Live data from Hacker News

Web Development: A Crazy World

rubiken.com

21–30 of 206 posts

Re: Web Development: A Crazy World

#22
This piece first appeared as a comment on this page in ENGLISH in July 2012: http://www.zemanta.com/blog/i-bet-you-over-engineered-your-s...

while the author claimed to have it translated from Chinese back to English, surprisingly almost exactly the same as the original English version. I guess English-Chinese-English translation has reached an amazing stage!

Re: Web Development: A Crazy World

#23
post #13

I agree and can see where OP is coming from. As a Chinese, I notice the difference in how we approach problems. Let me make an analogy: Chinese and western dude learns guitar. The Chinese player will worry about the scales, the hammer on/off drills, chord drills, picking drills, and maybe practice a song and focus on technicalities; western dude would pick a guitar, learn a few chords to play a song, realise that to…

You're right but I think this is a bit strong and perhaps a little generalist.

Anyway, IMHO it DOES matter which is best for you because if you choose the wrong one, and you're project is big and complicated enough, you'll spend a long time dicking about trying to fix the 'deficiencies' of your poor technology choice later. So in some sense the OP is right too.

IMHO, as with many things, a little moderation in both is the key. Do just enough analysis to meet your requirements, and future proof your choices, and then leave the crazy web world where you found it.

Re: Web Development: A Crazy World

#24
post #17

As with all technology choices, you've got to draw the line somewhere - new and risky, or old and proven, or somewhere in between. What matters these days is probably how well you and your team already know the technology, how good the documentation is, community size/takeup, how quick the devs are to respond to issues, and how mature and stable the API is from release to release. How popular/old the technology is al…

I used to develop with JavaScript, and JavaScript with Backbone before we switched to ClojureScript.

Backbone helped me to better structured my code but I found it also brings its own problems. There is a quite a lot of boilerplate code and views composed of other views were for me hard to manage.

I also wrote some macros to use Backbone from Clojure but the mismatch between object/mutable and functional/immutable is too high.

Finally the approach from http://clojurescriptone.com/ seems promising. It uses a simple yet flexible event dispatcher. Add that a templating engine and you are done.

Debugging ClojureScript within the browser is harder than normal JavaScript but it's easy nonetheless to find what generated JS functions corresponds to your ClojureScript code. If you compile in debug mode you can set a breakpoint as usual. Compiling ClojureScript adds a bit of overhead to the workflow but I compiled it continuously in the background via Emacs and forget about it, it just popups when an error occurs.

In general I found ClojureScript much cleaner than JavaScript and it's easier for me to write cleaner code with it when the data manipulation or the logic is not trivial.

Re: Web Development: A Crazy World

#25
post #13

I agree and can see where OP is coming from. As a Chinese, I notice the difference in how we approach problems. Let me make an analogy: Chinese and western dude learns guitar. The Chinese player will worry about the scales, the hammer on/off drills, chord drills, picking drills, and maybe practice a song and focus on technicalities; western dude would pick a guitar, learn a few chords to play a song, realise that to…

Agree~

Re: Web Development: A Crazy World

#26
Yes, this

The amount of 'meetoo.js' is amazing(ly bad)

Because of course everybody has to have "MVC responsive html5 cascading containers with chocolate covering"

Not to mention most of these are underdocumented, bug-ridden, too specific, etc

Need a js library? JQuery. period (and don't get me started with mootools, I need to ship, not swim around their docs figuring out how to do what in jquery is easy )

And focus on the backend, a competent backend development will eat your whatever.js "specialist" except for the most specific cases

Re: Web Development: A Crazy World

#27

Quite amusing and very true. Just had a little look at Handlebars - writing 3 to 4 times as much to generate very little. Also reminds me of this tongue in cheek site : http://html9responsiveboilerstrapjs.com/

3 to 4 times as much as what?

Re: Web Development: A Crazy World

#28
post #13

I agree and can see where OP is coming from. As a Chinese, I notice the difference in how we approach problems. Let me make an analogy: Chinese and western dude learns guitar. The Chinese player will worry about the scales, the hammer on/off drills, chord drills, picking drills, and maybe practice a song and focus on technicalities; western dude would pick a guitar, learn a few chords to play a song, realise that to…

I has nothing to do with being Chinese or not. I'm a "western dude" and I feel like it's my job to evaluate all of those things because picking the right tool IS my job. If I don't then I am doing a disservice to myself and clients.

I wouldn't want to get surgery by a doctor who said "well, I can do the operation in 5 minutes instead of 5 hours if I knew how to use Hyper Awesome Super Laser but sorry, I don't know how to use that technology".

Re: Web Development: A Crazy World

#29

Yes, this The amount of 'meetoo.js' is amazing(ly bad) Because of course everybody has to have "MVC responsive html5 cascading containers with chocolate covering" Not to mention most of these are underdocumented, bug-ridden, too specific, etc Need a js library? JQuery. period (and don't get me started with mootools, I need to ship, not swim around their docs figuring out how to do what in jquery is easy ) And focus o…

> And focus on the backend, a competent backend development will eat your whatever.js "specialist" except for the most specific cases

These days backend is pretty straight forward, mature, and usually easy compared to client side.

Re: Web Development: A Crazy World

#30
I've found that a large subset of programmers share the same trait as a large subset of photographers: they become so obsessed with tools that they lose sight of the fact that the end product is what matters at the end of the day.

For example: there are many people who spend thousands on cameras and gear, and still take terrible photos. There are many developers who spend a lot of time keeping up with what they think is the latest trend, and yet they never finish any projects that get used.

Then you have the photographers who take great shots no matter what gear they use, and developers who write and ship great code and products using Java and other unsexy languages.

Can better tools help you be more productive? Absolutely. But if you spend all of your time worrying that you're not using the latest and great tools, you won't get much done, and you won't be satisfied with what you do get done. I have 15 cameras and unfinished projects in 10 languages that demonstrate that.

My advice? Sit back, pick technology a couple of steps behind the bleeding edge, and focus on results. Choosing Ember over Backbone isn't going to cause your project to fail; building the wrong thing or failing to finish, however, will.

Post reply on HN