[TeMPOraL]

Viewed from the outside, webdev looks really funny. I'm doing web for a living, but I don't consider myself a web developer[0]. I'm not emotionally attached to any of the tools of the trade. For me, many web developers are the biggest cases of Stockholm syndrome I have ever seen in my life. Tools we use are ugly and broken. CSS is the worst offender for me; doing layout work with it is soul-sucking. It's an embodiment of the idea "make hard things one-liners, and the simple things half a screen long and not portable" (it's getting better now, but not that much). I'm really starting to doubt that switching from tables to div+styles was a good idea. We didn't gain any real content/form separation (I doubt the concept makes real sense in terms of publishing), and we replaced a tool that worked as layout engine somewhat with a kludge that can do anything but that.

So what did the kids do? "Oh no, our tools are fine, but those annoying problems? Hey, let's bolt on more tools, like various flavours of compile-to-CSS languages, client-side frameworks, framework generators, package managers, test frameworks, test framework generators, etc. etd.". I'm learning about a new tool that is "absolutely needed by any self-respecting JavaScript developer" literally every week.

The web nowdays is run by fashion designers and fashionistas.

[0] - and I hope I'll be able to find something more interesting to do soon, but for now, one has to eat and pay the bills.

[1] - http://en.wikipedia.org/wiki/Stockholm_syndrome

[innguest]

Not enough people are brave enough to speak out like you do. I agree with the Stockholm Syndrome argument. There's also an argument to be made about the fact that those who defend the broken web think it can be fixed with more technology and complexity, instead of less (ie. stepping back, coming up with a foundational model of computation, and so on, as functional programming has been doing).

If you live in NYC, let me buy coffee/beer/whatever sometime - I don't know anyone personally that agrees with me on those things.

[TeMPOraL]

Thanks for invitation, I'd love to talk this through over a beer :). Unfortunately, I live on the other side of the world. I'll let you know if I happen to be in NYC; shoot me a message if you find yourself in Poland :).

[Silhouette]

It could simply be that front-end web development is predominantly done by a new, younger generation of programmers, many of whom perhaps didn't come from programming backgrounds but rather a design background.

This generation is only just starting to learn important lessons that the wider programming community learned decades ago about maintainability, portability, the pros and cons of standardisation and of rapid prototyping/evolutionary development, how to design systems of non-trivial scale that can remain useful for extended periods, how to balance the overheads of introducing external dependencies with the benefits of using ready-made, tried-and-tested code and tools, and many other similar issues.

Consequently, that new generation is also only just learning how limited and unfit for modern purposes a lot of the foundations of front-end web development really are. As a direct result, we're now seeing a vast wave of tools, plug-ins, libraries and frameworks designed to paper over the cracks that should never have been there. And because there is little co-ordination or standardisation in the industry, there is also unfortunately a huge amount of wasted work, both for developers of those tools and libraries who are doing the same things repeatedly for no great benefit, and for the potential users who have permanent analysis paralysis.

That much will improve as the industry matures and consolidates, though I fear we are now stuck with HTML/CSS/JS as our foundation, at least until the successor to the Web comes along with things like interactive sites and security-by-default as primary design goals.