect sense, but it wont happen.
The business driving the development & progress want out of Microsofts shadow, and the web is their answer to this dilemma.
Once again, what will happen will be driven by commercial interests, and not by logic.
/Mikael Bergkvist
I like 'connected' desktop apps - from the Clipart viewer in MS Office on Windows, through to Flickr import/export in iPhoto - cut out the middle-man between my client and the server.
(Hey, we could call this idea client-server computing!).
I like interconnected desktop apps too (iLife/iWork suite) rather than 80s style standalone apps, which is what web apps are.
However, I think there are a couple of things mitigating against it. The companies I'm dealing with want web apps - even when it's a more expensive and inappropriate choice.
They solve a lot of deployment problems for companies - client machines only need a browser (and most of them know that will be IE6).
Also, a lot of these technologies (the various Widgets, Flex/Flash) are based around Javascript in some form or other. They're about using the web programming tools to develop desktop apps, because there's a massive base of web developers out there who don't know 'desktop' languages.
It seems like you're not looking forward to ubiquitous networked computing and you want to stick to yer own desktop client.
I see a future of many specialized browsers, some that work in the way you describe, but most of them focused on working on the network.
Search is contantly growing better as AI improves...and that's exponential growth!
Tara
It's good to challenge commonly held assumptions but I think you are wrong on this one.
We do of course already have networked desktop apps and they offer richer user interface options than web apps, but there are reasons that a lot of the action at the moment is happening in web-apps delivered through the browser.
One is increasing connectedness at higher speeds and lower cost. You are right though that it isn't everywhere and web-apps are a pain when you aren't online. But there are not many places you can't get online now, and I think it won't be long before we are online everywhere, to the point that solving the offline working problem is probably just not worth it.
The main reason for the success of the web and the web browser is that it offers a standardised way for us to communicate. A browser is just a way of processing those standardised messages. What you gain in interoperability, you lose in flexibility and richness of the interface.
So my feeling is that we will continue to use a browser, or some equivalent but maybe slicker standardised rendering and execution environment and everything will move to being always connected and you won't know or care where your data is stored in the same way that you don't know or care where your electricity comes from.
It would be good if either of us could remember to come back to this in 5 years! I'll bet you 1 gig of RAM I'm right (should cost about 1c by then!)
Bill
You can get a URL for your Google calendar that you can use to subscribe in iCal.
In Google Calendar, click the little down arrow next to your calendar's name on the left. Choose Calendar Settings. Then in the Private Address section, copy the ICAL link.
In iCal, choose Calendar / Subscribe from the menu bar. Paste in the URL you just copied. Then choose to refresh every 15 minutes and make sure you don't choose to remove Alarms or Todo Items.
Now your calendar will be mirrored in iCal and you can see it, sync it to an iSync device, or whatever.
Looks like a very timely story...
This from Ray Ozzie at today's Analysts Meeting...
Found at :
http://www.microsoftmonitor.com/archives/016742.html
"... Ray laid out a compelling "services transformation" that would touch all markets, and he said that all companies would have to embrace it. His vision isn't services in a browser as envisioned with Web 2.0. In fact, he disparaged the Web browser by comparing it to the terminal. Whoa..."
It goes on to touch on some of the things I think are important ala live data.
Great Post,
Christopher
WTF?
"Have your browser crash recently? Any backup? Restoration of where you were? Heh. I'd think that would be pretty basic. No?"
Actually, yes. It's been basic in the Opera browser for some time. In the event of a crash, I can restart Opera, and it will ask if I want to continue from last time, or start over on my home page or a blank page.
The "always-on-you web" brings web-based apps to your pocket for use on any PC, anytime, regardless of net connectivity. It doesn't force users to push their personal or company data up to a third party service. It integrates seamlessly with desktop apps.
For more, see my blog, Web 2.5 : The Always-On-You Web
The "always-on-you web" brings web-based apps to your pocket for use on any PC, anytime, regardless of net connectivity. It doesn't force users to push their personal or company data up to a third party service. It integrates seamlessly with desktop apps.
For more, see my blog, Web 2.5 : The Always-On-You Web
In two years this will be "obvious", as Vista users will be streaming down mixed-mode WPF apps on-the-fly. Though technically this may happen inside of "browsers", it won't be HTML-centric so your point is basically correct. And to someone who mentioned Amazon above, Yes someday Amazon will have a fat app that will be more popular than its HTML interface. HTML just isn't the end of the road.
I believe you are wrong. The problem with "desktop apps" is the are on a desktop, an XP desktop, or a OS-X desktop, or a (insert your favorite linux) desktop. Each has a distinct user interface and a standardize (by the manufacture who ever that may be) way of coding UI. I use bloglines for my RSS feeding simply because. I can switch machines without having to sync up what has been read. RSS is not an answer to the brower. It always looses the fact that when I'm angry I change words to red. No the brower is not dead it just needs to be more consistant.
I mentioned Adobe's Apollo project in another post, but it is releveant here as well. Apollo will take Adobe's new Flex 2 product to the desktop enabling developers to create desktop apps that communicate and update via the web, but run on the desktop.
On a side note, I personally can't wait till the day when we don't need huge hard drives and local OSs. Just a box that accesses an OS on the web (which is fully customizeable) and everything you save is saved to a virtual drive on the web. Your whole OS is, in a sense, a browser. Then you can access your "computer space" anywhere in the world. You won't have to worry about carrying a laptop wherever you go to access your files. Just use a public "access point" computer and boot up your remote OS/computer.
Think evolution Tara...
Browsers are already evolving.
iTunes is an example. It's a local desktop app with web technology (Safari) integrated.
Did it kill Safari? No.... will hybrid web apps kill the browser. No... they'll just evolve and be easier to use.
Ok..... now I'm going to blog about how something everyone loves is going to die....
Which one of these subject lines sounds better?
R.I.P Linux
R.I.P Apple
R.I.P Internet Tubes
R.I.P Your kitten!
Hm.....
:)
Obviously not a Firefox user? Session Manager for lost tabs? about:config editing to keep a watch on memory leakages, not to mention speed boosts.
I've decided why Firefox rules....
You know when you pay money for something and after a period of using it you decide that it lacks in some department and that there are things you might have included in its design to maximise usage/efficiency, problem being most products are the way they are and that's that...
Firefox is constantly evolving, fun to use and highly customisable.
All the little problems people find in every facet of Firefox are nothing but that, whingeing. They ignore the sheer useability and enjoyment achievable by exploring extensions.
And besided how would we browse the Net without browsers?
R.I.P. Web Apps?
An analysis of this article
http://guillaumeb.com/2006/08/08/why-browsers-wont-die/
sounds like you need a better internet connection :-)
Just thought I would offer up a different point of view here.
Craig
Post a Comment
Links to this post:
Create a Link
<< Home