Buy a mac.

It has cheatsheets, searches SO if enabled and more language support. It's almost worth spending $5,000 for a mac that's equivilent to a $500 PC to get those features.

Real world mostly meaning someone is paying you to develop software.

Where bootcamps could fit in.

Nope. I would re-phrase as follows.

There are a myriad of career paths and technologies that are applicable to software development. At the same time many are challenging and at the same time very interesting. Without a lot of experience developing software, it is hard to know what choices best suit one's interest and abilities. When you work for an entity, the focus tends to be on completing assignments in technological areas that the entity is focusing on. Getting exposure to these has its benefits, even if you decide that you are not all that interested in them at the end of the day. In short, being paid for developing software tends to focus on the practice of software development.

University course work tends to be involved with theory and principles more than the practice of software development. Learning theory is great. However, there are a lot choices one is presented with in course work and it is really difficult to know what areas of software development one wishes to practice in by just being exposed to theory alone. Knowing what areas of software development one wants to practice in can focus their coursework in learning theoretical concepts in those areas.

Not sure about that given the cost of a university education these days. If you end up taking courses in areas that it ends up that you really weren't that interested in when you had better alternatives, maybe even much better, then I think it is fair to say that you've wasted time and money.

I feel like you do, I don't feel there is anyway I could even pretend I'm good enough to have a job in the industry given my all self taught skill level.

but then I also read lots of stuff about how people with degrees cant even pass a simple fuzz buzz test, I dunno how much of that is hyperbole or not, but I have a friend whos in his last semester of a CS degree that I feel I prolly know more than, so I really dont know.

If I actually were good enough to get a job somehwere coding, I would prolly be equal parts shocked and appalled by how easy it is to get such a job and by how bad lots of professionals must be.

So I've been asked to rewrite some C# code into Java for an open-source project I've decided to volunteer on since my summer internship fell through. This method header is giving me some trouble:

The string x = null, I'm not sure what that means as I haven't written much C# and not sure how to do that in Java?

I'd assume it's specifying a default value of null if the function is called without that particular optional argument.

Here's a discussion on simulating default values for optional arguments in Java:

http://stackoverflow.com/questions/9...nal-parameters

That is what I was assuming, just not sure.

Am I going to have to use overloading to do the same thing in Java?

edit: Ah, I see now that I could do varargs. Never done that before. I wonder which way is better, varargs or overloading?

in Java I would prefer overloading over varargs, personally, and it's the most straightforward way to port that code

If you've been playing poker for a long time you know all you have to do is be slightly better than the average player at the table. How good do you think this table is? Hint: its not good. There's too much demand for people in this industry and not enough people who are not only smart and full of knowledge about these jobs but "passionate" about doing a good job and writing code and making products the right way as opposed to just half assing it. Everywhere I went while contracting was full of people copy pasting code, throwing plugins at things, not knowing or caring about industry standard best practices and technology and overall just not that good at doing this stuff. Don't overestimate these people and these jobs and these companies.

Yeah... I feel the same way you do. It is pretty obvious from the meetups I've been to that there are a ton of posers. To be fair, there are quite a few brilliant minds out there as well. It comes down to taking the dive and just going for it at some point. Where the point lies is a personal decision, and very difficult to know in a vacuum.

My biggest problem is that I despise building websites, and finding something beyond that is proving quite difficult. I *could* go for websites just for the experience, but why would I enter in on a position that I know I'd hate and why would I want to mark myself as a "web-dev" and make exiting out more difficult? Going to a classic CS school just for the opportunity to explore and say "I can do X" is worth quite a bit of personal gold, IMO.

I've got an idea for a web app, and would like to ask about how difficult it would be to make / the best technologies one would use to make it.

I have a database that I think is quite valuable. Basically, it consists of a set of companies and a few informational fields for each company. The data relates to sales, and people in the industry could gain valuable information from performing a search on the company before trying to make a sale.

Essentially, all I want the web app to be is a secure database on the back end and a pay-per-search charge on the front end.

I understand basic programming concepts and have written programs for myself in Python, Java, and JavaScript.

Is this the type of project someone with basic programming knowledge but no web app experience could pull off in a few weeks? If so what technologies would be best? If not, is it something someone with experience could do for a reasonable price? How easy will it be to keep the data secure on the back end?

re: entry level programming

Is that fizzbuzz test thing really true? I find that astounding. For me, i've always found simple logic puzzles like that really easy, but when it comes to broader concepts like design patterns and writing more complicated programs, that's where I tend to get lost.

I can write the **** out of a method like the fizzbuzz test but feel like i'd be a waste of space as a programmer in the field.

Has anyone here ever released an open source package to NPM or been a heavy contributor? I'm ready to release my node framework but definitely have a lot of questions about where to start. Particularly testing and creating boilerplate.

Everyone starts just figuring out how to make **** work. Then after a few years you start to figure out how to make it work and be maintainable. Then if all goes well a few years after that you start to figure out how to make **** that is relatively easy for other new developers to use to make their **** work and be relatively maintainable.

I'm not sure what happens after that, because that's where I'm at.

anyone know a good resource to learn tcl and tcl/tk? currently working through

https://www.tcl.tk/man/tcl8.5/tutorial/tcltutorial.html for tcl
http://bin-co.com/tcl/tutorial/ for tk

suzzer, I posted a few node packages to npm. I mainly used npm init's scaffold and kept the entry point to my lib in the root index.js file.

I setup a travis test too. I recommend looking at the express repo for the other boilerplate like a readme file, license, etc..

Can anybody shed some light on this program. I understand recursion and splicing, I'm just having a hard time wrapping my head around this:

Try drawing a picture of what's going on in each line of code or loop through the function, start with two easy sets (a, b) (b, c).

I think it looks like this:
0: union((a, b) (b, c))
_1: returns (a + union((b),(b, c))
__2: union((b),(b,c)) returns union((),(b,c))
___3: union((),(b,c)) returns (b,c)
walking back out
__2: becomes union((b),(b,c)) returns (b,c)
_1: becomes (a + b,c)
0: union((a,b) (b,c)) returns (a,b,c)

(I think, I don't speak much python)

If the first Set is blank, it just returns the 2nd set (nothing to join together)

If the current front element of Set 1 is in Set 2, we dont need to combine it, bc it already exists in Set 2, so it calls Union, this time starting off Set 1 at the next element.

In other words, if Set 1 was (1,2,3), its calling union with Set 1 set to (2,3). The 1 is dropped.

If the first element of Set 1 is not in Set 2, it puts it on the front of the result, and does the same thing as above, so now we get (1) + Union(Set 1 = (2,3), Set 2)


I have a feeling that didnt make much sense, but yeah, trace it out with Set1 = (1,2,3) and Set2 = (3,4,5) and see what you get

Oh thanks guys. I get it now. What I was misunderstanding was the logic of what a union of two sets is. I had it backwards thinking of OR rather than AND. Thus the 3rd else clause adding(unionizing) to the second set. Thanks alot, I was really hung up on this.

It's a classic case of divide-and-conquer via recursion.

Check out one of the more simple sorting algorithms based on a divide-and-conquer strategy, and you will find many similarities.

I also find that very hard to believe.
I mean, it basically has 4 elements. A loop of sorts, some if/else'ing, checking modulo, and echo/print/cout/whatever. If you can't get those lined up, you'd be delusional to apply for any programming job.

I am considered an "expert"* in three fields. I still cannot confidently say I'll crush a technical interview. It just depends on what they chose to throw at you.

Apart from that, from all the technical interviews I have taken and passed so far, the only topically relevant tests were from companies who knew exactly how, why and what for they were going to utilize me.
In all other cases (read: easily 80+%) real tasks and topics covered had negligible overlap.

Example 1: For my current project, I had a technical interview with a massive focus on design patterns. Primarily a combination of polymorphism, factory and decorator patterns. A wee bit of delegation pattern, too, and a smidgeon of test-driven development principles.
Now, having been on the project for a month, for the most part, I need to somehow figure out how some bunch of code works that has been programmed in embedded C between the mid 80s til early 90s and modify or extend the functionality thereof...
Overlap pretty close to 0%. Having said that, this is the most dramatic mismatch between interview and day-to-day stuff I have ever witnessed.

Example 2: For my very first job, the technical interview was based around polymorphism, pure virtual classes, class inheritance, and in addition other potentially tricky parts of floating point operations. In that case, I used the principles tested all the time.


* expert in quotes because from the project/program managers' perspective I know just about all there is to know but from my (and other "experts'") perspective I am somewhere at 7/10 max.

I completely agree with this.

I'm pretty much in the same spot. I have the suspicion that after this you're encouraged to be a project manager. Just to adhere to the Peter principle.