Develop Reference

How to get started with developing software? [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 5 years ago.
Improve this question
I love programming languages and trying to think of logical solutions to problems. A lot I have been reading on C# and the underworkings of WPF. (Also Python and PyQt) In all my college courses in programming, I learn absolutely nothing (What a variable is, what an array is, NOT what stack and heap are, NOT what CLR/JVM actually does) I feel I have a very good grasp on the basics and a decent grasp on the more substantial topics.
I have been trying to make my own C#/WPF app that is governed by input sensors and output relays for home/shop automation.
I am having trouble to just do certain things. For example. make a user control to be used by a central control that also controls another user control while having a database layer.
I would really like to learn how to structure logically programs and what best design is (for example, I hate having a program filled with if/else statements). The problem is, the books I find only have basic examples that don't really deepen my overall undestanding of how all the pieces need to work together.

How do I start to do things?
You find an open-source (or at least "source available at no cost" -- not quite the same thing, but the difference should only matter if you plan to redistribute your results) project that does "something like" (maybe "vaguely" like;-) what you're trying to do, and is coded in a language that you know or want to learn.
You duplicate the source tree for that project to your machine and put the duplicate under a version control system (svn, hg, git, bazaar -- whatever -- there's a bazillion good, free ones) quite independendently from any VCS it might have been in originally -- the point is being able to save your edits all the time, and revert them easily if and when you make mistakes.
Then you start hacking, looking things up in docs and forums, asking on SO, etc, as needed;-).

There is no "one shoe fits all" solution in programming world. But i do recommend you a book. Hope you like it.
Head First Design Patterns
(O'Reilly)

Honestly... truly... the best answer? Do stuff.
I learned a lot in school. Tons. But as you said, there is a level that schools don't touch. I've got a deeper understanding in my 2 months of programming/data analyst than I got in 2 years getting my AA in programming.
Realize one thing: School is there to teach you a wide range of things - to a minimum level. You won't become an expert from school alone (There are exceptions... as with any rule).
How have I actually started to understand stuff? Programming. Boss gave me a program task and I ran with it.
Personally, I've really liked Apress books. C# 2008 Illustrated (or the soon to be released 2010) has proved invaluable to me. I actually read 95% of the book, double checked examples... and most importantly... put them into practice in my horrible first programming task. I've re-factored at-least 5 times already, and already am seeing things that will make me do it again.
I don't know if you have a job... or a hobby... an open source program you'd like to help with... or a task you'd like to automate. Pick something, learn it inside out, then program it.
Personally, again... I'd say joining an already running open-source project is too much. I tried a few times and it was too much for me. To many things I didn't understand. I plan to try again once I get a little better, but I don't plan on hanging out on NBA courts when I can barely dribble the ball myself... if ya know what I mean.

How do I start to do things?
Find a problem that needs solving and writing a program to do it. It doesn't have to pretty or efficient. The program just needs to work. As you work with the program you'll begin to learn better techniques for accomplishing the task you set for yourself. Google and SO will be your friends and help you learn those better techniques.
As other have said, jumping straight into controlling hardware with C# may not be a simple task for your first program.
Good luck.

Read Martin Fowler's book Refactoring: Improving the Design of Existing Code.
Martin Fowler gives you the building blocks to think about high level design decisions by introducing you to all the low level decisions you may or may not be aware that you are making. He also gives a good introductions to the concept of code smells.
Look at design principles like DRY Do Not Repeat Yourself, KISS Keep It Simple Stupid, and keep it consistent(Pet Peeve).
More design principles here
Design Principles.
Practice these principles as often as possible especially in your course work and outside projects..
Get involved in open source
Hard to do but probably will help you the most especially when combined with the first two.
Personal Experience feel free to skip.
After three years of tough undergrad I realized I could solve some really complex proofs but I didn't know how to write anything but spaghetti code. The longest projects I had were 5 months long which was enough time for it really to catch up with me. I was usally very productive for the first 2-3 months but was crawling along from the code debt by month 4. At this point I had the same realization you did.
Everyone says the best way to learn was to practice but I didn't know how to practice.
I realized that the problem with thinking about design was I couldn't move one design into another. A professor mentioned Martin Fowler's book Refactoring: Improving the Design of Existing Code and that has made all the difference.

Considering moving to .NET from java [closed]

Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 11 years ago.
Improve this question
I am a 4th year computer science student, at university alot of my projects were Java based so I am pretty used to the language, platform and apis. However for a long time, I have been reading alot regarding the .net platform improvements and it's large career opportunities (at least in my country). Now I am considering shifting to the microsoft world. I have several questions:
Would it be severe to shift to .NET platform, as when I start my career I will be lacking the "4 years of java experiance" to a "1 year of c#/.net experience"?
How harsh transaction from java world to .net world would be ?
It feels like the future of Sun (and Java as a result) seems a bit blurry, can it be an issue on the long run ?
EDIT:
Sorry for the edit, I want to add that I am more into the telecommunication field (even tho I have no internship/job experience in this field yet). Plus indeed java is not my only language. I am really used to Python, LISP, Javascript,Php and some other markup languages.
Thanks

It will certainly be a shift, largely in terms of learning the .NET framework when you're used to the Java APIs. A lot of the concepts are the same, but it's a case of learning where to find things. You shouldn't expect to effectively have 4 years of C#/.NET experience - but you'll have 4 years of general software development, which is certainly important in and of itself.
Moving from C# to Java isn't too bad, although generics are very different in C#. You'll find there are a lot of features in recent versions of C# which aren't in Java, and it's worth getting to know them to use the language to its best advantage.
Personally I find C# to be a much more elegant and powerful language than Java - and certainly one which has evolved a lot more rapidly. I use both on a regular basis, but my heart is with C#... and that's having come from a Java background like yourself. I certainly don't regret "moving" from Java to C#, particularly as it isn't so much a "move" as just adding an extra string to your bow. It's not like you're giving up your Java experience to learn .NET.
Having said all this, I don't think you need to worry about Java disappearing any time soon. I think it's reasonable to have concerns over the glacial rate of progress of the core API and language (although 3rd party libraries and indeed languages are moving rather faster) but I'm sure companies will be actively developing new products in Java for quite a few years to come.

If you don't want to become the next cobol developer, get used to changing languages. Java and C# will be around for a couple of years, and so will the underlying platforms. But things will change, so better to get used to it.
You have exactly 0 years of experience. Anything done in school or university doesn't count. The reason is: the requirements are completely different. The school stuff might be way more difficult from a CS point of view, but it has fixed, rather clear requirements, and you don't have to support it for 10 years. So again get used to it: It was hopefully a fun time, but it does not count as experience.
C# was in the beginning almost a copy of Java. Now those guys have a lot of interesting features, that java is lacking. But still it should be easy to learn one, when you know the other.
For a job: It is irrelevant how many jobs there are. You only want one anyway right? So you should check what kind of job you are interested in? Web, Mobile, Embedded, Rich Client, Big Servers.
So what is the real advice? Go, find an employer NOW. Tell them you'd like to work in [whenever you are done] and you'd like them to tell you what you should learn in the mean time.

Personally I don't think Java is going away anytime soon, so IMHO Java is just as good a career move as .NET if not better. In any case, I would say that both options will give you plenty of job opportunities.
The languages are pretty similar so you'll most likely find the transition easy, but obviously you need to familiarize yourself with various libraries and frameworks as well (depending on what type of applications you'll focus on).
A lot of what you know from the Java world is really OO, patterns and so forth and that will apply equally well in the .NET world.

I work for a .NET shop, and I can say that as a hiring manager/interviewer, I don't think of a lack of .NET experience from a recent college grad to be a problem -- very few colleges and universities use Microsoft tools, and as other posters have said, the platforms are conceptually similar. What I expect from a recent college grad is energy, enthusiasm, and a solid grasp of CS fundamentals.
That said, I also don't consider a four year degree to be "4 years of java experience" -- student projects are nice on the resume, and are good to talk about in interviews, but unless you have been doing a lot of coding for hire on the side, one year of "real world" experience (where your job is to write software) teaches more about work ethic and delivery than 4 or more years in college.

It depends a bit on what type of personality you have and on what level you understand java. Allot of concepts are the same between the two platforms only with different implementation details, syntax, names etc... on one level I find .net easier since you often deal with all Microsoft products which tend to align with each other pretty well as where in the java space I've had to spent time figuring out how to get product A by supplier X to work together with product B by supplier Y. So if you have more then what I would call a "literal" understanding of java the move should be very easy.
I think you should be able to find jobs in both platforms just as easily imho
If perhaps this move is also being fuelled by eagerness to learn new things then I would advice you to go for something different like a dynamic or functional language.
There are a couple of books which focus on moving from java to c# (I haven't read them so I can not advice on one) but those should really get you up to speed in a matter of hours/days.

I wouldn't bother with learning C# for career purposes. And I say this as a happily employed C# programmer. :-) Oh, I suppose an explanation is in order...
It's much more valuable to know diverse languages. When I'm hiring for a position that's going to be writing C#, I'd much rather hire somebody with lots of Java experience and also some Scheme and Erlang (and no C# at all), than somebody with just Java and C#.
C# really seems to combine two things: the static typing and syntax from the Java world, and a handful of additional features and syntax from the functional world. If you know Java, Lisp, Python, and Javascript, then you can learn C# -- or pretty much anything else -- in almost no time at all. (You won't land an architect position, but you won't be getting that right out of school, anyway.)
The only trick is finding an employer enlightened enough to realize this, and not simply one that feeds resumes through a pattern matcher. But simply putting "Lisp" on your resume is often enough to turn the right heads.

java and .net are both fine i guess.
as long as you want to keep portability to linux, mac os or any other system, you might keep on java. if you want .net-applications for other platforms, you would need to rely on Mono, which is for sure a great project but in my opinion not yet stable enough for production.
if you want to stay on microsoft-world, i consider C# as one of the best development tool i've ever seen. indeed the visual studio, which provides many important coding features really makes the work easy.
they're both approximately the same from the learning aspect. java lacks operator overloading which is an important point in my eyes. .net might lack other features java is better at... and generally i guess in most of the applications, .net is faster than java.
hope this helped a bit
regards

If you write in C#, you will find a lot of similarities to the Java language, and the CLR shares many concepts with the Java runtime, too.
You could as well write both. For exercise i created a chat-program in C# and the server for it in Java and had a lot of fun doing that. You could concentrate on .NET and occasionally use Java. Nothing wrong with that, anyway, more important than being used to a language is the concepts you learn over the years, such as design patterns, best practices and algorithms. Those will help you to understand any new framework or library much quicker as you recognize the patterns and purpose behind it.

I agree with most of the advice others have already given. Both platforms are here to stay and both should provide decent career opportunities.
Just a word of caution though. You write "at least in my country" regarding .Net opportunities. I think you also need to consider "at least in my country" when deciding how large the career impact will be in the short term.
I have worked extensively in both the USA and Germany. My experience has been that, in the USA people are hired more for demonstrated ability than for resume points, but in Germany the situation was reversed. You will have to weigh the short-term impact in your country, unless you want to post which country that is and hopefully someone on SO has experience with that country.

Your Java experience and concepts will come in handy when transitioning to C# and they're not too different (to a certain extent, of course). I won't go into much detail about that aspect of your question since others have made good comments.
What I wanted to add is that although employers may not consider school experience to be equivalent to real world experience, don't let that stop you from referencing them on your resume. Be prepared to talk about them during the interview, as well as any side projects or interests you have. These are good for the "tell me about your favorite project?" and "what did you like most about that project?" type of questions. Be enthusiastic to learn and show a willingness to improve your skills. You can also show that you find certain aspects of C# appealing (perhaps contrast to Java) and that you're confident you'll be able to pick things up quickly and apply yourself.

Job experience is vastly different from the experience you get as a student.
Personally I think you should learn as much as you can while studying - this DEFINTIIVELY includes exotic languages - and only knowing one language, makes you a one-trick pony.
So, get your teachers to teach you exotic stuff. Like functional languages using pattern matching, logical languages for a vastly different approach to evaluating lots of possibilities, assembly language for doing the actual bit stuff, and all these other mind blowing things that people do for FUN and LEARNING :)

Django or Ruby on Rails [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
I'm a C#/.NET developer looking to mess around with something completely different - something LAM(*) stackish for building web apps quickly.
I'm thinking either Django or Rails. I kind of like the Python language better and it seems to be more full-featured than Ruby for statistical, scientific and networking (let me know if you think this is wrong). However, the RoR community seems to be much bigger - which might make it a safer bet. Do you think this is important? Anybody have experience with both RoR and Django?
Update: Specifically I'm looking to be able to build a community finance oriented site quickly. There are definitely a lot of Ruby Gems to speed up the process, but I noticed Pinax on the Django side which looks promising.
I know that Python is already fairly popular for financial/mathematical programming. Anyone else have an opinion?
Update 2: Noticed some comments about ASP.NET MVC. I have in fact done pretty significant work with ASP.NET MVC - a LOB app for Medical Equipment Servicing, and I loved it. It is IMHO a much better and more intuitive way to write web apps compared with ASP.NET web forms. But, I'm really looking for a non-.NET development environment to develop some chops in.

I kind of like the Python language
better and it seems to be more
full-featured than Ruby for
statistical, scientific and networking
(let me know if you think this is
wrong).
If you think you'll need libraries like NumPy or SciPy, probably best to stick with python/django. I've struggled to find ruby equivalents.
However, the RoR community seems to be
much bigger
Bigger is of course not going to be better. I've had great experiences with the django community in terms of support, and the pluggable app ecosystem is maturing quite quickly.
I would also add that I find the django documentation to be very easy to follow.

If you want a quick stack check out Sinatra.
Django will make more sense faster. It has awesome routing, simple file structure, nice clean syntax, not a lot of WTF moments, and understandable ORM functionality.
Rails you will have to dive into. You will battle activerecord. You will try to wrap your head around the many facets of a rails application.
I like them both. Try them both out. They don't take that long to get a feel for. You should ultimately base your decision on a per-app basis and factor in which language has better modules/extensions/gems for what you need.

Since your stated preference is Python, I wouldn't hesitate to recommend Django. The documentation available for Django is top-notch, and the community is very helpful. Browse the django-users archives or hang out in IRC for a while and you'll get a good sense of the support available.

I like Django better because it's less magic, with Rails i feel they pushed the convention over configuration principle too far, so i prefer the more explicit nature of Django.

funny... I'm in the same boat, though perhaps I've been looking around for a bit longer.
I came in from MS-land (C# too) and I went with Django, but only after I tinkered around with Google's AppEngine, which re-introduced me to Python. Django is well organized, internally consistent (as far as I could tell), and well supported with a vibrant community. Python is well organized, internally consistent (as far as I can tell), and extremely well supported with an amazing community.
Anyhow, there's more to rapid Python web development than Django. You might want to search for: python wsgi framework. Pylons and TurboGears offer enormous flexibility and focus on "best of breed" components, web2py appears to have a very cool DAL/ORM that I'm itching to dig into, and web.py and werkzeug (amongst others) are minimalistic and don't try to dictate how you should build your app. And if you're sticking with Django, definitely check out the Pinax project - it might help kickstart your development efforts.

I just finished building a small social networking site with Django for a client. I was new to Python and new to Django, but it was one of the most enjoyable development experiences I've had in a while (even with the headaches of being new to a language and framework). And it only took two weeks.
If you're building this for yourself then I would try both Rails and Django (or one of the other frameworks that c.batt mentioned) to see what feels the most comfortable. I ended up choosing Django/Python because I liked the language. It felt clean and efficient, and you can't argue with the speed at which you can get a bare-bones site up and running.
As mentioned previously, Pinax is great. If anything you get a solid base to build on top of. I found that I had to extend parts of it to get exactly what I wanted, and a couple of sections I decided to code from scratch because the changes would get "hacky" to match what I needed. Other parts, however, are perfect. The private messaging was used out of the box, just with some new templates to match the site.

i'm another C#/.net guy with the same itch to tinker. Django is appealing. Though the development is easy and fun, i've heard some stories about deployment and configuration headaches with both Rails and Django (mostly rails). It's not like just throwing an asp.net app at IIS or PHP at Apache. For those things you can set it and forget it. I'd be curious to get c.batt's comments on his/her experince with deploying django.
Lately I've been looking at Grails. It uses Groovy which is ruby like. It's very easy to develop in and you can build apps as fast as Rails but you get access to the Java platform via the JVM which could come in handy. There's a ton of open source java components you could drop into your Grails app, plenty of the types of components you'd be looking for. It's really not like building a java app. So far it does feel like Rails. Good luck.

You are at a point where you cant make a bad choice!
My personal preference is django, but I know for a fact rails is good too.
I think the best answer is what the BFDLs of django themselves say:
Try both, use whichever you like.

Rails is like Bible. Unless you are a Saint, you do not need to understand it and you probably will never understand; all you need to do is to believe in it.

Since you're a seasoned .NET programmer, you really should look at ASP.NET and MVC before turning to Rails and Django, at least if your goal is to get up-and-running quickly. It's always easier to leverage something that you already know in-depth.
I was in a similar situation recently -- I had a project that I wanted to push out on Windows and Unix, and so rather than leveraging what I know well (Ruby/Rails), I jumped ship and gave Groovy on Grails a go. It was a really rewarding experience, and I'm really glad I did a few small projects in Grails, but in the end I had a tiny fraction of the productivity I did in the environment where I had a few years' experience, even after months of hacking away on Grails.
So, pick up Django, Rails, Lift, or whatever you feel like doing -- it will help expand your mind, and give you different ways to look at solving problems. But if you want to get going quickly, use what you know (.NET) on a platform designed for you (MVC).

You express a preference for Python so I'd go with Django - it's a fine framework. My personal preference is for the alternative, but I'm not you and both options are good.
This is mostly a duplicate of https://stackoverflow.com/questions/91846/rails-or-django-or-something-else, btw

Like others suggest, trying both is one good way, but really, to me, I would base it more on which language you prefer to use. Sure, if it's a web app, you'll do a lot with the framework, but "every" line of code you write will be in that language. Since both frameworks are strong, it's really more a matter of what language you'll be happiest writing code in, day in and day out.
I personally use Rails and love it, but have many friends using Django. When we discuss things, time and again, it boils down more to language than framework. The languages are fairly different, and you'll typically find one you prefer a lot more than the other, at least in my opinion. I had originally thought Python sounded great, learned it, wrote a few things with it, etc., but it just never "clicked" for me. When I first saw Ruby many years ago it seemed a bit odd, but when I revisited it via Rails, it just felt so natural. When I write code in Ruby I feel like I'm pretty much writing English - if I can think of how I think the code should be written, I can pretty much type that and it works. It just fits my brain better.
Community wise, you'll get a lot of different opinions. I don't think being "too big" means it's bad, nor do I think the Ruby community is by any means "too big", especially compared to say the Java community. Ruby and Rails have certainly gone mainstream to some degree, and for those of us who've worked with it for several years, it seems mainstream, but in the grand scheme of things it's actually still fairly small.
Anyway, good luck, and enjoy looking into them and finding what will delight you to work with.

I'm curious. If you are a C# programmer, why have you ruled out Microsoft's MVC?
You have two different needs expressed in your question:
1) Want to try something new
2) Want to build community finance site quickly.
I understand the desire to look at RoR or Django for need 1, but for need 2 I'd expect you would be more productive more quickly with MVC.
Please note I am not saying that EVERYONE would be more productive more quickly nor am I saying that Microsoft's MVP is inherently more productive, just that someone who is currently a C# programmer would be able to come up to speed more quickly on something that uses C#...

Although I've never tried it personally beyond playing around, you might also want to check out ASP.NET with IronPython. Here's a blog post that details it.

I agree with the people who say: try both and use what you like best.
I think Rails will be an interesting experience because of the convention over configuration and forced MVC architecture. Also, I am not sure what people's experience with Django is on StackOverflow, but all of my Rails questions get answered very quickly.
Also keep in mind that you can call a python service to do some statistical calculations if necessary.
Here is an article written by a guy who went from .NET/C# to Rails to Django/Python, and then back to Rails. Might be worth a quick read.
http://blog.carlmercier.com/2007/01/30/why-i-moved-from-ruby-on-rails-to-pythondjango-and-back/

Your question must be : Python or Ruby ;)

Software that every programmer makes [closed]

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
Locked. This question and its answers are locked because the question is off-topic but has historical significance. It is not currently accepting new answers or interactions.
I'm looking for a personal project. One of the sort that "every programmer"* makes. The reasoning is that most developers have a tendency to work on personal projects that have particular traits interesting to all other developers. Software like:
Bug trackers (made one)
CMS (made one)
Blog Engines (Haven't made one; don't want to)
UI 'libraries' and graphics 'engines' (often for the game development crowd)
Password management programs
I'm looking to make something
fun
interesting
not particularly challenging (my work is difficult - or obscure enough)
doesn't have to be original by any means
preferably an application (ie. not a website)
not a hardcore learning experience, this is entirely for fun away from work
That being said, I find that the terms 'interesting' and 'challenging' don't necessarily carry the same meaning for developers as they do for other people (and often-times are interchangeable - I'm not looking for something easy or trivial). Ditto with the 'learning' part, I like and want to solve problems, but I'm not going to be getting into design methodologies or learning entirely new APIs and Languages to do this.
I ask because I think that certain projects, or types of projects, appeal to developers for particular reasons. I don't claim to know what those reasons are, nor do I care, but I do understand that they exist and that being a developer, there is a strong likelihood that the projects that are suggested will appeal to me.
If it helps, my background or preferred medium, without reciting my resume, is c# development for windows/tablet pcs; xna; db design/programming;
*by "every", I mean, "lots of".

You have XNA experience, and DB experience.
Develop a visualization tool for databases which is visually attractive and makes mining data easier.
This can start as a simple browsing tool and eventually grow to a full-fledged DB management application.

If you haven't written one, knocking together an implementation of Conway's Game of Life is always an interesting challenge. It's also the sort of project that once you get it done leads to seemingly endless tinkering ("I'll just add one more feature...")

Graphics are always fun to play around with, try writing a program to draw a Sierpiński triangle. It's a fairly simple fractal to generate. I once made a screensaver that drew them in random places and colors on the screen.
If you are looking for something more practical, try writing an email or RSS client. Learning to interact with SMTP/POP3 at low level is pretty interesting. Same thing with RSS, it will expose you to network, GUI, and XML design at a minimum.

Maybe write a compiler? I think every artist/craftsman should, at some point, learn how to build the tools of their art or craft.

Write a character generator for the pen-and-paper rpg of your choice.
It should be able to be used in 2 modes: Build-A-Character (which will let you build a character from scratch) and Generate-An-NPC (which, given some guidelines, will generate a decent NPC).
This is actually a project I've been working on for a while, and I've found it fun. YMMV, especially if you don't game regularly.
Edit - figured I'd give some reasons I find it fun.
It's something I and my friends will actually use
It helps me understand the game better, knowing what decisions make a good character
There's some (limited) AI in the random character generation(mainly a simple decision tree with some random jumps)
It's overall fairly simple, a reasonable project for a one-man development team.

I'm looking for a personal project. One of the sort that "every programmer"* makes.
How about writing your own scripting language interpreter?
That's a very common thing to do, which is usually not only a highly educational experience, but also fun and interesting.
Once you are finished with something working, you'll start thinking completely differently about programming, it will change your way of thinking!

A Sudoku game (or Connect Four, Clue, etc.).
Something that relies mostly on logic, and not artificial intelligence (like chess).

How about a ray tracer? It can be very simple if you do everything by brute force.
They are very fun to write and can lead to many interesting problems. Writing a ray tracer meets all your criteria =)

Each and every developer is different. About three years ago I asked exactly your question on some other site. It brought nothing because we all have different interests.
Maybe to make a website with focus on what you like? Like a small community within a university or your other fellows?

In order to learn Ruby, I just recently made a mini DSL/XML generator for a custom XML based language I use in a program of mine. I simply turn some real text into its XML representation. It was just a fun little side project that didn't provide much benefit in the long run but helped me to learn the language.
So... I suggest trying to create a DSL. The experience could actually prove invaluable as DSL's are pretty damn useful.

Write a 3D environment populated by artificially intelligent virtual beings, implemented on a public server, that we can all access using a WPF-based client.

Personally 'interesting' strongly overlaps 'difficult', so I'm kind of at a loss ...
... but perhaps you should take a non-programming interest you may have and try to solve a simple problem in that space.
A friend of mine once built a model railway database so people could track the trains they had. He was trying to sell it, that didn't go so well, but it was simple and fun.

Why not rolling your own
Web MVC framework or
Web templating system or
Persistence framework (+1 if you make an OR-mapper) or
Interceptor-based aspect-oriented framework
But oh, those were the sins of early 2000's.
Something cloudish then? :)

I had fun programming an audio player (like winamp / WMP) to learn .NET before I started my current job. There are free libraries to read MP3 files (and other audio formats), so that gets the complicated part out of the way ;)
An audio player is not too complicated and is good for learning threads, UI, and working with the file system.

I've always had fun trying to play with compilers/interpreters. I started out making pathetic attempts at my own Javascript interactive shell (I had no prior experience or knowledge of compilers/interpreters and just tried to wing it). And now I'm about to start working on writing my own lexer/parser from scratch.
Personally, I find compilers and interpreters interesting. You can use them, or their components as a start of other project ideas too if you wanted. Also, you can play with them in any language of your choice, so it might not be a bad side project if you want to learn or use a new language and aren't sure what to focus on to do it.

I've always enjoyed writing a little physics simulator or the canonical mandlebrot / julia set generator (parallel extensions anyone :)

Some kind of GPS tracker? I never made one but I believe that of lot of devs do.

Write an touch screen game for a Windows mobile device that competes with the iPhone. Maybe like,'Squash the Bug' or something. Chicks dig that kind of stuff.

Search for a good flash puzzle game and implement a solver for it :)

Personally, it based on your own preferences. For me, it was a grade book program. Since I always had teacher friends, they always wanted me to update it. So I first wrote it in Pascal, then wanted to learn C, so I wrote a C version, then a C++ version and so on.
My daughter was playing basketball, so I wanted to keep score and I had a PocketPC at the time, so I wrote a little app for it.
My $.02

How long would it take a non-programmer to learn C#, the .NET Framework, and SQL? [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
I am not that good at programming. I finished my masters degree in electronics. I want to learn C#, the .NET Framework, and SQL. How much time do you think it would take (if I have 5 hours a day to devote to it)? Also, what order do I learn them in? I have Visual Web Developer 2008, is that enough to begin?
Copied from a new question by the same user:
I should have specified that I wish to make a career out .NET Development. I'd like to give myself 2 months to get the 'basics' down. As a developer, what is expected out of you in IT companies? What skillset do they require and what are the responsibilities as a junior or mid-level developer? I don't have any industry experience, will I be able to get a feel of the job duties while at home?

That's like asking how long it would take to learn French:
1 day to learn what it is
1 week to learn it to an infant/elementary level
1 year to be considered a beginner by professionals
Several years to be considered an experienced professional
Plus there's "deep" knowledge of those subjects which a mere mortal such as you or I will never learn
Then again, plenty of people (most normal people, non-programmers) never learn those subjects, so if you're like "most" people then the answer would be "it would take forever" or "it will never happen".

Dive right it, make it a hobby, and have fun :)
Coming from an electronics background myself I can tell you that you should pick it up pretty quickly. And having an electronics background will give you a deeper understanding of the underlying hardware.
IMHO the root of information technology is electronics.
For example..
Think of objects as components.
The .NET framework is essentially drawers full of standard components.
For example you know what a 7400 (NAND gate) is capable of doing. You have a data sheet showing the pin outs and sample configurations. You don't typically care about the circuitry inside. Software objects are the same way. We have inputs and we have methods that do something to the inputs to produce predictable outputs. As developers we typically don't care how the object was written... just that it does what it says it will do.
You also know that you can build additional logic circuits by using two or more NAND gates. This of this as instantiation.
You also know that you can take a NAND gate and place it inside a circuit where you can modify the input signals coming in so the outputs have different behaviors. This is a crude example but you can think of this as inheritance.
I have also learned it helps to have a project to work on. It could be a hobbyist project or a work project. Start small, get something very basic working, and work up from there.
To answer your specific question on "what should I learn first".
1) Take your project you have in mind and break it into steps. For example... get a number from the user, add one to the number, display the result. Think of this as your design.
2) Learn basic C#. Write a simple console application that does something. Learn what an if statement is (this is all boolean logic so it should be somewhat familiar), learn about loops, learn about mathematical operations, learn about functions (subroutines). Play with simple file i/o (reading and writing text files). The basic C# can be thought of as your wiring and discrete components (resistors, caps, transistors, etc) to your chips (object).
3) Learn how to instantiate and use objects from the framework. You have already been doing this but now it's time to delve in further. For example... play with System.Console some more... try making the speaker beep. Also start looking for objects that you may want to use for database work.
4) Learn basic SQL. Lots of help and examples online. Pick a database you want to work with. I personally think MS Access is a great beginners database. I would not use it for multi-user or cross platform desktop applications... but it is a great single user database for Windows users... and it is a great way to learn the basics of SQL. There are other simple free databases available (Open Office has one for example) if you don't want to shell out $ for Access.
5) Expand your app to do something with a database.

Just do it! Don't sweat the details.

Well, it will take you forever. There is so much to learn about programming that 10 years are not enough.
http://norvig.com/21-days.html
Don't get me wrong, you will learn the basics quickly enough, but to become good at it will take much longer.
You should focus on an area and try to make some examples, if you choose web development, start with an hello world web page, then add some code to it. Learn about postbacks, viewstate and Sessions. Try to master ifs, cycles and functions, you really have a lot to cover, it's not easy to say "this is the best way to learn".
I guess in the end you will learn on a need to do basis.

The best thing about C# is that it is very catchy. Easy to pick up, and you'll also have fun doing it.
But .Net framework is a very big library full of wonderful things to discover, and yet again due to the nature of .Net you'll also have fun learning it. It's a coherent, object oriented, well documented library, and C# makes it so simple to work with it that you can simply discover your way through it while coding.
The vast majority of articles, books or resources about .Net and C# simply concentrate on explaining functionality and the framework and far less about how to avoid quirks, workarounds or exceptional cases, like it happens with other languages I don't want to name (*cough C++*) so in the end the experience of learning C# and .Net is very enjoyable from start to finish, and the things you can accomplish using .Net also makes it very rewarding.
You picked a good language to start with, in my opinion, and finally to answer your question, it will take you about:
2 to 3 months to learn the basics
1 to 2 years to become a versed developer
5 years or more to become a expert or, depending on your dedication, a "guru".
But then again, beating the numbers and breaking the limits lies inside human nature. Can you do it faster than this? ;-)

How long is a piece of string? I think this is subjective. I know programmers that have learned an extraordinary amount in a very short time based on the experience that they've exposed themselves to.
Basically, get your hands dirty and you're bound to learn more.

It really depends on what you mean by "learn". You could probably spend a week and get a couple of pages up on the web that had some minimal level of interactivity to save information entered by the user in some database, and then have some other pages for querying and displaying the information. You could then spend the next 10 years of your life learning all the intricacies of the .Net framework, SQL, and mastering using the IDE.

The period of which you are capable of learning really depends on your ability to grasp the logic behind programming while where to learn from depends on your learning style.
If you are a learn-by-a-book type of guy, just jump on Amazon.com and perform a quick search, pick up the book with the best reviews or wait for someone here to recommend a book (I'm not a programming by book guy)
If you prefer screencasts (video feeds demonstrating what to do) or tutorials, then go straight to the source: http://www.asp.net/learn/. There are tons of videos and tutorials explaining everything you need to get started.
Visual Web Developer 2008 Express should be all you need to get started. Basically, the express editions are Visual Studio chopped down to a precise set of functionality to accomplish one thing. They don't have some of the bells and whistles needed for large scale development, but everything you should need.

Can't tell how long it would take, it really depends on your existing knowlege. I managed to learn the C#/.NET 2.0 core basics in about 2 months.
My suggestion to you: Try to learn towards exams, they make sure your learning covers all important parts and also guide you through this new technology. See Microsoft Learning.

If you have any programming experience, you can probably learn the C# syntax in a few hours, and be comfortable with it within a week or so. However, you will not be writing complex structures unless you write a lot of code with it. It's really the same as learning any language: you can learn all the words and grammer fairly quickly, but it takes a while to be fluent.
EDIT
A book you may want to pick up for learning C# is C# in a Nutshell (3.0) which I found to be very useful, and has been recommended by several people here.

If you want to learn, REALLY want to learn, then time is not of consequence. Just move forward everyday. Let your passion for this stuff drive you forward. And one day you'll see that you are good at C#/.NET.

You'll pick up c# fairly quickly (the language syntax is not that complicated). It will take you a long time to really learn the .NET framework, but you'll pick up the heavily used parts of the framework fairly quickly, and you should start seeing patterns in the framework.
My advice to you: don't just learn from a book or website. They will teach you the language and framework, but they will not teach you how to program anything useful.
Writing little code snippets will teach you how to do a very specific tasks, but they do not teach you how to write applications. My suggestion is that you think of an app that might be fun to work on (and doable... e.g. don't think that you're going to write an operating system or crysis or something in a month or two). Personally, when I was learning, wrote my own full featured IRC app, complete with rich text, personal messaging, etc.

The answer in my view is related to whether you have a tangible problem to solve or if you just want to learn for example to be prepared for a possible new job. If you have a problem then you are in better shape. You can start by looking around and seeing how other people went about solving that problem. Languages in general you should be able to pick up fairly quickly (after all you hold an MS in EE, no small feat IMO).
What you need to be on the lookout for is good programming practices. You'll probably see yourself asking "why is this method so small", "why is this method empty and what the heck is this abstract word doing here". That will give you perspective beyond syntax towards good writing.

When I switched careers out of Finance, I took 9 months off to study C++ full-time out of a book by Ivor Horton. I had a lot of support from my best friend, who is a guru, and I had been programming as a hobby since high school (I was 36 at the time).
It's not just the syntax that's an issue. The idea of things like pointers, passing by reference, multi-tiered architectures, struct's vs classes, etc., these all take time to understand and learn to use. And you're adding to that the .Net framework, which is huge and constantly evolving, and SQL, which is a totally different skill set than C#. You also haven't mentioned various subsets of the framework that are becoming more widely used, like WPF, WCF, WF, etc.
You're an academic so you can definitely do it, but it's going to take serious effort for a long time, and you definitely will need some projects to work on and learn from. Good luck to you.

According to Malcolm Gladwell, it will take you 10,000 hours to get really good. So get cracking.

Simple answer: a lot longer than two months. Learning to program competently will take longer than that, no matter what. It took me years to learn to be a competent object-oriented programmer, and I'm good at this stuff.
More detailed answers: it doesn't really matter whether you learn C# or SQL first, as they're very different. I'd probably suggest SQL, as it's easier to learn and more independently useful.
You will have a hard time getting used to the on-the-job realities at home, much as if you were studying plumbing or quantitative finance.
You're going to have a hard time putting all the information together without one or more projects you try to do. You're going to need to have other people to tell you when you're being stupid, when you're being overclever and will pay for it later, and when you're actually getting it.
Try to find an open source project you find vaguely interesting. Study their code. Figure out why they do what they do. Look at the bug list, and try to find something as trivial as possible to fix. Work from there. Learning to contribute is going to teach you things that are useful in the work world, and it will give you something to point at. It will be far easier to get your first job if you have some experience to point to.

All "relativity" aside, not fast. Based on the fact that you said you never programmed before...to become a basic programmer, a few years.
And to become a good to outstanding (using design patterns and industry recognized standards that relate to common standards as defined by ISO/IEC 9126 Standard such as testability, maintainability, etc.) programmer, it takes years of experience and coding often.. you do not become "Sr." or an "Architect" overnight and the same thing is true for a mid-level developer who doesn't code slop.
It's always a process where you improve. So learning is relative. But to learn the basics, seems simple until you start to design classes and interfaces. And even Leads stumble on the basics..doing things wrong. Everyone does. There is so much to be aware of.
If you're just going to be adding features (using classes your Lead or Architect has stubbed out for the team) and not really adding new classes, etc. it's easier....but you should take care in coding using standards and you still have to know complex areas of OOP. But that's not really OOP. When you start to creating classes, interfaces and knowing about inheritance, heap, references, etc. yada yada...and REALLY understanding it takes time no matter how smart you are or think you may be.
So, for a new programmer. Not easy. Be prepared to code a lot. And if you are not, find a job where you are. It's all about coding as much possible so you can get better.
Read these books FIRST. Do not dive into any others out there because they are not geared toward teaching you the language in a way you can get up to speed fast:
http://www.amazon.com/Head-First-Object-Oriented-Analysis-Design/dp/0596008678/ref=pd_bbs_sr_4?ie=UTF8&s=books&qid=1231280335&sr=8-4
http://www.amazon.com/Head-First-SQL-Brain-Learners/dp/0596526849/ref=pd_bbs_sr_7?ie=UTF8&s=books&qid=1231280335&sr=8-7
http://www.amazon.com/Head-First-C-Brain-Friendly-Guides/dp/0596514824/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1231280393&sr=8-1
http://www.amazon.com/First-Design-Patterns-Elisabeth-Freeman/dp/0596007124/ref=pd_bbs_sr_3?ie=UTF8&s=books&qid=1231280393&sr=8-3
they will get you the fasted jump start into understanding, better than any books out there.
Also for these lame type of responses, ignore them:
"Then again, plenty of people (most normal people, non-programmers) never learn those subjects, so if you're like "most" people then the answer would be "it would take forever" or "it will never happen"."
Those come from developers (typically leads) who have some Ego trip that DON'T want you to learn. Everyone learns differently and at different paces and eventually you will become "fast". I get very tired of hearing Sr. developers say statements like this when their sh** also stinks many times no matter how good they are. Instead they should be helping the team to succeed and learn as long as their team is working hard to keep abreast and doing what they can on their own as well (not leachers).
Make sure you try to get a Jr. Level Developer position first...

2-3 months to learn the basics...no way. Unless you're gonna code spaghetti sure. Same goes for the 1-2 years. Spaghetti it is then..

Build on what you already know and have a look at lot of job adverts. E.g I have seen jobs asking for WinForms/WPF AND electronics for the writing of a UI to control a custom bit of hardware.
You may find the “robotics” .net toolkit interesting.
.Net is now too big for anyone to learn both WEB and Desktop so you have to decide the way you are going to go. Web has lots more jobs, but there are very few people with good desktop stills.