The advance of the web has been happening randomly. Standards have taken a long time to develop, and browsers has often done things differently. Interoperability has however improved a lot over recent years, to the point where developing websites, and archiving consistent results across platforms and browsers has become a lot easier.
This cross-platform compatibility is effectively removing the need to develop separate sites for mobile devices and desktop devices. There may still be advantages to mobile versions of a site, but ideally this should, in most cases, change as standards improve.
Emerging CMS systems
Then there's the whole briefcase of CMS systems emerging, which is very likely to take over a huge market of otherwise custom developed systems. In the process perhaps even decreasing the demand for developers.
I think people are increasingly becoming aware of the possibilities with free CMS systems, and the money they can save by using them. Ideally these systems should have a low learning-curve, or none at all (assuming that the general population is increasing their core IT skills).
The future of web standards
I would like to see the standards move away from versioning, and the developing bodies work closer together with browser developers. The fact that development of browsers and standards has been going on separately, is a core reason why we have lacked decent tools to design websites for decades.
You can think of HTML and CSS, as you can think of the human language – they share a common characteristic. Much like there's often invented new words, and the meaning of old words is changed, so it is with features of web standards.
We are not versioning the human language, so the question is, if it really makes that much sense to version web standards, just because we are dealing with software? I don't think it dose. Browsers have not been using the version information anyway, they have only used the document switch to turn on standards mode - where's a page with no doctype deceleration, would trigger quirks mode in some browsers, which is a kind of backward compatibility mode.
I thought of this long before the WHATWG officially ditched versioning on HTML, and started calling it a "Living Standard", but didn't really see it on the horizon at the time. Now I'm beginning to hope that the same will happen for CSS to, and that the w3c will realize the potential in all this.
The openness of the web can be a good thing in the long run, but it has killed innovation since the dawn of the web. If the web was maintained by one organization, and one browser, adding new features could be done overnight. No need to argue with anyone. That's why proprietary platforms are often way ahead.
The future of web browsers
I don't see a need to have multiple browsers. Multiple web browsers has traditionally led to problems in the way the browsers interpreted the standards, and Microsoft in particular has delayed, and continue to delay further development because of Internet Explorer.
At some point Microsoft decided only to upgrade Internet Explorer when a new Windows was released. This means that you can't easily install the latest version of IE on Windows XP if you wanted to, leaving IE users with a lackluster experience compared to users of Firefox and Chrome. In worst case, developers are forced to support older versions of IE as a result, which is not to be desired.
Ideally we wouldn't have to support older browsers at all, and browsers would just be automatically updated, like other any software. While its easy to update a browser, it can be difficult for people to learn a new language. Its important to note here, that browsers are not people.
While its not realistic to re-invent the web, we can learn from past mistakes. I still think there's room for improvement in interoperability between the different companies, and such improvements is only likely to speed up innovation on the web.