Thin-client has been tried many, many times before. For mobiles, it’s a trade off between local computing power and cost (availability) of communication. Thin-client situations have heavy communication requirements, but light computing power requirements. Thus I don’t think thin-client makes a whole lot of sense in a mobile phone context.

My prediction is that mobile computing power will soon rival computers from a few years ago. Communications networks won’t keep pace (at least not from a cost perspective, CPUs you pay for once, bandwidth monthly) so why would you want a browser-only phone? What would you run on a mobile phone that is running an ARM processor that had comparable performance to a Pentium class machine?

When that happens, and windows mobile comes to more resemble windows proper, that’s what I’ll be running. Maybe I could even leave my laptop at home upon occassion.

Further, being a city dweller, there are few times when I’m out of signal; however, I’d imagine that when I’m out of signal, that’s when I’d want to use GPS – what good is a browser phone then? The same could be said for using a mobile phone is subways, etc.

My 2c.

Browser based applications had their claim to fame in the early Web 2.0 days. Today it’s all about the semantic web (some people call it Web 3.0) and having your data accessible anywhere. To me that means having separate applications, widgets, websites and even a start page that pools information from a single data source.

Living within the confines of a browser isn’t where the future is heading. I fully disagree with you on that.

Why should I have to open a browser to check the weather? Why should I have to open a browser to write myself a todo? These actions can all be done via API’s today.

The best real world example is twitter. You can update your status via widgets, a website, sms, instant message, it doesn’t matter. Twitter lives outside the browser.

I think you’re missing the point of what I was saying though. I wasn’t trying to make specific technical predictions for the next five years, no one can really do that with accuracy.

I was trying to talk about the general direction smartphone software would be likely to go over the next few decades, because this browser-based direction already has such huge advantages on the desktop, and smartphones are getting closer and closer to desktop browser functionality. People invent solutions when there’s an incentive for them to do so, and I think there’s a heck of an incentive in going for web-based apps on both PCs and smartphones.

If you look how quickly browsers on smartphones have developed, they’ve almost completely caught up with PC browsers. The new S60 browser can handle almost every site except AJAX and Flash. Nokia’s recent N800 Linux internet tablet can handle almost every site including AJAX and Flash. The N800 is light and small and significantly cheaper than the N95 or E90, so it seems plausible that we’ll see the same kind of near perfect compatibility on smartphone models at some point in the next few years.

I also think you’re missing the point when you put the burden of compatibility on browser developers. If someone comes up with a web app now, the first thing they’ll do is check if it works with Internet Explorer and Firefox, because these represent such a huge potential market. If Gmail stopped working with Internet Explorer, Google wouldn’t sit around waiting for Microsoft to fix it, Google would fix their own code so it DID work with Internet Explorer.

In 5 – 10 years time we will probably see hundreds of millions of smartphones sold every year with web app compatible browsers, so people like Google would want to make sure their web apps could work on whichever smartphone browser standards are the most common.

I suppose saying AJAX would be THE standard for web apps was way too specific, what I should have said was that techniques like AJAX showed how web apps could be viable replacements for offline apps. AJAX has made many ordinary people (not bloggers, but the far far far more important non-techie majority) see that web apps could be as slick and easy to use as simpler offline apps. Someone who uses Windows Wordpad could just as easily switch to Google Documents for example.

If AJAX on Gmail etc is replaced by something else, Flash or whatever, it really doesn’t make any difference, in fact the end user wouldn’t even notice! The new web app would be at the old web app’s URL, with the old web app’s user interface and all of its functionality. The end user would just enter their password and username as normal.

If the standards for web apps are splintered, that doesn’t matter either. Web app developers could make versions of their software to suit all the most popular browsers and automatically deploy the version you need depending on which browser you’re using. You wouldn’t even have to choose a version. Deploying software on many platforms at once is not a challenge for these people, especially the biggest companies like Google and Microsoft, and especially for something as simple as a webmail or Wordpad-style service. At the end of the day it’s just about getting your product to as large a market share as possible, so if the market is evenly split between several standards they’ll make versions to cover most or all of the standards. It’s a fairly easy thing to do, especially on the web where the distribution costs are almost nothing compared to distribution of offline software.

About Apollo, in the road map there is an mobile version and the fact that the current Apollo HTML engine is the open source WebKit helps to say that Adobe is thinking in that direction. Adobe mention that one of the reason they pick the WebKit was that is already been ported to mobile phones.
There are no information about the mobile version pf Apollo but WebKit is been used by Nokia to create the S60 browser (code open source) and I am sure this work will be used in one way or another.

About Flash Lite 3, it will support Flash Video. There are no details yet and it will available to developers mid year. The FL3 player will have its own Flash Video codec.

Can Flash Lite replicate the success of the web on mobile? That’s a great question. First web and mobile are completely different world and many people just forget about it.

From my development having the same Flash Lite application running on Symbian, Windows Mobile 5, Java and the Browser is a pretty cool thing!!! No changes needed, only access the Flash Lite file.

Sure Flash Lite has limitations and you need to work around, but Flash Lite has shifted the attention to the usability and user experience on mobile. Something that until now sucked!

You can check some posts on my Nokia Champion blog about Flash Lite:
https://blogs.forum.nokia.com/author.html?id=8859

Alessandro