How to write a good article

Written by Dario Phong

The title is very obvious, isn't it? I don't claim that my ideas are the best ones, however I think that you can learn at least something, and helping a little is better than helping not at all. Most of the ideas are focused for a technical article, I mean explaining an algorithm or something like that, though they can be used for writing any other kind of articles.

You have a topic to talk about, and you want to convey it to other people, so you decide to write an article, but you want to write a good article, so the question immediately arises: how to write a good article?

The first thing is to know what you are talking about, this will help a lot. However, you can do good articles without being an expert at it, too. Now you have to organize your ideas. It isn't a good idea to just start to write, and see what happens, at least it's not good if you want to write a good article. I recomend you to create a new text file, and list there the different subtopics and everything you want to tell.

Now you have to know what the audience of the article is. If your article is directed to beginners, assume no knowledge, fully explain everything. On the other hand, if you plan to talk about an advanced topic, you know that the reader will have at least some basic knowledge about it, so there's no need to explain all the basic ideas. You can directly go to what really matters.

So far we have seen two different kind of articles. Introductory articles and focussed articles. Now decide what kind of article will explain your ideas, and based on this decision follow one of these steps:

1. If it's introductory. These are one of the most important articles, they'll lead to the reader knowing enough to be able to read the other kind of articles. Here you have to explain the basic concepts, assuming only minimum knowledge. Example: If you plan to teach how to create a simple fire effect you should explain how to use the VGA, putpixel and some other required things, too. If you don't explain that, all the readers who don't know it will not be able to follow what's going on and they'll lose interest in your article or even in the subject you were talking about. I suppose you don't want to write discouraging articles, do you?

I have a kind of law for new ideas: The first time you talk about something, fully explain it, with examples, pseudo code and how it works. The second time say nothing. And the third time remind the readers what it is.

As soon as the reader has read the main ideas, he can be in two states: Either he still doesn't know what you were talking about. In this case it will be good for him to put some pointers to other introductory articles or tutorials. Or maybe he fully understood it and is ready to use the new knowledge. In this case, your work isn't finished yet. You have to tell him about other similar topics or improvements of what you talked about.

2. If it's a focussed article. This kind of articles is very important too. It will help to further improve the knowledge of the readers. Here you have to talk about new ideas, improved algorithms or fully explain a given scheme. Of course you can assume that the reader you aim the article for has enough knowledge so you can start directly with the hard stuff. However, it's good to tell where you can find the basic concepts and link to other articles about similar subjects. This is called references. You should list the names of articles or books which you read and which inspired you to write the article. You should also include the date of the article and its author. It's also advantageous if you put a pointer sign before the reference.

In this kind of articles the reader wants facts and ideas. Base your ideas on the acts or validate your ideas with facts. List the pros and contras of the idea being discussed, so that the reader can choose whether he wants to use it or not.

When a reader looks for introductory articles he needs them, so he'll read the first thing that he get. But as for focussed articles, the reader may have tons of articles about similar topics and has to decide what's worth reading. Therefore you have to write a solid introduction, mainly talking about the following things:

1. What you talk about.
2. What idea the article is based on, or what idea it improves.
3. How good the article is.

With these three things, the reader can easily decide whether he wants to read the article or not.

Once you've chosen the kind of audience you can start developing the article. Every article contains some particular sections. Let's have a look at them.

1. Introduction.

2. Explain the idea the article deals with. Put some examples. Explain what the idea is based on, how it will work out in some particular cases. Here you don't need to talk about the details of the concept or the algorithm itself, only how it will do things.

3. Explain how to implement. Give pseudo code or example code. This is one of the most important things. You have to tell the readers how to implement what you've explained. However, avoid giving executable code, for some reasons: if one reader wants to implement it in another programming language the code will be no help. Another reader may skip all the explanations and just get the code. So, why did you write the article at all? But I think that example code has some avantages, too. It shows how to actually implement the concepts. This you can do with pseudo code, too, which has a good thing, as the reader has to implement his own version in a programming language of his choice, and so he will learn twice.

4. Conclusions. There you should do a little overview about your scheme versus others, and tell how it could be further improved or used. If you want to be polite, now it is time to thank the people who helped you develop the article.

5. References. If the reader wants to read more, there he should find what he is looking for.

Now, choose a topic which you can talk about and write an article. Use any of my advice if you want to. The more you write the better you write, so don't expect your first articles to be the best ones.

Also never expect feedback. The readers are too lazy, so most of the time they won't react to your articles. Sad but true. Maybe this makes some persons refrain from writing more articles. If you want them to believe in you, start by believing in yourself.

Good luck, and see you in my next article!

- Dario Phong, Barcelona 09/07/1999