Ruby and Rails for Attractive .Net Developers | Hacker News Reader

Ruby and Rails for Attractive .Net Developers

david.ing.name

38 pointsdaviding11 comments

Open articleView on HN

I use both Ruby and C# daily. I love them both. My only complaints about Ruby and the Ruby ecosystem are these:

1) Minor versions of Ruby are incompatible with each other. 1.9.1 to 1.9.2 requires some code changes.

2) Developers of non-popular gems often lose interest and stop maintaining them.

3) WPF is truly a remarkable UI toolkit. I wish there was an equivalent.

What I love:

1) The Gem system makes moving your app from system to system trivial.

2) Most Ruby devs use Git/Github.

3) Ruby's expressiveness.

4) Metaprogramming.

If you use both languages, what are your opinions?

I too like both, but as you tell from the post, I use less C# recently for web development.

With the C# ecosystem there is more stability and often better introductory resources for learning. Most innovations will come out for .Net via Microsoft in some form or another (either 'in-housing' an existing community project or just plain reinvention). This 'Wise Parent' model of .Net community stewardship is both its greatest strength and its greatest weakness.

.Net and Microsoft are now starting their own package management project (brought in-house again) and it will be interesting to see how that works out. Their MVC framework is very good, produced in this model, but I can't help feeling the 'mainstream developers' that buy all those Visual Studio licenses aren't that interested.

I like WPF a lot, but compared to Spark in Flex, it's a giant bag of headache. If I could somehow force C# and Flex to have a lovechild with Spark for UI and C# for the code, I would be an infinitely more productive programmer.

1.9.2 was not a minor version bump.

Ruby indicates bug fix level changes with the patch level, so Ruby 1.9.1-p430 is compatible with the earlier 1.9.1-p429, but not entirely compatible with 1.9.2-p0.

"I’m not sure why Ruby doesn’t have named parameters yet (See disclaimer up top on my ignorance), but the language certainly seems alive and active, in that perhaps the next version will?"

Matz stated during his keynote at Rubyconf 2010 that Ruby 2.0 would have proper named parameters for methods, although in practice they will be largely the same as they are now.

Ah, thanks for that. I thought I might have missed them, but the symbols/hash convention is something I'm growing used too.

And hashes with symbol keys became a bit nicer in 1.9:

  Account.create(:firstname => 'Bob', :lastname => 'Smith') # old

  Account.create(firstname: 'Bob', lastname: 'Smith') # new

I've got to say, that's probably my least favourite of the new syntax changes. It adds a new, context-specific syntax for both symbols and associativity, hiding what's actually going on in the method call, to save 3 characters per parameter. That's a fair cognitive load and backwards-compatibility problem in exchange for a relatively minor typing and readability gain.

Maybe I'll get used to it, but it doesn't seem worth the change to me right now.

After using Ruby 1.9.2 for a few months now the new/alternate hash syntax has grown on me quite a bit, especially for keyword arguments. It typically ends up being much more readable (unless your value is a symbol (key: :value) reads awfully).

If you don't like it though, there's no reason to use it yet.

It does seem very much like a kludge to get pseudo named parameters working. It would be great to see this dropped as a hash syntax and used for only named parameters in 2.0. The genie does seem to be out of the bottle though :(

The summary does a good job of summarizing:

> It is interesting in that a lot of the ‘magic’ side of Rails is actually not that mysterious at all, in that follows the idioms of Ruby and that whole stack is still vibrant, changing and moving forward. At times it seems chaotic, but I think that’s part of why it’s still alive and kicking. So far so good.

Reply on news.ycombinator.com