Rails has always been bloated. That's what you get from frameworks that come with automation for common tasks.
However, let me provide a counter-point ... I think that Rails 3 is finally a good framework that I want to use. Blogs written in 5 minutes (which you can still do in Rails 3) have never impressed me.
What impresses me is code-reuse, modularity and readability. Which is why I preferred Django instead of Rails 2. The biggest reason was that I can read Django's source-code, but Rails (prior to 3) was really awful in that regard and cargo-culting is OK, but only when you're starting out.
Node & Coffeescript are indeed cool and have their place. However I prefer mature frameworks that save me effort when building yet-another-admin or yet-another-facebook-login.
And I think one of the things that I would, that would sort of come first to me is this idea of, don't just believe that because something is trendy, that it's good. I'd probably go the other extreme where if something, if I find too many people adopting a certain idea I'd probably think it's wrong or if, you know, if my work had become too popular I'd probably think I'd have to change.
Contrarianism is not an end in itself. If you deliberately take the polar opposite of what is popular, you're following fashion just as much as someone who always has to be using the new hot thing.
How about we just use the tools that allow us to both do a good job and enjoy doing it.
If part of your enjoyment comes from feeling that you're using the optimal modern configuration, that's great. If part of your enjoyment comes from blazing trails with exciting new technology, go for it! If part of your enjoyment comes from using familiar tools in the way you feel most comfortable, that is also sweet.
In all likelihood, when you're building a compelling and performant web application, they will each serve more than adequately.
The idea that "If my work had become too popular I'd probably think I'd have to change" is terribly cynical.
> The idea that "If my work had become too popular I'd probably think I'd have to change" is terribly cynical.
Not really. For some people, the goal is to do useful work that only you can do. When your work becomes mainstream, that means there are people willing to take it and run with it. At that point, you've made a great contribution. Leave it to the next generation and move onto the next project. It keeps you from being complacent, keeps you moving forward and creating new things.
Case in point - how many new things has Steve Jobs created since he re-joined Apple? How many new things has Linus made since he started Linux? Two very different approaches to projects and work, yet I wouldn't call either of them cynical.
I think it's the ideal example for demonstrating a web framework.
1. It gives basic introduction to the framework and installation notes.
2. It's simple enough to follow, yet large enough to
show major aspects of development with rails.
3. It shows all major aspects of the web application development - db access layers, db creation and migration, models and validation, request routing, controllers, views, partials...
> What impresses me is code-reuse, modularity and readability. Which is why I preferred Django instead of Rails 2.
I use both Django and Rails. Rails has various stuff which I need for a basic app out of the box, which Django doesn't have - asset pipelines, coffeescript integration, swappable templates, bundler for dependencies, rvm(not exactly rails) for isolating ruby version and gemset, tested gems for basic tasks(kaminari, devise, paperclip) etc.
I know I can do all of it in Django/Python(pip, virtualenv, coffee --watch, webassets etc etc), but rails sets up things for me and I can begin coding.
Rails does require more peripheral knowledge, but it's optimized for people who know rails. Beginners will be more at home with Django, but you are a beginner for a very short period of time, and optimizing for beginners while forfeiting comfort for people who are on-board isn't a good compromise IMO.
In an ideal world, we can have frameworks which is both beginner and power-user friendly, but I seldom see it happening it in the real world.
As far as reading code goes, I have trouble following Rails code, and my only friend is the interactive debugger - I set breakpoints, read code in the editor, continue, break...
I haven't read Django's code in a while, but it used to be quite heavy on magic - not sure about the current status.
Notice how I said that Rails 3 is finally a framework that I like.
Also, Django may not be as feature rich, but it does have the Python ecosystem, which is more comprehensive as a result of being used in academic environments.
> I haven't read Django's code in a while, but it used to be quite heavy on magic - not sure about the current status.
Those times are over, fortunately. Whenever I encounter a lower-level problem I tend to look through the code rather than the documentation, because it's usually much easier to find what you're looking for.
What it ultimately comes down to is you do what you feel is best for you. And this has been said many times, but, a tool is a tool is a tool. Whether it be js, Ruby, Python, Assembly, Lift - they're all tools to get the job done.
I started a new project a few months ago and spent a day almost agonizing over 'the new hotness'. I talked with a friend about doing it the 'old way' (Rails mvc) or the 'new way' (backbone, node, etc). In the end, I just wanted to get my work done, so I chose the 'old way'. I don't see anything wrong with this. I feel like I'm still creating, I love getting up every morning and making progress on the app.
At some point, I'll probably go back and revisit. Maybe start sprinkling in some new hot here and there. The one thing that DHH has said that's really stuck with me is (paraphrasing): "Love the tools you're working with". So, I leave it at that.
After everyone saw how it easy it was to write your own blog in Rails, everyone ... used Wordpress anyway.
The Rails use case was always truly limited. I still don't know any application that achieved maturity and stayed coded in Rails. For a while that was Twitter, but then they started moving stuff over the jvm.
Node will probably end up in a similar way, because at the end of the day no one wants to manage thousands of lines of Javascript (or any other language that doesn't enjoy the benefits of a statically typed ide).
1) Meh on node. I don't really see the advantage of using node aside from specific and well defined performance requirements. It's ironic to be saying this coming from Rails, but it seems like Node's toolset is somewhat immature. Is there a good way to use plugins in Node? There must be. Anyhow, server side? Meh.
2) That said, Javascript isn't going away. We will all be managing thousands of lines of Javascript because… that's where applications are going - client side. Good luck convincing all the browser vendors to swap.
Anyhow, I know you're just trolling but you have to up your troll game - 98% of apps fall in between "toy demo" and "twitter".
In my experience with Ruby development (2 years), I have never, ever experienced a single issue caused by duck typing. Static typing is great for performance, but it has nothing to do with being able to maintain a project.
Have you worked with more than 5 developers on a rails project? I ask because all the problems I've had with dick typing appear around interfaces between code from 2 programmers, and 5 seems to be the threshold where communication starts to break down.
The "Blog in 5 Minutes" demo was just that, a demo. It was a showcase of how easy it is to create simple CRUD apps in Rails as compared to competing frameworks (or DIY stuff) of that era. Even today, it remains a decent beginner's example that touches on many aspects of MVC.
I don't think you really believe a web development framework and a feature rich blogging product were actual competitors but are simply creating strawmen to support your view. Drop the crap.
I love Giles. His first Archaeopteryx talk was one of the best Ruby-related talks of all time. He's way too harsh on Rails though. The original "blog in 15 minutes" demos that catapulted Rails into the limelight are still possible with Rails 3. Things like 'gem install rails' still work. If you are determined to skip bundler, you can do that too. What's more, unlike 2005, in the 15 minutes it takes to build a Rails blog, you can deploy to Heroku and have a live site running.
Node is fun. V8 smokes MRI. But Node's web framework ecosystem has a long, long, long way to go before one can reasonably say that Rails is a bad investment. If anything, we should be celebrating the number of options that are open to us. It's a great time to be developing for the web. Good things will come from Node's ascendancy, whether its frameworks dethrone Rails or not. The negativity is unnecessary.
I'd go further and say, that it will be very hard to have a good MVC framework with Node before the async callback mess is brought under control. Maybe we'll even need something like Iced CoffeeScript for this.
I'd caution against any "I'd go further to say" statements. Either you know because you've tried, or you don't and are simply guessing.
As someone who has made an MVC framework in Node.js and launched a production site using that framework, I can assure you that it's not hard to avoid async callback mess. I can also assure you that many others have had success doing this.
It's really not difficult to write beautiful, modular, and readable code in node.js without going past 80 columns and without using Iced CoffeeScript.
I suspect that comments like this only further reinstate the theory that most people are dismissing Node primarily due to the annoying amount of hype. There are better reasons to not use Node, but imo, this isn't a very strong argument.
All the examples, github repos, books, and tutorials I'm seeing with my dive into node say exactly the opposite - the code becomes a mess of callbacks. It looks like enormous work goes into avoiding callbacks, similar to the amount of work that went into Rails in 2005 so that server-side folks could avoid writing JavaScript.
I'd love to see this framework you wrote. Is it open-source? or is it something you're unable to share?
You have to think about async code differently than you think about synchronous code. In PHP you might write:
$data = getData();
And people want to substitute the return for a callback in js.
getData(function(data) {
// Do something with data.
});
That indeed does lead to callback mess. But when you are coding in js you rarely should use anonymous functions. I mostly just use them when I might want to perform recursion. Going back to my example, I would write the operation as an object like so:
Wrapping things in objects doesn't seem to help me.
I'd still end up with this, thanks to callbacks.
app.get('/documents', function(req, res) {
Document.find().all(function(documents) {
// 'documents' will contain all of the documents returned by the query
res.send(documents.map(function(d) {
// Return a useful representation of the object that res.send() can send as JSON
return d.__doc;
}));
});
});
Seems to me that what's missing is something that lets me write it in a more simple way. What I have here looks like an absolute nightmare of coupling. So if you have some suggestions you could point me to that show a little more detail, I'd be really grateful.
I would create a DocumentRequest object that takes your req and res, then break apart your callbacks to separate functions on the DocumentRequest prototype. Then you would simply consume it with:
The async callback mess is not really a problem, there is about zillion patterns and libraries to handle it. The problem is error handling in callbacks, capturing stack traces from the async parent callers, and generally not crashing Node if you forget to try/catch your async handler. However, this problem is known, and this is being worked on in the form of "domains". Btw. Iced CoffeeScript doesn't solve this a single bit.
If you have callback hell in your code, it's not because the problem hasn't been solved, but simply that you didn't put the effort to npm install a simple library.
I agree that it is a problem if neglected, but most people don't allow it to be a problem in the first place.
It's like knocking on PHP because it could be used for templating and business logic on the same file. Just because you could, doesn't mean you have to (or should).
I honestly don't feel that it's as big of a problem as people make it out to be. It's simply a minor annoyance. The error handling is the killer.
Could you give a brief code example how monads eliminates the problem of error handling? Do you mean to put try catch in the monad around every async step it takes in? Just the API for how this would even look would be great.
this is just a sketch to give you an idea of what I am talking about.
right well you've probably used jquery before, so the api would look something like this. You start with a sequence of async actions you want to perform.
Now this is a bit different from jQuery since all the methods here do is push method names and arguments onto a stack. it's the .end() method that reads in the stack and goes through and executes it. This means that each method along the way we've constructed a partial computation value- that is, the stack, which we can pass around as a value- just leave off the "end" method...
var scrapeHN = $(ctx).geturl("http://example.com).parseExample().renderTemplate();
now we can modify the computation by turning it into an error handling monad...
var ecm = ErrorCatchingMonad(scrapeHN, errorCallback);
this wraps each method in the monad with a try catch block, and calls the callback if an error gets thrown, and stops the chain from getting executed further. in fact this is how the origial monad was created
var $ = DeferringMonad(scrapingAPI);
which takes a normal jquery style plugin interface/chaining api and turns it into that deferring stack type api with an end(); method.
real answer: It's easy and fun to whine about things you don't get in a language/paradigm you don't know. So easy and fun in fact that's almost entirely what programmers talk about to each-other. But whining hardly gets anything done, does it. The answer is: It's not a problem if you know what you're doing. And that applies to all other instances of this programmer discourse pattern.
This came across to me as a long rambling rant which amongst other things interchangeably referred to ruby as both "ruby" and "rails", but i got the feeling it was more about bashing rails.
Rebuilding something because the older stuff is is "bloated" and the new stuff is somehow superior always seems like a weak excuse to me. The new stuff quickly becomes the old "bloated" stuff and you're back to square 1. In fact I have no doubt that within 5 years something else new will come along and perhaps Giles will be writing another article about "Node Went Off The Script: Why I'm Rebuilding Clyde In TheNextBestThing".
Personally I would rather concentrate on making something which is actually good in its own right.
In fact I have no doubt that within 5 years something else new will come along and perhaps Giles will be writing another article about "Node Went Off The Script: Why I'm Rebuilding Clyde In TheNextBestThing"
Which possibillity he mentions in TFA.
But this is what we call progress (or, to a more bitter viewpoint) change. It doesn't stop.
tl/dr Rails is not the new hotness anymore, the Rails 3 rewrite was a waste of time and it's strayed far from its initial ease of use. Node & Coffeescript, the new hotness, are awesome, here's my new video series, the first one's free.
Funny, I've mentioned "ease of use" and "rails" to a few rails friends, and most have said something along the lines of "Rails was never pitched as easy to use - I've no idea where people get the idea that Rails was intended to be easy to use!". It may be just them, but I've heard it from more than a few Rails people (though not universally).
Personally, it feels a bit revisionist, because I distinctly remember the Rails fans in 2006/2007 talking about how easy Rails made web development. But... perhaps my mind is playing tricks on me as I get older.
Rails was easy to use/learn in 2006/2007, and, coming from a background of 5 years of writing Java web apps, it was vastly easier than doing that mess.
I'm not saying it was or wasn't easy to use or learn, but I've had a few people tell me that it was never promoted as such by Rails proponents, only that it was a "better" way.
Again, my memory tells me that supporters at the time promoted it as easy as well, but I'm not finding much particular support material from 2006 on Rails (perhaps my google/archive-fu needs work).
I feel like a lot of complaints about Rails 3 being harder to use are coming from Rails 2 developers who don't know it as well.
I've only played around with Rails 3 (I'm following Michael Hartl's tutorial) and have never used Rails 2, but I don't see how new Rails features like `has_secure_password` or the asset pipeline (which is super dead simple) make Rails 3 any harder. It seems like the opposite — perhaps someone who has developed on both major versions can enlighten me?
And as for overly ceremonial code, let me tell you about my balls. I actually have a rake task to scratch my balls. Somebody told me a good entrepreneurial programmer writes code to scratch their own itch, and I took them literally. The code uses a serial port to power an Arduino board. The Arduino's got a servo with a backscratcher attached. I keep it on a shelf at scrotum height underneath my desk. I literally need to type bundle exec every single time I want to scratch my balls. I don't like having to type bundle exec every single time I scratch my balls. I named the task balls, so I can type rake balls, because it rakes my balls, and Bundler pisses all over my tidy syntax, but am I going to complain? Am I going to point out that asking me to type bundle exec every time I want to scratch my own balls shows incredible contempt for me as a user, particularly considering that the whole reason I'm a user of this particular thing is because I work in a language which was designed to optimize for programmer happiness? Of course not. It's not productive or helpful to say that, so I simply type bundle exec rake balls and scratch my balls in sullen silence. Typing bundle exec every single time I want to scratch my balls is the epitome of the overly ceremonial code which Rails defined itself against from 2005 to 2007.
The problem with tl/dr is that it misses out on the funny. :-(
Correct. The problem isn't the difficulty, it's the education.
In my own project (SproutCore), we've used Ruby-based buildtools since 2007, and there is unanimous support to move to Node, precisely because running a modern Ruby stack requires so much education.
We want to spend time in our IRC channel on SproutCore itself, not Ruby and it's myriad problems (all of which have "fixes" that "aren't so hard").
We're also moving to Node because the performance of the Ruby tools sucks, despite three rewrites (one by a member of the Rails core team, so not Ruby n00bs).
The Node tooling? It's so fast that we can literally rebuild the entire project from scratch every time we reload the app and still load everything into the browser in under a second. Oh, and it took about a day to write. :/
That, to me, is why Node is taking off. The "I can get shit done quickly" factor is very high, the resulting code runs very fast, and the community is large, energetic, and helpful. Plus, pretty much everyone has to use JavaScript today.
> precisely because running a modern Ruby stack requires so much education
Education about what? Can you give some specifics here? I'm genuinely curious as my day-to-day ruby development is pretty smooth and I haven't been doing it very long.
or if that's too much use --binstubs to use "bin/rake balls" or if you are really, utterly crazy, use --binstubs AND set $PATH to contain './bin' and then use "rake balls".
You can also use rvm with gemsets and just not use bundle exec at all. I'm amazed how quickly people forget about the things that bundler tries to fix. If you don't have any conflicts across your gems, you can basically use bundler as "just install my gems".
I agree Rails has evolved to become many of the things that DHH first complained about in so-called Enterprise frameworks. But that's because a wider audience has wider problems to solve than Basecamp 1.0. At the end of the day, Rails 3.2 for all its complexity still "gets" server-side web development better than the antiquated Java frameworks that it replaced ever did.
What Rails really never got was client-side development. Even as it ushered in the era of easy AJAX support in 2005, it was based on a poor fundamental understanding of how to do good client-side development (to be fair, no one really had that understanding for javascript apps back then because we were lost in the weeds of browser compatibility). Over the years it's improved only in fits and starts, with bandaids and ancillary features like the asset pipeline and coffeescript (RJS was probably one of the worst ideas ever hyped).
So to me what the new crop of frameworks bring to the table is rethinking things from a deep understanding of client-side development, in much the same way that Rails rethought things from a deep understanding of server-side web development. I don't hold much hope for Rails to get good at client-side development due to the fact that DHH doesn't really like client-side development (as indicated by some of his commentary in the Basecamp Next caching article). He likes coffeescript because it's more like ruby and thus less painful, but the fertile ground of client-side development techniques will never be tilled by someone with this attitude. Javascript may not be the prettiest language, but it is powerful and paradigm-shifting when embraced to its fullest.
All that said, server-side ain't goin away, and Rails is still one of the best ways to develop server-side apps. If Rails was fast I would happily stick it on the server as I explored client-side development independently. Unfortunately I'm really starting to feel like Ruby performance is holding me back and that it will be impossible to fix regardless of how seriously Ruby and Rails core get about performance. Who was it that said "everything can be fixed with another layer of indirection, except performance"?
> All that said, server-side ain't goin away, and Rails is still one of the best ways to develop server-side apps.
I'm not sure. You only need a server-side if you have users, there are a lot of apps that don't need users. For those that do there's no reason why your server-side code needs to sit next to your web server as it does today. What you need is a good API framework, which Ruby does well today, but perhaps should be its primary focus in the future.
Has anyone made a comprehensive comparison of the top web frameworks and how mature the libraries/plug-ins/services around them are?
Frankly, Rails and Node.js seem to have much more in common than apart. Both are based around relatively inconsistent languages (though conforming to some BDD or Good Parts style can help), both have deeply flawed runtimes in terms of performance and especially parallelism, and they seem to have extremely high overlap in the people who program in both frameworks.
What I care about is libraries, plug-ins, and services. Specifically, if I have some new web task, what I care about is that there is some library/plug-in/service that exists to perform that task (e.g., user auth, Facebook API, exception logging, Heroku, resque, state machines), that the library/plug-in/service is mature enough that I can pretend it works, and that it has enough of a community around it that it won't break in the future.
Is there some comprehensive list of use cases, plug-ins/libraries/services, and relative maturity of those plug-ins/libraries/services across web frameworks? For example, the maturity of the various ORMs across frameworks, or the various testing tools, or the various Facebook libraries. It would be great to see it change over time, and to know when to jump ship to the next flawed language/runtime that has good plug-ins/libraries/services.
"It's a bloated meta framework that requires enormous amounts of peripheral knowledge to understand."
Unfortunately I think that might be because web development requires enormous amounts of peripheral knowledge to understand. It might be overkill for the first prototype of a new web app, but after a while that web app will probably acquire lots of peripheral hacks, too.
Glad you published this Giles. I really think you hit the nail on the head in your criticism of rails, and not necessarily ruby. I think some of the framework decisions in the 3.0 release were kind "abstraction for abstraction's sake" moves. In other areas, I think the framework has pushed beyond what the language is capable of handling in a reasonable manner, from a performance perspective (I'm talking about development efficiency performance, not production performance).
No doubt, these features were implemented with good solid principles in mind. But, imo, at this point in Rails' lifecycle as a framework (something that has been in public release for 5+ years) performance (both production and in development) should have held more weight in the decision-making process than doing things for what appeared to be because they were the "right" things to do.
For one, the startup time of the framework is just, atrocious. If it takes me 15-30 seconds to boot up the framework (and on my quad core dual processor 8gb ram MBP running a small-to-medium-sized project, it does) just to run one test, that is just plain ridiculous. A lot of that time is spent loading the large dependency graph of gems with which most new rails apps these days are packaged. I think this trend -- rigorous packaging open source library code as a first-class dependency, instead of writing it yourself, or snapshotting that actual code in your SCM -- is a good thing. But I think the performance characteristics of the ruby language, and of the ruby community, make trying to cram 150+ gems into the load path and then require all of them, serially, in a single thread, unacceptably slow for modern web development.
I will say that I think the situation has improved quite a bit more recently, and the focus seems to be honing in more on development performance, but it might be a tad too late for me. I'm focused more on building api's than I am building "web apps". For that, sinatra is just a much simpler, faster and elegant stack IMHO.
But here's the thing. How many technology choices does fashion dictate? I've seen fashion determine huge amounts of technology choices, and wheels get reinvented constantly in this industry as a result.
In 2006ish I did a talk about MVC JavaScript at a tiny FooCamp in Albuquerque. At the time people were all excited about Rails MVC, and I was like, "hey, initially MVC was for GUIs, and Ajax is headed towards GUIs." Because of that, it was easy to see that this particular wheel was headed for imminent reinvention.
It's easy to sneer at individuals for making technology choices based on fashion, but that's petty. Take a minute to look at communities making technology choices based on fashion. The periodic re-invention of wheels in programming mirrors the cyclical nature of fashion in terms of what people wear.
Embarrassing fashion confession: I used to be a raver in the 90s. I wore big baggy jeans. In the disco era, people wore pants like that. They called them flares in the UK and bell-bottoms in the US. Clothing fashions cycle, and programmer fashions cycle too.
If fashion is dictating an entire community's technology choices, you can learn something about the fundamental problem they're solving by asking yourself what wheel they're re-inventing, and looking into the history of computer science a little to see how programmers solved it last time.
I can't say much about Rails, but I will say that after doing a CoffeeScript, Backbone, and require.js with clientside compilation project -- debugging is pain and tears. Even though Chrome supports source maps, CoffeeScript doesn't, and the error messages raised by require.js are of the "you fucking peon" variety referred to in the article.
I like CoffeeScript and AMD, but they aren't ready to be in the same room together. Hopefully, jrburke's almond will let me build a serverside compilation strategy that isn't awful to debug.
On the other hand, Backbone kicks ass, and I wrote a DOM-twiddling Knockout.js inspired data binder called _outback.js_ to make it better. It's up on github (politician/outback) if anyone's interested.
I'm not ready to ditch rails yet, but I think the framework (and Ruby itself) needs an intervention. Bundler, Ruby 1.9.2, 1.9.3, and a ton of gems are all making life harder than it used to be.
I have faith the community is self critical enough to identify these problems and fix them. Posts like this are needed though to give us all a kick in the ass and make things better.
I'm not sure the community is self critical enough to fix the issues.
The maintainers are highly opinionated and have shown very little concern about the bloat up to now - mitigating it by targeted optimisations not by removal of bloat itself.
The community will just move on to something less bloaty - many Rails developers I know spend more (side project) time with JS now.
I don't think the author is hinting at ditching rails altogether. IMO, he is just pointing out his disappointments, starting at the start and doing what he thinks is good for him.
Having to learn the ins and out of bundler `bundle exec, update, --without :darwin`.. just to use a package handler. I had growl_notify installed, and now we've forced our team to be mac only because we got tired of dealing with linux issues.
Ruby Debug is impossible to get working in 1.9.3. The maintainer of ruby-debug has gone AWOL so we're forced to deal with solutions like this http://blog.wyeworks.com/2011/11/1/ruby-1-9-3-and-ruby-debug that don't work within bundler or a team environment.
1.9.2 + Rails 3.x is attrociously slow (30+sec). With falcon (a very unstable ruby hack) the startup speed goes to 10seconds. But 10sec is still way too slow
Sorry, it sounds bad when I just rant. In the grand of scheme of things, Rails makes me more productive as a web dev than I ever thought possible. It's just easy to pull out the daily frustrations and complain. I've been trying to help fix these issues but they are so dependent on a web of gem authors, and core ruby fixes that it takes an extraordinary amount of time.
I'm almost to the point where I think things need to be fixed top down. Matz needs to take control and try to make stuff not suck as soon as possible. Since he's working for Heroku now, it's in his best interest to push the platform and keep new users engaged.
Thanks for the heads up. I've searched far and wide to find any sort of docs for the built in debugger. ruby-debug was incredibly easy to use. require 'debug' sets a breakpoint, but I cant figure out any way to interact with it.
I use pry, however I've found it's very unreliable for stepping into shoulda tests, and often it errors out. I haven't been able to find stable alternatives to the most common ruby-debug commands 'l', 'e', 'n' and 'c'.
The problem is that there's no third option- something to manage dependencies that's not bundler. We all want and need dependency management, but bundler should not be the One True Solution.
I've used Isolate[1] pretty happily on a few things (command-line tools and/or Sinatra apps). I haven't tried it with Rails, so I can't speak to how well it works in that context. But the readme's first example is Rails, so it's worth a look if you really want an alternative.
Is autorequire even a real problem? You can quite easily put your gems in a separate group if you want to manually require them. Installing is another issue, that is opt-out, but I don't see the install process as the real bottleneck.
Write your own? I'm sorry but if Bundler doesn't solve your problem (or introduces other problems) either try and fix those issues OR write your own dependency management lib.
The unfortunate reality about writing your own new Bundler is that you'll inevitably end up rewriting Bundler. Isolate is basically just a mini-Bundler with a different (arguably simpler) storage location, but the storage location is really not where the complaints about Bundler are coming from.
Rewriting things just to shed complexity or "bloat" will often just put you in a position where your library just can't do things it will need to do, and then you end up rewriting those things and adding back the complexity. It's the circle of life. Lack of SCM support in Isolate is a good example-- I'm pretty confident that Bundler's support of git is heavily used and taken for granted, but if you were to rewrite or use a simpler dependency manager, you're throwing this out.
This point feeds back into the larger issue of Rails vs. Node, I think. Rails ended up with all the complexity it has now because it was [mostly] necessary-- not all of it was, but much of it was. The idea that you're somehow liberating yourself by shedding complexity is often just the short-sighted pleasure of running faster test suites. And it's usually followed by a little bit of claustrophobia when you realize you can't do X, Y, or Z anymore. Wait until the claustrophobia kicks in, because it will. And the solution to this claustrophobia? More cowb-- complexity.
I agree. I just hate to see this ― "zomg bundler sucks at <some feature usually speed>! I hate using it! It has to be fixed for me!"
Any open source project needs help. The maintainers of Bundler have 200+ issues on the issue tracker. If people could help them replicate and confirm issues, write failing tests and if possible write patches ― it would do a world of good. But instead it's EASIER to complain and whine. So...
getting ruby-debug to work with 1.9.3 was one of the first things I done after trying 1.9.3.. it wasn't hard or time consuming, I don't even remember what was the fix.. so I don't get this complaints...
I don't have bundle exec issues...
time rake environment takes 4 seconds(1.9.3, 3.2)
I agree with a lot of things in the article, but I don't have the issues a lot of people are complaining here..
agree plus add Rails is very well suited for the client side hotness with the asset pipeline in Rails 3.
However I just tossed Rails after many years for 2 primary reasons:
1) Anything that might take a sec or more has to be moved out of the request cycle and into a background process. That just gets annoying after a bit. Node.js can serve requests, grab data from to world's slowest API's, send emails, etc, all from a single process.
2) Although I think Ruby's strong point is readability and succinct LOC, it's hard to beat using the same language for everything and not context switching. Maybe my brain is broken, but I used to drive myself nuts writing client js and using Ruby's nil instead of null and Object.new instead of new Object.
For the project I'm working on now, I had to make my own asset pipeline. It uses less for css (and bootstrap), precompiles handlebar templates and concats all the js into a single file and uglifies it. Cost me 2-3 hours, about the same time it took me to learn Rails asset pipeline and was about 10x more fun.
Last point I'll add is when Rails was a baby it made fun of all the xml configs etc that other frameworks bogged developers down with. Now it's grown up a little, and although sprocket and yaml files aren't xml, the concept is the same. If you've been around Rails for awhile and dive through the code of project like Spree, you just scratch your head and little simple things like node appear shimmering sexy!
> For the project I'm working on now, I had to make my own
> asset pipeline. It uses less for css (and bootstrap),
> precompiles handlebar templates and concats all the js
> into a single file and uglifies it. Cost me 2-3 hours,
> about the same time it took me to learn Rails asset
> pipeline and was about 10x more fun.
Rails is not bloated it's modular and quite simple if you understand ruby and can read - look at the source. That said his to address his _complaints_:
* Rspec / Testing is slow as it has to reinitialize a whole rails stack every time it runs - run it async smartass it's what you're preaching, use inotify or fsevent with watchr or guard.
* Bundle exec for Gemfile changes? This guy is inept.. async yet again with inotfiy / fsevent
* HTML5 - It's a server side framework, if you want HTML5 you do it in the markup or the included coffeescript pipeline, why not pure coffeescript / javascript? because I value my time and sanity over the new hotness
> * Rspec / Testing is slow as it has to reinitialize a whole rails stack every time it runs - run it async smartass it's what you're preaching, use inotify or fsevent with watchr or guard.
> * Bundle exec for Gemfile changes? This guy is inept.. async yet again with inotfiy / fsevent
I like Rails, but these two points right here are bullshit. Rspec and Bundler being ridiculously slow are valid problems with those projects. Papering over the problem is not solving it, and I'd rather be polite than state what I think of calling people idiots for not papering over the problem.
You're not papering over anything, rspec is slow because it _has to_ initialize a whole rails stack if you want speed make it happen in the background - you don't stare at compiler output while compiling a large program do you? Again bundler isn't slow due to bundler being slow, it's slow due to IO.
> Rails is not bloated it's modular and quite simple if you understand ruby and can read - look at the source.
You've got to be kidding. The Routing has got to be the canonical example of bloated and confusing code.
Is inotify/fsevent built into the "full stack" framework? No?
"Hey guys, I wrote/found/whatever this code to solve a problem that another bit of code I chose to use caused for my development. Oh BTW you're a complete tool if you aren't already doing the same."
Confusing? you map resources to URL's it's confusing if you are inept. Inotify / fsevent and such aren't built into rails just as they aren't in _EVERY OTHER FRAMEWORK_ because it's complete voodoo to make side effects of editing a file default.
> you map resources to URL's it's confusing if you are inept.
Inept? Kids these days. Forking my code then calling me inept. It's precious is what it is.
But you're moving the goal-posts. You said read the source. Since you're clearly unfamiliar with it, here's a couple handy links to the Routing code I mentioned.
FTA
> I literally need to type bundle exec every single time I want to scratch my balls. I don't like having to type bundle exec every single time I scratch my balls. I named the task balls, so I can type rake balls, because it rakes my balls, and Bundler pisses all over my tidy syntax, but am I going to complain? Am I going to point out that asking me to type bundle exec every time I want to scratch my own balls shows incredible contempt for me as a user
Really? This what passes for A vs B discussion at the top of HN?
Actually, node is also a web framework. It's just not a web framework to build CRUD apps like Rails is tailored for. So, it all depends on how you define what web is.
This is not a fashion show, or "my framework is leaner and executes faster than yours" competition. "Rails is old and busted; Node.js is the new hotness." Seriously? is this the mentality we should adhere to? What happened to the problem solving, or building stuff? Old is bad, eh? If that is our credo we should quit javascript altogether because it is oldness itself!
Please don't get me wrong but I see a problem here. The attitude is quite problematic. You choose a programming language based on what do you want to achieve, and how that language helps you to achieve, not on buzzwords. If I want to build a weblog application, I won't try to learn c++ for example. Not because "it is old and bloaty and requires a peripheral knowledge", it won't help me to solve my problem. And I wonder, what kind of programming language does not require peripheral knowledge?
I can go on and on about how hard is it to do such a shift in a corporate environment and maturity of a codebase is good, and code is not a thing that rusts but I think I made my point.
tl;dr: you choose programming languages based on what you want to do and how that language makes your job easier; not because it is new or old or bloaty or "lean".
As far as Node goes, there is still no "holy grail" framework that can help one build a web app seamlessly across both client and server side. The momentum will definitely shift heavily towards Node if and when that happens. Right now, there is still a lot of manual glue work needed to set-up a client side MVC app using Node + Backbone while also taking care of graceful degradation etc.
That's called throwing the baby out with the bathwater.
There might be a few issues when your newbie & want to kickstart a rails app, but they're easily overcome with a good template.rb (the later being as small as possible, many of them are bloated).
Also I find the Rails 3.0 bashing cheap and unfair since it's a huge step forward in terms of robustness, modularity & coherence.
But yes I agree the fine tuning you need to get your quicky rails app up & running with your special flavor (eg: rspec, haml, inherited_resources etc) should be better advertised & explained.
I also spend more time on js nowadays but with backbone.js, throwing out the legacy of rails on the server seems a huge waste of time at this stage.
I guess the first of many things I don't understand about this post is that Archaeopteryx is not built in Rails, has nothing to do with Rails, nor is it a web site nor has anything to do with http nor even requires active_support or similar.
I know some people think Rails is going in the wrong direction - I disagree, but it's a topic on which reasonable minds can differ. This post may have something valuable to say on that subject, or not. Regardless, it's got nothing to do with the author's music library and I suspect the only reason Rails was mentioned at all was to stir controversy and bring attention to an otherwise unrelated announcement.
tl/dr What I take from the post: Guy has no fun with underlying technology, rewrites from scratch with more fun framework, does not care about current users, thinks this is offset by future users who can more easily use coffeescript.
Rails is missing the V8-like phenomena Node.js has.
Imagine if we would see something like V12 in the future, a JavaScript engine running on nVidia CUDA. That's where I would like to run my Node.js from ;-)
>Imagine if we would see something like V12 in the future, a JavaScript engine running on nVidia CUDA.
For that concept even to make sense, we would have to have massive leaps forward beyond what is currently even believed possible in the realm of auto parallelization.
Too short, expanding: Giles BS reasoning of what 2012 should be resonates so well with you, that you not only consider it the truth, but also "so yesterday, dude". i.e another fad-chaser.
Yeah, but you have to give the guy credit, his was the only comment that made me laugh.
And it's basically true. Not in the sense that this is new to me, I did a bunch with node and jasmine-node (etc.) in 2010, maybe earlier, but I went to a Ruby conf and was shocked how many people hadn't heard of it. even heard people saying you don't have to test JS because it's hard. so I decided the Ruby community needed to have a look at all the stuff that's already been going on for quite some time in JS-land.
Being old enough in this industry I have seen technologies come and go, while a few are able to stick around for a little longer.
The Node.js hype will be gone in the next 5 years, being replaced by whatever is hype, while giving me and other consultants lots of money to port existing applications to more mature enterprise level technologies.
I don't get the frequent comparisons (mostly by bloggers) between Rails and Node.js. The things you most want Node.js for you don't really want to be doing in Rails, and vice-versa.
I'm going to preserve the best thing I've read in awhile in a HN comment:
"And as for overly ceremonial code, let me tell you about my balls. I actually have a rake task to scratch my balls. Somebody told me a good entrepreneurial programmer writes code to scratch their own itch, and I took them literally. The code uses a serial port to power an Arduino board. The Arduino's got a servo with a backscratcher attached. I keep it on a shelf at scrotum height underneath my desk."
We are developers and we can learn and use any language if we just know the concepts of what we are doing. Same thing, different language... who cares. Hope it's the last "why I switched" blogpost because I never found them useful.
The people who learned with Rails always make me chuckle. The only reason Rails got anywhere was because of scaffolding and the hype that surrounded it - which of course already existed in some form in other platforms. It's completely revisionist to think that platforms like Java or .NET didn't already have application creators and ORM.
The genius of Rails and DHH was the marketing which isn't surprising seeing as how much he loves Apple for it. They can always try to say the other platforms were bloated, but this is coming from people who thought full Unicode support at launch wasn't necessary.
I don't know where you got that from. Struts was my first proper web framework- the kind where adding a column was an all-day affair.
The simplicity of Rails, even now, is that you'd only need to put it in two places- the migration and the view. You don't need to mess around with Hibernate XML configurations all day. Just add the damned column and get on with your life.
Recently, I actually started to learn the Ruby spec because I was offered a well paid work in Ruby for many months to come, and I thought it would be cool to work in something new. Besides web dev, my past is C/C++, then many years of Java, and recently Node.js. I turned the Ruby project down because very early down the spec I realized that I don't want to work with technology belonging to 5 (or at least 2) years ago, language design-wise. I'm now on Node and I have great fun. The JS community is great, the Node.js subset even greater. The NPM repository is exploding with new libraries. You can see a new framework literally every week, but you also have tested ones to rely on. Good patterns are taken from anywhere, anytime, and new ones are being created. I'm not saying this to bash Ruby, it's just my personal story. I approached Ruby with optimism, but what I saw was a badly designed language. I can understand it being so popular compared to Java, I know I would be Ruby coder today had I discovered it during my Java days. But today I have Node.
You refused a ruby job because the language (Ruby) was "5 (or at least 2) years old language design wise" and instead you went for.... node.js? Technology based on a language designed...17 years ago!? (God that makes me feel old)
I feel there is a flaw in your reasoning, or at least other reasons which you are not going into.
EDIT: I know ruby has been around a long time. Just found it first appeared around the same time as JavaScript.
I turned the Ruby project down because very early down the spec I realized that I don't want to work with technology belonging to 5 (or at least 2) years ago, language design-wise.
Ah, so you don't want to work with mature, proven technologies, language design-wise, that bring benefits to the company and the user base.
Good thing that you don't WANT to work with this kind of technology, because I too don't think you SHOULD be working anywhere near client projects with that mentality.
However, let me provide a counter-point ... I think that Rails 3 is finally a good framework that I want to use. Blogs written in 5 minutes (which you can still do in Rails 3) have never impressed me.
What impresses me is code-reuse, modularity and readability. Which is why I preferred Django instead of Rails 2. The biggest reason was that I can read Django's source-code, but Rails (prior to 3) was really awful in that regard and cargo-culting is OK, but only when you're starting out.
Node & Coffeescript are indeed cool and have their place. However I prefer mature frameworks that save me effort when building yet-another-admin or yet-another-facebook-login.
Not pursuing hotness is the new hotness IMHO.