News for hackers. Stuff that matters.
Re:Rails has won: The Elephant in the Room by vonklaus on Tuesday May 24, 2016 @12:09AM (#11758931)
I haven't used rails in about 3 years (when I first started programming) and I only built 1 application which was a super shitty weekend instagram clone that didn't really work well. I didn't want to use rails anymore because it didn't feel like programming and nodeJS was getting hot.

The thing is, I knew fuck all about programming then and now, I have the luxury of knowing very slightly less than fuck all. Which gives me the perspective that this article has, rails is pretty fucking awesome for building a crud application that uses a relational datastore. The author indicated ~80%.

Wordpress is also something I swore off when I was learning because it straight up wasn't programming. I installed wordpress on a server for the first time last week to setup a blog. I browsed a few thousand themes, it took me a while to find a theme for a blog. The just a blogging platform is actually more ironic now, than it ever was.

So, I agree with the author the design patterns shape the community and then the community shapes the design patterns. You can pretty much build any type of app with any language if you are motivated and talented, however, from my observation, the following languages seem to give you the highest leverage in the domains the community has evolved them for:

PHP/Wordpress: Blogging sites & CMS building.

Node/JS: Applications and APIs typically with a noSQL database.

Ruby/Rails: Building CRUD apps with relational datastores.

Python: Data intensive applications & programming.

erlang/elixir: Never used this, but seems ideal for highly scalable concurrent architecture and highly scalable process choreography.

So in essence, I agree with the author. I would love to switch from node to Rails as I like the framework aspect where you can just get productive super quick. I also see rails as going away, so I wouldn't make the time investment. I think learning node was the right thing to do ~3 years ago, but I wish I spent another 6 months using rails and got really comfortable with it first.

> This is not serious programming/you aren't a serious programmer

fairplay. I am a jr. web developer, explaining my conception of some frameworks/communities/languages, so I agree with you that I am not a particularly talented programmer (it isn't my core skillset) and I am not solving hard problems in this arena(as I don't conetend that I am). I disagree that I claimed any of the above was my definition of serious programming. Of the hard problems I will work on/what I think needs to be solved, I have spoken of at length elsewhere.

Re:Rails has won: The Elephant in the Room by greenleafjacob on Tuesday May 24, 2016 @03:11AM (#11759539)
> erlang/elixir: Never used this, but seems ideal for highly scalable concurrent architecture and highly scalable process choreography.

Erlang's primary reason for existence is fault tolerance. High concurrency and scalability is just a bonus. Consider hot code reloading, supervision trees, links and monitors, D/ETS and Mnesia, location-agnostic message passing. These have nothing to do with scalability. You use Erlang when you absolutely have to have the system continue to work 24/7/365 in perpetuity even in the presence of software errors.

Re:Rails has won: The Elephant in the Room by davidpatrick on Tuesday May 24, 2016 @12:49AM (#11759040)
It sounds like you are just trying to justify your decision for Node. I use Node and Rails in production for different products/services. They both serve different purposes well.

"Going away" was said about .NET and PHP and they don't seem to be going anywhere, albeit they are less popular than they use to be.

Too many discussions on HN have this paranoia lurking around of whether or not "I chose the right path". I find it unhealthy, and find that the longer life of a product exposes the weaknesses of any language/framework.

Re:Rails has won: The Elephant in the Room by mastazi on Tuesday May 24, 2016 @12:59AM (#11759079)
Language/framework fads are just that, momentary fads. I believe that, in the long term, things like good architecture, sane dependency management, well-managed life cycle (which includes provisioning and deployment) are far more important than the language/framework you are using. You may not like PHP, for example, but I would still choose to work on a good quality PHP project than a bad one in the latest flashy language/framework, any day.
Re:Rails has won: The Elephant in the Room by rvense on Tuesday May 24, 2016 @03:23AM (#11759574)
I don't know. I just ripped out basically half the rug under my Scala app and put in a new one without the rest even noticing. It's not a very good architecture I don't think, but because of flashy language features like generics that are often dismissed as unimportant or even an impediment "in the real world", I was able to change a core part of the code that basically everything else touches and still be sure that everything works exactly as before once the code compiles again.
Re:Rails has won: The Elephant in the Room by mastazi on Tuesday May 24, 2016 @07:36PM (#11766485)
I don't want to dismiss the importance of certain language features. I'm just saying that any Turing-complete language with a reasonable ecosystem is good enough in most situations (where good enough != perfect but hey, perfection is the enemy of the good!).
Re:Rails has won: The Elephant in the Room by vonklaus on Tuesday May 24, 2016 @01:00AM (#11759085)
Again, I am not trying to justify my decision, I think if you asked most developers to give advice to a kid learning programming 3 years ago how to get started(with the benefit of 3 years glance into the future) they would probably lean towards node if it was between nodejs v ruby.

However, I totally agree that the longer life of a product exposes the weaknesses of any language/framework. and there are multiple paths. Just that, if you want to get good at writing software you probably need to pick a starter language with a versatile community. People still write cobol, and wherever there is code in production, there will be demand for a language, but .NET and PHP are probably not growing in demand. I actually don't know about .NET tbh.

However, there is such thing as choosing a path as, without extensive time investment, skill and knowledgebase, it would be difficult to be a truly capable python, .NET, javascript and rails dev.

It would be wise to be capable in those languages, but to be a master or attempt mastery of the one most likely to embody your interests/career would likely be a better move than being able to be less than mediocre in numerous languages. I think we are largely in agreement, and I hope I didn;t mischarachterize you.

I think rails is cool, and I now regret not learning it as I do like the scaffolding nature and I am sure I could pick it back up quickly, I just do not have a need to right now. Elixir looks super cool though and if I do learn a new language that is markedly different it is a toss up between Go and Elixir. I am messing around with python right now, but javascript and ruby are pretty similar, and pyton seems fairly similar. A functional language or a compiled language will be a radical departure from what I am used to.

Re:Rails has won: The Elephant in the Room by pjmlp on Tuesday May 24, 2016 @01:51AM (#11759223)
.NET is alive and well.

Enterprise only cares about Java and .NET as standard backend stacks, regardless of the technology of the day HN posts how everyone is doing Go and node.js.

Re:Rails has won: The Elephant in the Room by vonklaus on Tuesday May 24, 2016 @02:00AM (#11759257)
i actually have been interested in Java because apache has open sourced what seems to be the best collection of resources on the internet and they are almost entirely written in java for the java ecosystem. I also think I would learn a massive amount. I also am strongly considering Java and less strongly C. What are your thoughts on Java in terms of:

* How much it would teach me about programming/software design, e.g. just as a general learning excercise?

* What does Java lend itself best to/is a typical set of programs a beginner could build after say 1 month?

* Is google planning on phasing Java out and using Go?

Re:Rails has won: The Elephant in the Room by reitanqild on Tuesday May 24, 2016 @03:52AM (#11759669)
1. Java has had huge amounts of open source code for years. (Case in point: course provider sales representative telling my colleague somewhat annoyed that Java people are the worst because "they are used to get everything for free.)

2. Im going to read that as "... 1 month after finishing studies".

I will recommend starting somewhere where you get good colleagues. When I finished school I was sure I could never work with Java: it was hard and unforgiving compared to VB and PHP that I had picked up on my own.

After getting introduced to ides and working build systems however I was starting to get work done within a week and within a year I was confident enough to make a real difference.

Also I picked up a feel for just how annoying it is to be left with maintaining code based on abandoned commercial tools (old Delphi code).

3. No idea, but IMO no way banks are leaving it anytime soon so plenty of jobs and momentum for the next few years:-)

Re:Rails has won: The Elephant in the Room by vonklaus on Tuesday May 24, 2016 @04:05AM (#11759716)
I mean, could I teach myself Java and build something interesting in a month. Sounds like a no?
Re:Rails has won: The Elephant in the Room by reitanqild on Tuesday May 24, 2016 @05:37AM (#11760058)
Depends on you as well as your definition of interesting.

I'd say you should start in a good team anyway because real life experience IMO is crucial and getting on a good team is a way of getting it fast and as painless as possible (I.E. without breaking your bank.)

If you just want to make "something" go with whatever is less friction for you (for me it was php), just be prepared that there will be friction in any language as soon as you step outside of the tutorials.

Also keep in mind that if you get a good job you can do quite a lot on your spare time. I used to code a utility program in php while commuting. It never became a commercial success but a rewritten version became a major hit with the church I belong to, was maintained and in use for years and gave me both experience in sw architecture as well as bragging rights in job interviews for several years. (Oh and it totally helped in getting my head ready for my paid work in the morning as well as unwinding after work.)

But I also think Java has come a far way towards beginner-friendly. If you do just accept that in Java land IDEs are a must. Getting used to IDEs was were Java "clicked" for me. Use either Netbeans (all features free, my preferred choice), IntelliJ (lots of features free) or Eclipse (all features you'll ever want is free but takes some getting used to, I would never picked it as my first IDE if it wasn't the preferred choice on my first team and they where happy to help me up to speed.) And dismiss everyone who tells you that one of them is crap or way better: it is just a matter of preference in most cases. (Notable exception: android development where IntelliJ is the officially supported alternative.)

Also learn to use Maven or gradle.

Re:Rails has won: The Elephant in the Room by pjmlp on Tuesday May 24, 2016 @03:47AM (#11759647)
Java might have its issues, but it is one of the languages with best tools available.

For example, most languages eco-systems lack tooling like VisualVM, Mission Control, JITWatch on their canonical implementations.

https://visualvm.java.net/

http://www.oracle.com/technetwork/java/javaseproducts/missio...

https://github.com/AdoptOpenJDK/jitwatch

Also note that although many seem to be focused on Java being Oracle JVM or OpenJDK, there is a pleothora of JDKs both open source and commercial to choose from. With many other features.

For example, most commercial JDKs support AOT compilation to native code.

Or targeting micro-controllers with a few hundred KBs.

Also Java gets bashed by being all about a specific OO model, but VB.NET, C#, Eiffel, Smalltalk follow exactly the same model.

> Is google planning on phasing Java out and using Go?

Until they embrace Go in Android, ChromeOS, their Google Developer SDKs and Google IO sessions, I would say no.

Re:Rails has won: The Elephant in the Room by pritambarhate on Tuesday May 24, 2016 @09:55AM (#11761494)
In 2001, I learned Java on my own from Java 2 Complete Reference without an Internet Connection. It took more than a month. But I have been building lots of intersting stuff with it since then. I think learning Java was one of the best decisions I ever took.

Great thing about Java is that it lets you solve wide variety of problems, right from Mobile, to Desktop to Big Data. For Machine Learning, Image Processing, some of the best libraries and SDKs are available in Java.

It's a good skill to know which I recommend to all new programmers. One important thing I like about Java is that it's low on features and it's possible to master almost all aspects of the language in relatively short period. This is not possibles with languages like C++, C# and Swift. I think simplicity is one of the reason Java and Go are popular.

Re:Rails has won: The Elephant in the Room by nikdaheratik on Tuesday May 24, 2016 @03:04AM (#11759497)
1. My beginners class in Uni taught Java for the 101/102 sequence, so as far as that goes you can use it to implement most beginners projects just fine. It even compiles to something if you want to deal with that experience.

2. I'd say the set of programs can be pretty sparse in that it doesn't tend to play well with its native computer environments (partially be design), so it's more work to generate stuff with a GUI, somewhat more work to handle CLI programs (but not that much more), and it has an interesting set of web tools that have hopefully been updated in the 8+ years since I played with them.

3. I have no idea what Google is doing, but it doesn't matter as Java will live a very long life in Enterprise land if nothing else.

Re:Rails has won: The Elephant in the Room by scient on Tuesday May 24, 2016 @10:17AM (#11761677)
So when you tell people that 3 years down the road the package ecosystem mess is going to be so bad that a package containing a one-line function broke half of the internet, they would go ahead and pick that technology to learn? Oh buddy.... say it aint so.
Re:Rails has won: The Elephant in the Room by rubiquity on Tuesday May 24, 2016 @12:38AM (#11759015)
> I didn't want to use rails anymore because it didn't feel like programming and nodeJS was getting hot.

> Wordpress is also something I swore off when I was learning because it straight up wasn't programming

> Python: Data intensive applications & programming.

I would love to hear your definition of programming.

Re:Rails has won: The Elephant in the Room by vonklaus on Tuesday May 24, 2016 @12:49AM (#11759042)
I still sort of stand by those statements, but again that was based on a beginners completely incorrect view of the world. However, to add a bit more color to those statements which maybe you can infer how I think about programming? I meant something like:

Rails/Ruby was very popular bit it was a framework. You could do a massive amount with very little work. You could build an entire app without any knowledge of what Rails did for you under the hood. NodeJS was a developing technology of which I could gain massive leverage, both learning why asynchronous processes work(I still have trouble with these)/general programming concepts as well as double down on a front end language. The v8 compiler and ryan dahl seemed to provide a good avenue for which to learn lower level(at least than rails) about programming and it seemed to embody the unix philosophy. This seemed like a good place to be and I could grow within a community that was going to be influential. HTML, CSS and SCSS were pretty easy to learn, so by learning a single database technology and javascript I could be a full-stack developer. Not in the sense of how I would market myself, but that I could go out and build a full application if I wanted to and that was poweful and awesome from a learning experience. Given the proliferation of nosql The database, front-end, orm, and back-end were entirely javascript so I didn't pay a high context switching penalty.

wordpress: I dislike wordpress. I don't think it would make any sense to tell someone interested in programming front or back end software to learn "wordpress". Wordpress is a great convenience for setting up a blog, but I presume they have a CMS as well. I don't know much about it, only doing the most basic PHP stuff, but PHP & Wordpress seem to be synonymous with client facing software that requires a CMS/non-technical user to make updates.

python. I mean, python at least from my understanding is a great language for processing data. It is why tensorflow ships as a python library and numpy is used for a lot of data work. The communities using R and python work with data at scale and thus those communities have developed a lot of tooling to accomplish this, not that this stuff needs to be done in python, but machine learning, textual parsing and classification libraries and some NLP stuff all seem to hit python first. However, oython does have a great community writing secure applications. Django and Flask, sqlalchemy ORM as well as zeroDB exist and there are communities using python for apps, data parsing, and microservice deployments.

However, I define prgramming (like most people) as a schedule of television or radio broadcasts.