Actually I think we still have that kind of innovation. There are still standout technologies that arrive essentially from the mind of a single person. Think about Node for example. Ryan Dahl designed the initial version and the majority of the original technology and code still powers that entire platform. Some of the bigger client side frameworks are typically the design child of a single developer and only after they gain some traction does it become a much more involved operation.

I think we're solving very different problems today than the ones we solved back in the in the late 90's and even longer ago. Distributed systems involve a lot more complexity. Different platforms that have completely different development platforms are lot more complex. Different screen sizes, different features on mobile devices - trying to build a one size fits all application seems futile to me at least if you try to do it 'natively'.

What we need to support is much, much more complex than what we build 20-30 years ago, it's that simple. We couldn't build the kind of stuff we're building today with relative ease, not unless you learned C and assembler and started digging into OS Apis.



A Web application is very different from a desktop application is very different from a mobile application and so on. There is no one size fits all and even if there was it would probably be a pretty watered down solution. We're talking different platforms, different environments each with their own little quirks. It's hard enough to build a decent Web framework and account even for the majority of browsers and mobile devices let alone trying to run on a Windows desktop, a Mac and a Linux box.The way people do this today is actually using HTML container technology like the Atom editor shell and the Chrome shell, that run cross platform but essentially are HTML/Javascript applications running inside of a shell.

We're seeing more and more of this because HTML and JavaScript for all of its problems is what works on every platform. everything else may have a run at it but in the long run I think HTML/JavaScript will win this race for the simple reason that it's everywhere.

+++ Rick ---




>>>As bad as that is though, I think for the first time since I've been doing development (which is a long time now :-)) I actually see the current trends leading to minimalistic approaches. We're going backwards from ever more complexity to smaller, more isolated and simpler pieces in the development process. And although there's now a lot of fragmentation, I think once the pieces find their way together in the near future we'll be much better off with component technology we can combine to build solutions more easily.
>>>After the maelstorm there's usually a period of calm - I think we're at the height of the storm right now where it's next to impossible to keep up with the newest trends. It'll change - this stuff goes in cycles and will consolidate.
>
>IMHO there's no excuse for lack of a cross-platform 4GL that runs natively on any platform so that individuals with niche business insight can create useful work, as used to be the norm. I used such a product 30 years ago on the TRS-80 and saw my stuff working on an Apple II afterwards. The compiler was written by a single programmer, Andy Gariepy, whose work created tiny machine code suitable for the 16K computers of the day. Whole accounting systems could be rendered in 16K RAM along with spreadsheets, word processors (e.g. Electric Pencil that introduced us to word wrap) and most applications of the day. More recently, a clever fellow wrote something called "West Wind Web Connect" ;-) but by the time he did that, it was unusual to be able to make such individual headway. Today it's almost unheard of because of the massive complexity pollution and expectation that every development task is worthy of Rube Goldberg.
>
>My question is WHY. Why have we allowed this, and who does it benefit?
>
>Give me a (wo)man who delivers a 4GL that delegates UI to the vendors for defined commodity delivery but gives me consistent 4GL control over everything else and I will kiss his/her feet.
>
>The problem now is that the end-buyers are weary of all the churn and don't get excited as once they might have because Visual FotD has arrived on the scene, hoorah! Uh huh, they say, and go back to downloading free software on their Android phones. ;-)