I'm thinking about general modeling (diagramming) standards and tools such as BPMN, UML, data modeling, etc.
Why do we need modeling?
Usually on what phase is modeling essential for you? communication? document and report?
How can you benefit from them in work or daily life?
Thank you!
You coined the most important word already: communication
Not only with others, but also with your future self. If you write an application now, and return to it after a few months pause, it will be hard to find your way around. Certainly, this depends on the size and complexity of the application in question. But having some visual representation will always be helpful.
Also, diagramming the application before actual development is a great way to wrap you head around everything. Normally, you will detect problems in you initial idea that are easy to fix at that stage. And as said, these diagrams will become helpful in the future.
Showing others how the application works internally is equally important. Everybody has a different way to approach problems. So something that is intuitive for you might seem weird in someone else's eyes. This makes it hard to grasp the concepts of an application someone else has written.
Also remember that only few of us (is any) write perfect code, and do a perfect architecture. We are human, we make errors.
Let's say for example that there is this nifty design pattern you heard of called "decorator", and you want to use it. Naturally, the word "decorator" will appear in your code, and people reading it will think: "Hey, I know this pattern. No need to read the gory details. I'll just handle it as a black-box and use it." But what if you misinterpreted the pattern and implemented it incorrectly. Now the person using it as a black-box will run into problems. These may range from small bugs to non-compiling API-calls. If you have a diagram of the whole thing, it will be much easier to pinpoint the cause.
The biggest problem with modelling is to keep the diagrams in-sync with your code. During the project life-time you will make changes. Small ones and big ones. In a perfect world, you would update the diagrams, think about it, let it sink into your brain, and then implement it. But the world's hardly perfect. So you may end up implementing something before diagramming it (if it's actually diagrammed at all). This whole process is cumbersome. Be it because the diagram software you use is crap or just simply because of time constraints.
Personally, I like to create an initial diagram. And once I'm happy with it, I dive into implementation. I won't revisit the diagram for small changes. Yes, after a while the code will deviate from the diagram, but the Big Picture is still there. If I need to make a bigger architectural change, I will revisit the diagram for sure.
What I will do however is keep the in-code documentation up-to-date. Most documentation extraction tools (like javadoc) give you the possibility to use markup which is useful to make the generated docs readable and usable. Especially when using hyperlinks.
So, coming back to one of your points, where you ask what the benefit is during the daily developer life, I think that the larger diagrams don't make that much of a difference there. Simply because once you grasped the concepts of the project, you will not need the diagram anymore as you peruse the codebase on a daily basis. What comes in handy though are proper code docstrings. Primarily because many IDEs display them while coding, or at least make it easy to jump to them. With diagrams, that's not so much the case.
Diagrams however are useful to get started quickly with a project.
One more thought about flow-charts/activity-diagrams: I find these mostly useless. except for complex algorithms, as it helps you visualize the cyclomatic complexity. But quite honestly, I have never needed to write a complex algorithm myself. I always found ready-to-use implementation in either the standard library or in a third-party library.
And one final note: This question should have been posted on https://softwareengineering.stackexchange.com/ ;)
I can only speak for flowcharts and diagrams for state machines, but the answer might apply also here: If you chart something, you are forced to think about your underlying stuctures. Usually bad structures tend to be more hard / ugly to draw as a diagram.
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.
Its always said that more you program, the better you become. Sounds good and true.
But I was wondering if there is a proven route to becoming a better programmer.
Something like:
Learn a
Learn b
Learn c > 'Now you are good to burn the engines'
Try stuff around based on your learning.
The answer might be similar to a CS course roadmap, but I want to hear from successful programmers who might want to pitch in with something notable.
Thanks
It's not true that practice makes perfect.
It's perfect practice that makes perfect.
If all you do is keep repeating the same bad practices again and again, you'll only make it possible to create bad code faster.
By all means keep coding. But at the same time be critical of everything you do. Always have a jaundiced eye that looks for ways to do things better. Read widely to get new ideas. Talk to others about how they do things. Look at other people's code, good and bad.
There's no "sure" way to learn anything that I know of. If there was, anyone could master this.
All questions are rhetorical and meant to stimulate thought.
Technical parts:
Design Patterns - There are probably some specific to a domain but generally these are useful ways of starting parts of an application. Do you know MVC or MVP?
Basic algorithm starting points - Divide and conquer, dynamic programming, recursion, creating special data types like a heap, being greedy, etc.
Problem solving skills - How easily can you jump in and find where a bug is? Can you think of multiple solutions to the problem?
Abstract modelling - How well can you picture things in your head in terms of code or classes when someone is describing a problem?
High level versus low level - How well do you understand when one wants something high or low? This is just something I'd toss out there as these terms get through around a lot, like a high level view of something or a low level language.
Process parts:
Agile - Do you know Scrum, XP, and other new approaches to managing software projects? How about principles like YAGNI, DRY and KISS? Or principles like SOLID? Ideas like Broken Windows?
Developer Environment - How well do you know the IDE you use? Source Control? Continuous Integration? Do you know the bottle necks on your machine in terms of being productive?
xDD - Do you know of TDD, BDD, and other developments driven from a paradigm?
Refactoring - Do you go back over your old code and make it better or do you tend to write once and then abandon your code?
Soft skills:
Emotional Intelligence - Can be useful for presentations and working with others mostly.
Passions/Motivation - Do you know what gets your juices flowing and just kick butt in terms of being productive? Do you know what you would like to do for many many years?
My main piece of advice would be: don't be afraid to rewrite your own code. Look at stuff you wrote even a month ago and you will see flaws and want to rewrite stuff.
Make sure that you understand some fundamentals: collections, equality, hashcodes etc. These are useful across pretty much all modern languages.
Depending on the language you use - use lint and metric tools and run them over your code. Not all their suggestions will be applicable but learning which are important and which are not is important. E.g FindBugs, PMD etc for Java.
Above all refine and keep refining your work. Don't treat your work as abandonware!
Learn your 1st programming language a new programming paradigm or a
find a mentor you can learn from
Apply what you've learnt in a real world project
Learn from your mistakes and successes and goto step one
The trick is knowing what to learn first:
Programming languages - this is the place to start bcause you cannot write software without knowing at least one of these. After you've mastered one language try learning another.
Programming paradigm - i.e. object oriented, dynamic/functional programming etc. Try to learn a new one with each new language.
Design concepts - S.O.L.I.D, design patterns as well as architectural concepts.
People skills - learn to communicate your ideas.
Team leadership - learn how to sweep others and how to become a team or technological lead.
After that the sky is the limit.
I would look at improving roughly in this order, in iterations with each building on the previous one:
Programming concepts. Understand things like memory management, pointers, stacks, variable scope, etc.
Languages. Work on mastering several modern languages.
Design concepts. Learn about design patterns. Practice using them.
Communication. Often-overlooked. You can only become a highly valued Software Engineer if you can communicate effectively with non-tech people. Learn to listen and understand the needs that people are expressing, translate that into a set of requirements and a technical design, but then explain what you understood (and designed) back to them, in terms they can understand, for validation before you code. This is not an easy one to master, but it is essential.
Architectural concepts. Learn to understand the big picture of large, complex systems.
Learning a programming language is in many ways similar to learning a spoken language. The only way to get good at it is to do it as often as possible. In other works
Practice, practice, read and then practice more
Take time to learn about all sorts of coding techniques, tools and programming wisdom. This I have found to be crucial to my development. It's to easy to just code away and feel productive. What about what could be if you just had some more knowledge / weaponry under your belt to bang out that next widget.
Knowledge/know how is our real currency. The more we know the more we can make a better decision about how something should be done and do it faster.
For example, learn about:
•Development Practices, Software Design, Estimation, Methodologies Business Analysis Database Design (there are a lot of great books out there and online resources)
•Read Code - Open Source Projects are a good place for this. Read
Programming blogs
•Try to participate on Open Source
Projects.
•Look for programming user groups in
your town and/or someone who can mentor you.
And yes, as mentioned practice. Don't just read, do and watch how you will improve. :)
Practice, practice, practice.
Once you're over the basic hump of being able to program, you can also read useful books (i.e. Code Complete, Effective Java or equivalents, etc.) for ideas on how to improve your code.
First and foremost write code. Write as much as you can. Tackle hard problems. If you want to be a really good programmer you need to get into the guts of what you are doing. Spend a lot of time in debuggers looking at how things work. If you want to be a good programmer who really understands what is going on you need to get down to the metal and write highly async code, learn about how processors work and why SSE is so awesome. Understand threading primitives and be able to write them as well as describe what is actually happening in the processor. I could keep going here but you get the idea.
Second find someone who knows a lot more than you and learn. This relationship will work better if you are already deeply immersed in writing lots of code.
Third, spend some time in a large high quality open source code base. I learned a ton from the Quake I and Quake II code. Helped me be a better programmer.
Fourth take on hard problems. Push your limits. Build things that you thought were impossible. Right now I am writing a specialized compiler. I have learned so much just working on this for the last couple of months.
Sure, strictly speaking, the more you practice programming, the better you become at solving those sorts of problems. But is that what you really want?
Programming is a human activity more than a technological one, at its heart. It's easy to improve your computer skills, not so hard to improve your interpersonal skills.
Read "Journey of the Software Professional" by Hohmann. One of the concepts the concepts Hohmann describes is the "cognitive library," which includes both programming skills and non-programming skills. Expand your cognitive library, and your programming skill will improve too.
Read a lot of non-programming books too, and observe the world around you. Creating useful metaphors is an essential skill for the successful programmer. Why do restaurants do things how they do? What trade-offs is the garbage department making when they pick up the garbage every few days instead of every day? How does scaling affect how a grocery store does business? Be an inquisitive human to be a better programmer.
For me, there has to be a reason to learn something new... that is, unless I have a project in mind or some problem I need to solve, there's no hope. If that prerequisite is met, then I usually try to get "Hello, world" working, and after that the sky's the limit. So much of development these days is just learning new APIs. Occasionally there's some kind of paradigm shift that blows your mind, but that's not as common as people like to think, IMHO.
Find a program that intrigues you, one that solves a problem, or one that would simplify many of your tasks. Try to write something similar. You'll get up to speed very quickly and have fun doing it at the same time.
You can try learning one thing really well and then expanding out to programming areas that are associated with the things that you have learnt, so that you can offer complete solutions to customers.
At the same time, devote part of your time to explore things outside your comfort zone.
One you have learned something, try to learn something a little harder. Read and practice a lot about things that seem confusing at first time (lambda functins, threading, array manipulation, etc). It will take its time, but once you have practiced enough, what seemed confusing at first, will be familiar and easy.
In addition to the rest of the great advice already given here, don't be afraid to read about coding and good practice, but also take everything with a grain of salt and see what works best for you. A lot of advice is opinion.
Good sites to read:
-thedailywtf.com
-joelonsoftware.com
-codinghorror.com
-blogs.msdn.com/oldnewthing
A great place to get practice is programming competition websites. Those will help you learn how to write good algorithms, not necessarily maintainable code, but they're still a good place to start for learning.
The one I used to use (back when I had time) was:
http://uva.onlinejudge.org/
Learn more than one language. One at a time, definitely, but ultimately you should be fluent in a couple. This will give you a better perspective I think, and help you to become an expert at programming, rather than being an expert at a certain language.
Learn the ins and outs of computers at all levels, hardware, os, etc. Ideally you should be able to build your own system, install multiple operating systems on it, and diagnose just about every problem that can arise. I know many programmers who are not "computer tech people" and their failure to understand what is happening at every level becomes a major hindrance in diagnosing and fixing unusual bugs or performance issues.
As well as looking at 'last weeks code', talk to users of your work after delivery - be one yourself if possible.
Its not my bag, but some of the best coders I know have spent time supporting applications. The experience improved their product I'm sure.
eat breath dream the programming language your using (no seriously, it helps)
There are two kinds of learning -
1. Informal (like how you learned how to function in society- through interaction with peers and family)
2. Formal (like your high school training- through planned instruction)
If you want an entry-level programming job, formal training via an undergrad Computer Science/Engineering degree is the way to go. However, if you want to become a rock-star developer, it is best done by informal training- make unintentional mistakes and have senior developers curse at you, learn a design pattern because an app you are updating uses it, almost cry because a bad developer wrote a huge messy program lacking documentation and best practices and now you have to do several updates to it ASAP; thing of these nature.
It is hard for anyone to give you a list of all you need to know. It varies per area (e.g. a web developer vs. to a desktop developer) and it varies per company (e.g. Microsoft that sells software vs. General Motors that mainly just use it in their cars.) Informal traiing and being engaged in trying to learn to do your job better and get promoted is your best bet in my opinion.
To prove my point, everyone here has great answers but they all differ. Ask a rock-star developer how he learned something or when, why; they may not know- things just happen.
Practice, individually and collectively
Keep an open mind, always learn new things, don't limit yourself to what's familiar. Not solely from a tech perspective, ui design, people skills, ... Don't be afraid of what's new
Peer review, talk to people about your code, let people talk to you about their code, everyone has a unique way of looking at a problem and you will learn a great deal from peers
Love coding. If you love what you're doing, putting in alot of time seems effortless. Every coder needs the drive!
One small addition to these good answers. When I work on someone else's code, usually I pick up something new. If you have the opportunity to work with someone else that is of equal or greater skill, noticing their programming style can teach you tons.
For example, in C++ & Javascript I no longer use if() statements without braces. The reason is that it's just too easy to mistakenly put:
while (true) {
if (a > b)
print a
print b
}
This is an obvious typo, but very easy to introduce, especially if you're editing existing code. I just call it defensive programming in my mind, but little tricks like this are valuable at making you better.
So, find a peer or mentor, and work on their code.
I am not sure if the OP was looking for general advice on how to be a good programmer, but rather something more specific.
I know I am reviving this thread, but I found it because I was trying to see if anyone asked this question already.
What I had in mind was, can we come up with a "knowledge-map" of programming concepts similar to the map that Khan Academy uses.
As a programmer, I want to be able to visualize the dependencies and relationships between different ideas, so that I can understand what skill level I am currently at; what I need to know before tackling a challenging subject; and be able to visualize my progress.
The very belief in the roadmap's existence blocks the road to perfection.
I have a decent idea in mind. I don't see a lot of stuff around for what I have in mind and stuff that is around is implemented badly. Kind of like expert-exchange vs. stackoverflow.
I have not worked on a project by myself before, I've just been involved in working with teams for last couple of years. So I am scared to tackle what I have in mind by myself and doubting skills I already know.
Language/framework I consider myself proficient in is Java/Struts2 However, I've been wanting to learn RoR or Grails. So I need advice whether on I should implement initial version of the idea with tools I already know (Don't worry, be crapy) or start with learning new tool and side by side implement the idea?
It depends:
Are you willing to wait for the learning curve?
Is this idea so original that it will wait for you?
Is this idea so great that I would have very few bugs/ enhancements when released?
If so, learn a new language, If you have the time then it would be just great!!
but if
There is a business time limit ( ie. you want it before the next SuperBowl or Christmas etc. )
Nobody is thinking about it ( everybody thought Q/A was a solved problem when SO came into scene )
You may be victim of your own success and the lack of updates would allow your competitors add the requested features in copycat products and that would kill you.
Then make it in the technology you command right now. But make sure it does what you need and that technology may be up to the challenge.
Using the right tool for the job is the best, always, but I you need to cut a table TODAY an you only have a Hammer, start punching already!!!
I would go for a new language (and I did when I was in your place).
It makes it even more fun, and you will "kill" two birds with one stone.
If you want business approach, then do it in the technology you are most familiar with.
b.t.w You might want to make it a WIKI, or be down voted.
It depends on your goals, and what languages you're looking at.
If your goal is primarily (realistically) to learn and to have a fun side project, learn the new language. Learning a new language is a fantastic way to improve as a developer, especially if the new language is radically different (not C# to Java).
If this is really a business venture, and the new language isn't drastically more suited to the project, stick with what you know. But be realistic here about whether it's likely to be a real business venture or not. Knowledge has value, even if the project fails.
If this is a business venture, and the new technology/language is much more suited to the problem domain - you're still probably better off learning the new language. Using rails/ASP/PHP for a web page, even if you don't know the language up front, is (likely) going to be a lot better even in the intermediate term than trying to write a web-based service in C++.
I would use this as an opportunity to learn a new tool. Start out by implementing a few tutorial examples in each the new tools you're considering, and go with the one you feel is the best fit for your new project.
Even if your idea doesn't pan out, you'll gain valuable new experience.
I would say it depends on what you're trying to do. If what you know is what is the most commonly used thing for what you want to do, then stick with what you know and expand that knowledge. If a new language has been known to do what you want to do better, then by all means, go and learn the new thing.
Is the idea is business related or simply a pet project?
If it business related I will probably go with what I am best at. I don't want to have a learning curve at the beginning of my business. If the idea works, then at some point I could ported to a more suitable language.
If it just a pet project, then learn a new language. That's the best way to learn a new language.
I understand that Spencer's answer seems subjective ( java is dying part).But having worked on both Struts and ASP.NET , I would definitely recommend the latter for better tooling, community and resource availability. It definitely made me more productive.
I would also recommend learning something new. Could be asp.net or python,ruby or php. Because it is rewarding in that it gives you a different perspective.
The title may seem slightly self-contradictory, and I accept that you can't really learn a language quickly. However, an experienced programmer that already has knowledge of a few languagues and different styles (functional, OO, imperative etc.) often wants to get started quickly. I've seen a few websites doing effective "translations" in the form of "just show me syntax equivalence". I can't remember the sites now, but for related languages (e.g. Perl/PHP) it's quite common.
Is there a better resource that covers more languages? Is there a resource that covers idioms as well as syntax? I think this would be incredibly useful for doing small amounts of work on existing code bases where you are not familiar with the language. Looking at the existing code, as we know, is not always a good indicator of quality. Likewise, for "learn by doing" weekend project I always have the urge to write reasonably idiomatic, clean code from the start. Such a resource could also link to known good example projects of varying sizes for those that prefer to learn by reading. Reading a well-written medium sized code base can also be much more practical when access to development environments might be limited.
I think it's possible to find tutorials and summaries for individual languages that provide some of this functionality in disparate web locations but I'm hoping there is a good, centralised, comparative place that the busy programmer can turn to.
You generally have two main things to overcome:
Syntax
Reference
Syntax you can pick up fairly quickly with a language tutorial and a stack of samplecode.
Reference (library/API calls) you need to find a proper guide to; perhaps the language reference, or perhaps google...
With those two in place, following a walkthrough (to get you used to using the development environment) will have you pretty much ready - you'll be able to look up what you want to say (reference), and know how to say it (syntax).
This, of course, applies principally to procedural/oop languages; languages that require a paradigm switch (ML/Haskell) you should go to lectures for ;)
(and for the weirder moments, there's SO!)
In the past my favour was "learning by doing". So e.g. I know a little bit of C++ and a lot of C#.Net but I must write a FTP Tool in Python.
So I sit for an hour and so the syntax differences by a tutorial, than I develop the form itself and look at the generated code. Then I search a open source Python FTP Client and get pieces of code (Not copy and paste, write it self to see, feel and remember the code!)
After a few hours I get it.
So: The mix is the best. A book, a piece of good code, the willing to learn and a free night with much coffee.
At the risk of sounding cheesy, I would start with the language's website tutorial and/or FAQ, followed by asking more specific questions here. SO is my centralized location for programming knowledge.
I remember when I learned Perl. I was asked to modify some Perl code at work and I'd never seen the language before. I had experience with several other languages, however, so it wasn't hard to figure out the syntax with the online Perl docs in one window and the code in another, side-by-side. I don't know that solely reading existing code is necessarily the best way to learn. In my case, I didn't know Perl but I could tell that the person who originally wrote the code didn't know Perl either. I'm not sure I could've distinguished between good Perl and really confusing Perl. It would've been nice to be able to ask questions here at the time.
Language isn't important. What is important is learning your ways around designing algorithms and the proper application of design patterns. Focus on the technique, not the language that implements a certain technique. Once you understand the proper development techniques, any programming language will just become real easy, no matter how obscure they are...
When you put a focus on a language, you're restricting your own knowledge.
http://devcheatsheet.com/ seems to be a step in the right direction: it aggregates cheat sheets/quick references and they are (somewhat) manually reviewed. It's also wide-ranging. It still comes up short a bit in terms of "idiomatic" quick reference: for example, the page on Ruby doesn't mention yield.
Rosetta Code appears to be an excellent resource that includes hints on coding idiomatically and moves from simple (like for-loops) to things like drawing. I haven't checked out how comprehensive it is, but there are a large number of languages and tasks listed. The drawbacks re: original question are:
Some of the linking is not accurate
(navigating Python->ForLoop will
take you to the top of the ForLoop
page, not the Python section). It's a
wiki, this can be improved.
Ideally you could "slice" the wiki
however you chose to see e.g. the top
20 tasks for two languages
side-by-side.
http://hyperpolyglot.org/ seems to be an almost perfect match for what I was looking for. The quality is not always there, or idiom can be lacking, but it has the same intention and is pretty comprehensive.
Next year will be my graduate year to be an informatics engineering person and I am trying to find ideas about the jounior project. Actually, I have an idea of making an expert system engine. I worked with clips and prolog and I liked clips very much but it seems to be an old engine. Can any one advice me about this idea or give me sources for papers or any topics that may help me? I am thinking to use C language to obtain the high performance, and to build a robust data structure. Also, I am thinking about an idea (I dont know if it could be done) of writing facts and rules (like clips) and then generate a C++ optimal code from these rules such that I can obtain the speed of the machine and use exe file.
I need help to make this idea more clear and how it can be done. Specially because I read about fuzzy logic, nueral network and heard about the new generation of expert system, so I dont know how that can be related to such topic.
For your junior project, I would recommend against writing it in C. Your problem sounds like it needs correctness more than it needs speed. Writing it in C will take longer because you will need to implement a lot of primitives that are not included in the language or any standard library. Also, since C is relatively low-level, there are a lot of opportunities to make low-level mistakes. Write it in a higher level language that is closer to the problem domain. You will have more time to focus on your actual problem because you will spend less time getting the framework set up. If you already know Prolog, it would be good to stick with that. Perhaps you might consider Mercury. It is similar to Prolog, but also designed for speed.
JBoss Rules (also known as Drools) offers the best approach to rule-processing. It's written in Java. It allows you to integrate program components in the rules, and rule-bases into your program components. You can even build or modify rule-bases on the fly.
I've heard that Java is catching up in its ability to do math, but outside of that, you have nothing to fear from performance.