HTML Web App Development Still Has a Ways To Go 279
GMGruman writes "Neil McAllister was helping out a friend whose web developer disappeared. Neil's journey into his friend's website ended up being an archaeological dig through unstable remains, as layers of code in different languages easily broke when touched. Neil realized in that experience that the ever-growing jumble of standards, frameworks, and tools makes web application development harder than it needs to be. Although the Web is all about open standards where anyone can create variations for their specific needs and wants, Neil's experience reminded him that a tightly controlled ecosystem backed by a major vendor does make it easier to define best practices, set development targets, and deliver results with a minimum of chaos. There's something to be said for that."
Web development is hard for even talented people. (Score:4, Interesting)
Even for excellent software developers, web development is difficult. It's not the concepts that are difficult, per se, but rather the jumble of half-backed hacks that make up ever layer of the web stack. The foundation is so weak that anything built upon it just can't stand well, even if it itself is well-designed (given the constraints of web development).
Just look at the common open source technologies used by many web sites. MySQL is one buggy hack upon another. PHP is much the same, plus some security holes.
HTTP has been over-extended well beyond its original use (cookies are a hack to get around its statelessness, it's caching mechanisms are fucked to high heaven, SSL and TLS are hacks).
JavaScript is perhaps the most horrid hack of them all. Something meant for adding minor interactivity to a page has been misconstrued as being suitable for large-scale application development, although it lacks many of the most basic features necessary to do that sort of development effectively.
It's difficult enough to fight with unclear and conflicting requirements alone. Toss in shitty technology, and it becomes very difficult even for the best seasoned professionals to develop even just mediocre software systems.
HTTP is the problem (Score:1, Interesting)
HTTP needs a replacement for application development. It was designed for document retrieval not applications. For the past 15 years all these frameworks have been a hack to get over the limits of HTTP as a protocol used in application development. AJAX is the latest hack to overcome limits in this aging protocol.
Re:Web development is hard for even talented peopl (Score:3, Interesting)
Which features are missing, exactly? Cappuccino [cappuccino.org], for example, implements almost exactly the same APIs and conventions that native Mac/iPhone developers use, only in Javascript.
Re:"tightly controlled ecosystem"? Bullshit... (Score:3, Interesting)
If it was 'never intended for interactivity', why are POST and DELETE part of the spec? They are designed for interactivity.
If you want to get picky, 'computer' were 'never designed' for media playback, using your criteria. (That criteria being that only the initial thought counts, and not the years and years of changes afterwards.)
The problem is not the tools... (Score:5, Interesting)
The problem is not the tools (well, not *always* the tools), but the developers. You can provide the best development tools in existence to an incompetent developer, and you will end up with a crap website. It has nothing to do with the quality of the tools or the maturity of the application frameworks.
Hell, humans have been building houses for 1000s of years, yet an incompetent builder can still build a house that will fall apart. I don't think the problem is that the hammer and saw still have a ways to go.
It's just new (Score:3, Interesting)
x86 frameworks have been around for how many decades? HTML5 has been around for how many months?
Web development may seem like a pain, but last I checked, it's the only truly cross platform development environment that has ever existed. If your bank, for instance, had to develop binaries for every platform, you'd never be able to login and move your money around at 2 in the morning.
It's taken time to develop the standards that enable platform agnosticism of more vanilla HTML standards, and it will take time to flesh out new standards that compete with native toolkits for interaction. But it will happen, and compiled, single platform application development will be nearly unheard of.
Wanting better tools??... (Score:3, Interesting)
As a webmaster, if found the article strange.
We have better tools today, than years ago. Firefox and his amazing extensions (like Firebug or Webdevelopper) make analizing what is going out here, much easier. JQuery itself make writting javascript predictible (so if it works here, it will work on other browser). There are not lacks of good Code Editors. We use to have good HTML Wysiwig editors, but these are outdated by now (I could be wrong, but the better pad nowdays seems the brain of the designer).
So.. what you need? REDMINE to track proyects? Good debuggers? Inspectors? Documentation?
Working with HTML could be hard (specially if you have to support webpages that have to run on Internet Explorer 6), HTML itself is crap, Javascript is much more evil than people know (but I love it, like a bad girlfiend)... but.. the tools? what tools you need? tools that automagically write code (good code) for you? Visual Studio seems to have some of these, and while I don't use these myself, I have ear good things... (if you are into wysiwig and how bloated any code touched by microsoft looks. ). What you want exactly?
Could be, maybe, that you don't know what tools we the professional webmaster have available? :-)
Re:"too many frameworks" (Score:4, Interesting)
Off the top of my head, here are a few problems with the mryiad of many frameworks for the web:
1) The super-ultra-awesome slider you want is for YUI but the rest of your site uses jQuery. If you want to use it, you'll have to have the browser pull down both jQuery AND YUI.
2) Many of the frameworks conflict--prototype, for example, doesn't play nicely with a bunch of other frameworks.
3) Each framework added to your stack increases the number of moving parts on your site. More moving parts = more chance for error.
Seriously, it is a cruel joke when you find the-most-perfect-rich-text-editor but it was for MooTools instead of YUI.
*That* is my problem with having so many frameworks. The world would be a better place if we all just used jQuery :-)
Web Development (Score:4, Interesting)
There are a couple of problems I see in web development:
1) Unlike the systems programmers, myself included, for a given topic area tools are adopted and standardized.
Web developers seem to get jobs based on the flavor scripting language of the year.
(All of which is crap in my opinion....i.e. php, javascript....python...)
It always seemd too me, that XML, XSLT CSS and Java servlets are really all you need and you can build marvelous interfaces. Tried that once, but the response I got was (thats too hard, lets use javascript).
2) The closest I have come to a decent application framework for building web apps is Java. It has clear security controls, recognizes the importance of Virtual Machine technology to compartmentalize access in a dangerous online world. It even has a very straightforward debugging environment which is quite impressive to track down bugs.
But curiously, it is shunned because if you don't know the scripting language flavor of the day, people don't want to build web sites or won't hire you.
Which is one of the reasons why I don't write web applications anymore. Because when your job and pay is based on how fast you can memorize the scripting flavor of the year, and it doesn't bring anything new to the table (in many ways it can be even worse) to solve the problem of writing a web app, well...it becomes just a money game.
I mean really, I don't mind learning new languages, but I haven't seen anything new since Java 1.6 was released that is any better...just mostly worse.
3) Finally the field has become too greedy. I mean, there is no reason why it has taken this long to standardize video and audio, except for the fact that greed is everywhere.
It is really sort of disgusting, and the crap you have to go through to get video onto a persons browser is just way over the top, mainly due to Adobe and Apple being greedy idiots.
Maybe when the Video and Audio tags get full support for open protocols I will write web apps again. It isn't rocket science, but it is currently a science of idiocy.
-Hack
Re:"too many frameworks" (Score:3, Interesting)
It is when the choice is put into the hands of monkeys with darts. That's the problem he's pointing out. Most small companies don't hire seasoned developers unless they specialize in software development. They get whoever they can for cheap, or, at best, they get what their limited experience tells them is the best person. They hire:
1) Kids fresh out of college who know theory but have worked with very little.
2) People with no actual formal computer training who simply "know about computers" and and will work for what a real developer (or even a kid fresh out of college) would consider shit. But which beats what the guy was making selling tires.
3) People with experience, but in the wrong things. I'm a very senior sys admin type. My resume looks very fancy and technical. I'm sure I could convince some mom and pop to hire me as a web dev, especially if I was willing to take a pay cut (say I lost my job and needed something now). Could I build a web site? Sure. Could I make it functional? Absolutely. Would you want to come behind me and maintain it? I'd like to think so, but reality is I'm not an experienced web developer. I'm likely to make bad choices, especially in the beginning. Especially if mom and pop are breathing down my neck because they hired me to make them a web site dammit. It's been a week, where is the damn thing already?
I'm not saying you're wrong of course. More choices is a good thing in theory. For the expert, it's nearly always a good thing. This doesn't change the reality of the situation though... most people aren't experts. Even the ones working in the field. Sit an inexperienced kid, an enthusiastic amateur, or a sys admin who needed a job (now) in front of the current embarrassment of choices, and we're all likely to make bad choices. Probably for the best of reasons. Then we're going to go get new jobs, and the next kid/amateur/sys admin is going to make new bad choices and pile them on top of the old bad choices with a little duct tape and some Gorilla Glue.
It's a thorny problem. Fewer choices is bad, but the current situation isn't great either. Luckily I'm a sys admin, so you developer guys figure it out :-P