Slashdot Mirror
Tips for Independent Learning?
Austin asks: "As an undergraduate seeking a degree in Computer Science, I am very interested in expanding my programming skills and tools. Much of the platform specific knowledge, such as the usage of unique languages and APIs such as .NET and Cocoa, are rarely taught on an academic level, and independent projects are looking very appealing as a way to broaden my abilities. I find it difficult to maintain motivation developing software that there are already excellent alternatives to, and contributing to mainstream open source projects is an intimidating endeavor due to their size and complexity. What have many Slashdot readers done to gain experience with tools not taught inside the classroom?"
Start with yProxy...
And if you are not learning stuff there, toss it and get another one.
Should holding down a part-time intern/development position be impossible for you while you are in school, get on rentacoder.com or similar sites and learn by doing shorter projects for other people. Eventually you will have your own ideas. The first ones are likely to be systems for keeping track of billing and worked hours.
When I want to learn a new programming environment, I pick out a simple project that I have already written in a known environment and port it to the new system.
This lets you focus completely on the new environment and not get bogged down in (high level) design decisions. It also gives you an apples to apples comparison in possible strengths and weaknesses of the two systems. (although any first project is likely to miss out on some possible advantages.)
# (/.);;
- : float -> float -> float =
magazines, not books, if you want to see more current information. Obviously, web pages ... )
can be even more current, but keep in mind there is lots of misinformation on the web
( hell, this comment could be just that
Get to know people who are already out there doing what you are interested in. Do things which
make these people happy to have you pestering them for knowledge.
Avoid seeking credentials, if knowledge is what you really seek.
Next, put the code aside; use the program, but don't look at the code for about six months or so, while doing other things. Then, come back, look it over and see if you can improve it. You'll have forgotten how you did things, so this will be similar to maintaining sobody else's code. By the time you're done, it will be better written, easier to understand and probably working better, and you'll have learned quite a bit more.
Good, inexpensive web hosting
I find the best way to learn is to have a real life project you are working on. Their are too many people in the world with "paper certifications". That is, they can pass a test, but could not do anything with that knowledge in a practical way.
I am in the process of studying for certifications. Three tests I was ale to pass with mostly work experience. One, I really needed to study... I found excuses to introduce some of the topics into my work. It helped a lot.
You, likely, are not working in the industry yet. Find a way to work on a real project. It might be one you dream up. It might be working cheap at a local company. I don't really know enough specifics to advise. I'll just say that dreaming up your own project requires a lot of motivation to continue to work on.
But bottom line, you need to be applying, or it is just book knowledge.
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
I too am an independent learner. Although I have taken several language courses at my local college, I have found them to be limited at best. One thing you might consider is starting a project on sourceforge for something simple. Do you find yourself doing an repetitive tasks that you could write something to automate? Do you have anyone that needs some web development that you could use Ruby on Rails for? There is also TopCoder. They have a lot of coding competitions that you can participate in - many have cash prizes. Although I don't have the links handy, there are also some sites that have student programming project suggestions that you could look at.
My
The first thing to learn about independent learning is to: not ask the world.
Proof by very large bribes. QED.
I would suggest getting a school project that you enjoyed working on and do the exact same thing with an API that you want to learn. By doing this you can easily compare your new work to your school project to make sure everything is working as expected. You will also quickly learn the advantages and disadvantages of the tools you are using because you will have a 'control on your experiment' so to speak. You will be able to have a more informed opinion on which tool is better for certain situations.
Abaddon: An Xbox 360 Indie game
There are thousands of them. Find one that uses the APIs/Languages/etc. that you are interested in and then get it source code. Spend time perusing that for a while. Next, look at the open bug list for the project and see if you can delve in an sumbit a patch.
You will increase your skills and you will be contributing to a project that other will benefit from. Every one wins.
I've made lots of small Cocoa applications on my own, and the two things that have helped me stay motivated on individual projects are:
.NET programming I would join it. Also, it can be easy to sign up for 1-2 units of individual study credit with a professor if you have a project idea (this is a great excuse to come talk to them during office hours).
1) Keeping them short (1-2 weeks to finish)
2) Having people to show them to
So if there is a group on your campus related to Cocoa or
You're going to get told this by everyone commenting: do. Just do. You'll pick up on it very fast.
Second, you have to have some ideas. Don't go out there with the mind set that you have to make something everyone will use. If it can make your life easier or amuse you, that's enough reason to start working on it. Even if software already exists that does what you want, chances are it won't do it specifically enough or it doesn't do it exactly how you want.
Third, re-write it. If you get something finished, I guarantee you won't be completely happy with the code the first time around. This will teach you a lot about your code.
Fourth, document it. Maybe this can go before or after three, but you'll learn a lot here, too.
Did you ever notice that *nix doesn't even cover Linux?
A couple of other people have suggested going to an open source project. I think you should as well. One of the best things that an OSS project will do is show you the true complexity of a working application, as opposed to the theoretical concepts presented in an education.
You will also learn far more good practices from a well known / respected OSS project than you will in your average organization. I learnt a ton just reading through the Mozilla source code and following how they broke out interfaces and practiced defensive programming.
Finally, you will have the chance to work with people who need you for something. Programming is very much a team exercise, and you can learn about teamwork as well as programming through the activity. Apprenticing yourself to a senior mentor is a great way to piggyback through your mentor's experience, as long as you have the necessary humility to have your work ripped to shreds...
Finally, don't sell yourself short. Alan Cox has a story about a random guy who hung out on the Linux IRC channel until someone told him to make himself useful. He ended up writing most of the IPv6 stack.
It's not the answer you're looking for, but after trying what you're about to try myself and fail at it, I can only give you the advice to learn the basics before trying to do more than tiny projects. Anything else leads to frustration.
.net. What is it going to be in 5 years? Nobody knows.
It might seem pointless to do the projects you're now doing in class. But those are the tools that give you later the ability to pick up anything coming your way. Technology changes fast. And people make the mistake to dig into the tech without learning the basics. 10 years ago it was RAD tools. 5 years ago everyone went crazy about Corba. Now it's
But when you have the ability to understand new technology quickly, you're already ahead of those that only learned a certain technology and don't understand the wiring under the board. Learn to understand that instead of hunting some specific implementation that's most likely outdated before you understand it if you don't have the foundation to learn it quickly.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
Yada yada yada.. And I'm not talking about your balls. You will need creativity.
0 00932.html
And for that...
http://www.gapingvoid.com/Moveable_Type/archives/
Easier said than done though.
I am a freshman computer science student. Here is my advice.
Remember that a computer science curriculum teaches the theoretical foundations of computer science. You start with basic programming (C, Java, or [if you're really lucky] a functional programming language like Scheme), and then learn about data structures, algorithms, software engineering, architecture, systems programming, OSes, compilers, and theoretical computer science, as well as some CS topics of your choosing (numerical analysis, graphics, more software engineering, AI, etc.). Discrete mathematics is also very important; you can't do much in computer science without it. And don't forget your EE courses like digital design (designing circuits using Boolean logic) and assembly programming; you'll get to know about computers from an electrical engineering standpoint (different, but cool, IMO). As much as I like mathematics and theoretical CS, sometimes you need to get close to and examine the metal. I personally liken computer science, mathematics, and electrical engineering to be the holy trinity of computing, but I might be out there....
Computer science departments typically do not teach the ins and outs of APIs and industry programming languages unless it ties into learning about the fundamental concepts explained above (Java's usage in CS courses, for example). However, this doesn't stop you from learning these APIs independently. My suggestion are the following:
Enjoy your life as a computer science major. It is truly rewarding and interesting. Now I have to return to my own programming assignment....
If you find it hard to keep yourself motivated on a project, one solution is to pair up with someone who shares your interest in the project. Admittedly, it's not always easy to find someone with similar interests and complementary skills. But if you do, having another person to bat around ideas with can be a wonderful experience.
You want the truthiness? You can't handle the truthiness!
The work experience itself is worth several years' worth of graduate school.
I'm sure you've heard that one before, and you need to be careful of it. That said, you found yourself a hammer (your new knowledge of whatever) so start looking for nails.
I've run into the same problem many times. I understand not wanting to re-implement things (I thought about making an accounting program for a little bit before deciding to just use Quicken, for example). But you need to be creative.
When I first learned Python I made a simple little game based on one of my favorite games that I used to play on my TI-85. This also let me learn OpenGL. The game was Blocks.
Trying to do more, I made a simulation (I'd call it a game but it wasn't interactive) called Itty Bitty City which also included more OpenGL. It wasn't too complex but was fun to watch (even if I did have Pie-In-The-Sky ideas about what it would do as usual).
I never updated my website because it was a hassle. I didn't want to do it in DreamWeaver or something like that. I wanted to automate things. Python was my favorite language at the time, so I wrote a little program I called SiteMaker to do it for me (ironically, that page is quite out of date). Python was rather well suited to that (it was command line, and needed good text processing) so I used it and gained a good knowledge of Python in the process (including the build-in modules you can use to open and send data across FTP connections).
After that I wanted a program to make it easier to make blog style entries on my website (which would call SiteMaker). I had been wanting to learn Objective-C for use on my Mac so I used that for the project and it was fun, and I learned quite a bit (even if I don't use the program because I didn't add enough features to make it useful). This was SiteBlogger.
When I wanted to learn Java 5 to get back into it (since it had features that fixed my biggest gripes) I made another simple game (which I took rather far) called Pond Game. I got to use all sorts of stuff in Java and got a good working knowledge (after quite a bit of time off).
I was still weak in GUIs in Java (have done no Java GUI programming ever) so I used Java to scratch a major itch I had and made Scheduler which I am now updating for a Senior Project. I learned how GUIs were done in Java and got an even deeper understanding of Java from this project. It also fixed my problem of hating to figure out a school schedule.
I had been wanting to learn PHP, so when a project came up at school to build a system for them (not an assignment, I work there part-time) and they asked me if I wanted to do it, I jumped at the chance to make such a system (which I hadn't done and included a large amount of DB programming) and used PHP (which they were perfectly happy with, and were actually going to suggest).
There are other little ones here and there. The main point of all this is that most of these didn't bug me enough to make them (Scheduler was the only real problem) but I saw them as opportunities to try out a new language or some such. I've learned to do this because of a problem that I would suffer from which I assume you suffered from: I'm curious and would read about a language or some such but not do anything and thus forget it all. I needed the practice, so I found places to do it. Whether these were things I needed for myself (Scheduler, SiteMaker) or just little games that I wanted to make or to use to try something (Blocks which was an OpenGL test, Itty Bitty Citty which was going to be an AI test).
My last suggestion would be to enter a programming contest. I've been enterin
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
It's supposed to be fun for someone with at least 1 ft of humour.
Damn, what is that organ inside your skull for? Holding hair?
Nearly everything you will need to survive will be learned outside of the classroom. Or at least, many great programmers that I know had to do their most productive learning outside of the classroom.
It is expected. That's how things work. Pay attention to the O of N stuff in the classroom, but really get down and dirty on your own.
That having been said, I'd start a few smaller projects that interest you and that have a very limited scale. An end-to-end e-mail system that solves all of the cryptography and accountability problems with the current mail system would be neat, but you'll never finish that for you first project. How about some simple scripts that let you control your home server by sending it e-mail? Your first stuff may get hammered out in a few days, and may not get terribly elegant code. But you'll learn and stay motivated. Be sure to finish your first project before you get bored.
If games are more of your thing (Go games!) start by creating mods of reasonable scope... Like jousting mods in Counterstrike that propel people forward when they fire their weapons, or simple gravity mods that alter attraction based upon the distance to the ground below. Or to really stay motivated, join a modding team. There is nothing more helpful than a good modding team.
Whatever it is that you want to do... start by doing it. Nothing in the classroom will really teach you to write scripts to be a sysadmin. Or to write graphics code. Or to control robotics systems in a production envrionment, find real information in databases full of junk, or how to resuscitate someone else's crappy code.
Find something you like, and figure out how to do it. Sometimes you get lucky and find mentors, but don't expect to find it in a college classroom.
The ______ Agenda
This is a really handy soft skill because so many employers are going to tell you what they want without being able to provide anything technical-- learning to analyze features from the outside-in will help develop the mental tools you'll need to deal with the non-technical types and that can really pay off.
Height: 38U, Weight: 0 Newtons, Eyes: #0000FF, OS: Gray Matter 1.0 (Alpha)
Go around and talk to people. Express interest in what they are doing. Explain your level of skills/need to acquire skills, describe something you have already done and how you went about doing it.
Consider:
- Web content management system. Good returns for modest effort, a gradual introduction to problems with a good ability to fix them, and it's a project you might keep working on for a few years so you get that lived-in experience.
- There are any number of small games you can write in a reasonable amount of time. Don't try to be too original the first time. You can also learn how subtle UI can be, as with games the difference between "nice technical demo" and "fun game" can literally be in the milliseconds.
- A text editor can be educational too. Again, promise me not to share it.
- Look at all the open-source software you use. Pick one to work on that you know, isn't Apache. A lot of open source projects have one or a very small handful of people, and you can both get into them in a reasonable amount of time, and actually contribute.
Once you've got a couple of things under your belt, you'll find all kinds of things to do. But when you're young, it's hard to know what you're even capable of. Any of these projects will be a good start.But promise me you won't inflict it on the rest of the world.
Do not attempt a Roguelike from scratch. Pick up Angband and mod it, maybe, but don't start one from scratch. A lot of people make this mistake, but to make any decent Roguelike from scratch requires some algorithms and design experience you don't have yet.
Get yourself working for a company for the summer. Accept just enough pay to keep a roof (maybe even your parent's) over your head and to keep you in beer and skittles. In exchange learn how to actually do the job you think you are training for.
By no means accept just any internship. Even if you like the name on the door, do your research. The best way to do this is to get clear in your own mind what you want to learn before you apply for the internship. Go into the interview with a plan and see if the company can provide what you want. Sometimes the right answer is a huge firm with formal mentoring. Sometimes the answer is a tiny firm where they will accomodate your individual needs and talents.
We try to give our interns real for profit projects or at least internal projects. Nobody is well paid, but they learn a lot. They tell us that in their exit interviews and their code sure looks a lot better in August than it did in June!
After a few years of updates and improvements (and graduation), it became my full-time job! So I would recommend this to you. Create a real product that you (and others) will use.
A Multiplayer Strategy Game for Mac OS X, Windows, and Linux
What should you do when you've gone through school and NOT done what Austin suggests?
Here is my story:
I went to school, having no clue what I wanted to do.
I figured out I LOVED coding, (yeah, doing it since age 7 kinda makes you like it...) and tried to make it my goal.
I got distracted by a lot of stuff, (Kicked out of house, working min. wage job because I couldn't network to save my life...) kinda fizzled out in the middle of my education.
I knew how to program, but had no idea how it actually worked in the real world.
After barely doing enough coding to pass my classes, got two programming jobs that I bombed out of cause I couldn't develop the social or organizational skills to either do the job or get help.
Working retail like 4 billion hours a week to survive, I squeeze out graduation and barely stay above water for a few years.
Now I finally recover my wits, (I had some legit mental health issues that are now more resolved.) I am faced with this quandary.
What next?
Right now, I finally have a somewhat computery job, doing consumer tech support for a major ISP, but the pay is garbage and there is no mental challenge.
My resume is a nightmare. I have little definable real-world programming experience, and my coding is rusty to boot.
Is it possible to break into real programming jobs later on in life? I feel 10 years late for the prom.
I figure I either need to come up with the plan from hell to make me a paid software developer in two years or I gotta find something else.
So if anyone has any ideas?
If there are other IT paths people could suggest I'm interested too.
(Right now my best solution is trying to go back to school to become a teacher, but I haven't given up on "DO" yet...)
Innovation is continuous. That is to say that you don't have to have an original idea to begin with as long as you keep asking if there's a better way of doing things or functionalities that can be added. This is a problem that plauged me until I realized that it doesn't matter that much. So basically just do it. If you have an idea for a project you'll find theres lots of new stuff to be done once you're in the thick of it. Otherwise look at other people's projects and see if there's anything you can innovate there.
"What have many Slashdot readers done to gain experience with tools not taught inside the classroom?"
I find money is a good motivation. Find a job that gives you room to learn (or in fact demands that you learn). Most CS, even most of the theory, can be readily learned from textbooks. The professors are just reiterating the textbook to you. The computer and manuals are the only tools you need, and presuming you actually have access to a computer all the other stuff is dirt cheap (compared to physical sciences...).
It's 10 PM. Do you know if you're un-American?
1. make sure you really understand the fundamentals of computer science. once you get the foundations really captured you will be able to learn a new technology pretty quickly.
2. don't make the mistake of believing that your technical skills will keep you employeed. as important as they are they are only about 30% of what you need on a daily basis to survive.
3. instead of doing "toy" problems to learn a language or an api, try doing an "end-to-end" project. write user requirements, do a detailed design, code, write formal test cases, use a configuration management system.
4. start defining your "personal process". use something like the SEI's Personal Software Process model.
good luck
I find it difficult to maintain motivation developing software that there are already excellent alternatives to
Once you get into the "real world", you will find that your statement above is ultimately false.
I have been programming "out here" for about 15 years now. Not as long as some, but long enough. What I have seen continues to astound me.
The worst? The number of companies that continue to "reinvent the wheel", and do so each time badly. A big one is "data conversion", the converting of data (in a database, flat files, etc) from one formatting convention to another. This is something that is extremely useful, wanted by about every company out there, and they pay big bucks for it - most of the time to the vendor they are moving to (who invariably writes "one shot" conversion tools, thanks to the myriad of proprietary undocumented formatting schemes dreamed up by vertical market vendors).
One of the big pushes in an attempt to alleviate this problem was to move to create interfaces to communicate to the "outside world", using what became known as EDI. Good in theory, a nightmare to develop for. Standards were drawn up (and you had to pay for them, unless you coded to the drafts) by the UCC and other bodies. Ultimately, companies ended up extended and fragmenting the standards, so you never really knew what you got was what you wanted, and what you sent was what was needed. Things were still "flat files", but at least in a "semi-documented" (if not always rigidly followed) format. Various GUI-based tools and such were created to work with and develop conversion between the EDI "standard" and the actual needed format (for each implementation). It was a little better. Then XML came along - we all know how that turned out.
I am not saying any of these things are bad. All of them had their good points, and all are still used. What I am saying is that there is still a need for all of this. If you can come up with a tool or toolset to allow you to set up conversion templates from any format to any format, and sell it cheap, you could make a fortune. Yes, there are other companies that have attempted to do this, all to varying degrees of success. It isn't glamorous or anything, but it is something that companies pay for, and would make a good project to enhance your skills. You can't do any worse than the previous guy.
I can think of numerous other projects you could work on for which there aren't good solutions for certain target markets. For example, if you did what Peoplesoft (Oracle, now) does, and you did it better (probably not too hard to do, given what I have heard from every user of their software solutions), and cheaper (ie, so small to medium sized businesses could afford it), you would have people beating down your door. Once again, this isn't glamorous or sexy software, but companies pay for it through the nose. On a similar note - a low cost, very easy to use and develop for reporting (business intelligence) package, on the near-level to Actuate's software (good luck here - they know their stuff, and have great support to boot), would also be a boon to business. The two big players are Actuate and Crystal, but there are others. Either of these type of projects would be great to develop and learn from, and might lead to excellent business opportunities.
Finally, something else that is not quite "needed", but is a currently niche market that needs lower cost alternatives. This area is the "personal fabrication" and "rapid prototyping" market. A complete, but very easy to use end-to-end solution is needed. Right now, you buy your fabricators from one vendor (or your custom build your own) and software from another (or use various open-source solutions), and depending on what you are using, things may "snap together", or they may need some "customizing" to get them to work. Even once you have the toolset down, you still need to use generally arcane CAD/CAM software to design your parts, then convert that using other tools (which may be a part of the set or external to
Reason is the Path to God - Anon
I tried google answers... let's see if it works for slashcode on ubuntu. http://answers.google.com/answers/threadview?id=71 5917