The pleasantness of Om

The pleasantness of Om(danielsz.github.io)

63 points by danielszm 12 years ago | 22 comments

Udo 12 years ago |

I know the use of object-to-HTML-mappers is only going to grow from here on out, but I really don't see the appeal in most cases where pure HTML would have been a perfectly readable and lightweight choice. We're becoming so scared of angle brackets that we seek refuge behind object-like constructs that are nothing but yet another impoverished and leaky abstraction layer above an already abstract markup language.

Look at that nightmare of a rendering code and tell me you'll still want to touch that when the time comes to go back and modify it in a few months.

I had the displeasure of working with Jade on a recent project. Not only was it more difficult for humans to parse than the HTML it produced, it was also slow to execute and generally cumbersome to modify. The main purpose of its existence might just be so programmers can lie to themselves about separating code from templates, and making those templates look really sophisticated at the expense of productivity.

The battle cry "nobody should write literal HTML anymore" is not a good enough reason to do this - in fact it's not even a reason at all to begin with.

The example code in the article, well, I'm sorry to say that it's unreadable, convoluted, and huge for what it actually does. Both versions are terrible in their own way, only the JavaScript one is also deliberately sabotaged in order to really stand out as an abomination.

That being said, if there are people looking at this going "my, that does look mighty pleasant" - all the more power to you. A huge part of programming is finding technologies that map well to your brain. It just doesn't work for me at all (I needed to vent that just now, thanks for bearing with me) and I kind of hope I'm not alone.

woah 12 years ago | |

It's pretty indisputable that jade, haml, etc takes about 10% of characters to write vs plain html. There is nothing in html that can't be expressed in jade, and if you would prefer to use html you can put it in as plain text behind a | (I often do this for links in text). The lack of closing tags eliminates the most common cause of html mistakes completely.

If you don't like jade, I can respect that, and most of the time I use html for work because of factors that I can't control and it's no problem.

But you need to come to terms with the fact that your dislike of the language is caused entirely by your (I'm guessing) long experience with html, which has made you unwilling to gain proficiency in something new. Jade is objectively simpler than straight html.

benaiah 12 years ago | |

It actually looks much more pleasant than HTML. The syntax is almost identical, except that HTML is much more verbose and difficult to read. This is a very different beast from Jade - Lisps translate to HTML very easily and transparently.

mercurial 12 years ago | | |

I'm not the world's foremost Lisp advocate, but you're absolutely right. It's even better than HTML in the sense that you don't have the "sometimes close the tag, sometimes not close the tag" inconsistency issue.

dustingetz 12 years ago | |

The difference is that a traditional approach requires you write code that updates your templates (the function `update` in his javascript example). The React version has no analogous logic. So you end up writing less code, and the code you do need to write is simpler.

catshirt 12 years ago | |

"nobody should write literal HTML anymore" isn't really relevant. React actually goes out of it's way to enable HTML syntax with JSX.

the templating is subjective. i for instance prefer writing my template in the same language as it's logic.

IanCal 12 years ago | |

> only the JavaScript one is also deliberately sabotaged in order to really stand out as an abomination.

I hope this is the case, otherwise someone somewhere thought this was the best JS code to solve the problem:

for (i = _i = _ref = this.length - 1; _ref <= 1 ? _i <= 1 : _i >= 1; i = _ref <= 1 ? ++_i : --_i) {

pselbert 12 years ago | | |

I would guess this is JS transpiled from CoffeeScript or something similar. That makes it an even less fair comparison, as ClojureScript would compile down to something just as wacky looking.

paulwithap 12 years ago | |

Weird. I've come back to our Jade templates months later, thankful I decided to use to Jade. My only beef with it at the time was the lack of documentation, but that seems to have improved since.

twfarland 12 years ago |

Om looks interesting for more complex stuff, but I agree with others that the jquery version shown for comparison is a bit baroque. Here's a simpler one for comparison https://gist.github.com/twfarland/9285304 Short, magic-free, and sufficient for something this simple.

bshimmin 12 years ago |

> Is it a rhetorical question were I to ask which of the two versions, Om + Clojurescript vs vanilla JavaScript, looks more pleasant?

Well, the JavaScript is harder to read because of the colour choices and the fact that at least in part it looks to have been compiled from CoffeeScript (or something else).

Conversely, the ClojureScript is probably quite hard to read if you don't know what "peek", "reify", or "did-mount" mean.

masklinn 12 years ago | |

* did-mount is a mapping to React's componentDidMount[0] lifecycle method (and, in Om, part of the IDidMount protocol[1])

* reify is the creation of an anonymous type instance, implementing 1..n protocols (not entirely dissimilar to an anonymous class in java). This could also have been done with a deftype (no difference in efficiency, but Om's creator seems to prefer reify):

    (deftype Widget [data owner]
      om/IRender
      (render [this]
        ; render code here
        )
      om/IDidMount
      (did-mount [this node]
        ; mount code here
        )

* peek's a bit more complex because it requires knowledge of interesting features (?) of Clojure collections: all clojure collections are immutable, and often built with conj(oin). conj "adds an item to a collection", but where it does so depends on the collection (and its efficiency profile): (conj '(1 2 3) 0) -> '(0 1 2 3) but (conj [1 2 3] 0) -> [1 2 3 0] (lists can be efficiently extended from the head, vectors from the tail). `peek` essentially returns the last conj'd item, so on lists it's equivalent to `first`, on vectors it's equivalent to `last` (but faster)

[0] http://facebook.github.io/react/docs/component-specs.html#mo...

[1] https://github.com/swannodette/om/wiki/Documentation#wiki-id...

jah 12 years ago |

Om and React are cool, but this comparison is disingenuous. The author should have compared the Om implementation to a native (JS) React implementation.

I recently converted a React project to Om and found it to be a very enjoyable exercise. However, contrary to most tweets/blog posts, my Om code ran slower than the JavaScript version, which ultimately caused me to abandon the conversion. Full disclosure, the port was a naive implementation of a single page React component which probably wasn't the best use-case for Om ( (om/transact! ...) was causing a full page re-render to the VDOM).

Om/cljs is very beautiful and concise compared to JS ... immutability rocks! If I could achieve performance comparable to JS I'd switch in a heartbeat.

bostonOU 12 years ago | |

Is your project open source? I'd love to see what you're doing. From my understanding of React/Om, Om shouldn't cause a full re-render if React wouldn't.

The immutability of Om/cljs means that the code for determining re-renders (shouldUpdate lifecycle method) is (always?) faster than React.

brandonbloom 12 years ago |

I'm a huge proponent of both ClojureScript and Facebook's React. However, this is a severely broken comparison.

jbeja 12 years ago | |

Its give me sympathetic shame.

nickfargo 12 years ago |

The "looks more pleasant?" test can be valid only between cljs/Om and the CoffeeScript from which the generated JS was compiled.

By presenting Coffee's not-quite-idiomatic JS output as the counterpoint, the otherwise well made argument for Om's pleasantness is weakened.

elwell 12 years ago | |

yeah, it's riddled with '_ref's and the like.

stevejohnson 12 years ago |

Your post content keeps moving up and down when the ticker changes size. Makes the post unreadable.

darklajid 12 years ago |

That sample widget was not a good idea. Tried reading the article on mobile & tablet, but I cannot: Every iteration showcasing a different testimonial changes the height of the widget, causing the _complete_ article to jump up and down.

Might be only a problem on a device like mine (S3 or Nexus 7), but it completely breaks the whole page.

Cless 12 years ago |

That JavaScript code is ugly because either:

1) the author tried to make it ugly. 2) the JavaScript was generated by a compiler.

Or, I suspect, both.

elwell 12 years ago |

Widget delays the load of the rest of the page.