Write Like Programming

OK, this will be an evolving guide, meant to be directly applied to the real process of writing on this blog, and hopefully, at least a subset of this guide will apply in other writing contexts.

I want each "rule" to be a solution or remedy to a specific problem. So what are the major problems that make writing hard, and even harder to sustain?

Problems

1. Time-consuming: From drafting to iterative improvement and refinement, or even major revisions, high-quality writing is no trivial task. Perhaps one of those hard to nail AI-research problems as well?

2. Can be wrong: It's embarrassing, isn't it. But think about it, the moment we find out that we did it wrong, really we have learned something, something "more right". So it is a good thing. When we find a bug in a program, what do we make of it? Emotionally, I admit sometimes it's hard. But deep down we know it's a good thing. As long as the intention is to make good software, we never intentionally put bugs in our programs: we want it to work, but almost inevitably we can only write imperfect code (assuming we're talking about non-trivial software). We can only make it less and less buggy. So why does natural-language writing feel very different when it comes to imperfection? What we do want to avoid is intentionally writing something wrong, for depraved purposes of misleading people. As long as we are honest, rational, and always open to solid evidence, revising what we wrote to make it better is just like fixing software bugs. You shouldn't be ashamed of making mistakes. But do correct them.

3. Keeping motivated is hard: Because for a personal blog, there is no external rewards. So you got to have fun, if long-term sustainability is what you aim for. Don't burden yourself that every piece has to be a major article. A short, easy one is fine, as long as it has substance. Yes, whether something "has value" can be subjective, but in spite of that, learning to enjoy writing on a persistent basis is an important life skill, as important as the habit of self-reflection and inter-personal communication (in fact, writing can be used for both of these essential components of life). So make it part of your living, and make it your friend, a true friend who listens and inspires. You tend to have fun when you are with a true friend, don't you?

Solutions

So, what are the corresponding solutions?

1. Time-consuming: Monitor your time spent on writing. Start with a lower budget: don't over-commit yourself, otherwise you will burn out soon. Sustainability is the key, just like learning. In particular, if you find that you tend to spend way too much time refining a rough draft, then don't overdo it. Just express the ideas using a crude, straightforward vocabulary. Admittedly, sounding like an achieved, mature writer is a wonderful feeling, but this goal is way beyond reach for any beginner, and is in fact only a superficial one. After all, the essence of writing is in the ideas, not finding the most elegant and fancy words to express them. Rhetoric is a powerful skill, no doubt about it, empowering people to describe and explain in artful, attention-grabbing ways, but it is meant for later phases of the learning curve. Getting your ideas out in a simple, understandable way is a vital prerequisite. As in programming, beginners, and even seasoned coders, don't aim for the most optimized design or implementation in the very first iteration: you prototype, make a proof of concept (or minimum viable product, etc), and then repeat through multiple cycles of redesign and reimplementation, to the extent that you deem worth. The benefit of this easy-onset approach is such that you get to play with it early and have fun, and if it turns out to be a good idea, then you will be naturally motivated to improve it further, and at that stage, it stops being a burden; instead it will be an indulgence when you work on it, gone is the notion of "time-consuming". So, take baby steps first. Don't be ashamed of being simple and crude in your writing. For example, try not to look up fancy words in the dictionary. Be straightforward, be honest, with your ideas, they are the real stuff, presentation is secondary after all.

2. Can be wrong: This is simpler to fix. Understand that you, like anyone, can be wrong, and fixing that is the job of "you in the future", not the current version of you. Just be honest and write to the best of your ability. Correct your mistakes when they are found. Another tip is, if you care very much about being right, then avoid putting subjective thoughts in it. If you just want to express your emotions, then go ahead and record whatever is in your mind at the time of the writing, but be aware that emotions do not tend to stay the same over a long period of time (there are of course exceptions, but not common). So if you write an emotional piece now, just don't expect that even you, yourself, are going to be as deeply empathetic when reading it in a few years, or even a few months! To be clear, no longer feeling the same emotions does not mean that the record of the past real feeling is "wrong"; time just does its magic on such things. So if you prefer something of stability, or something more practically useful both to the readers and yourself, then create content from facts and rational reasoning. Again, expressing emotions is a very helpful way of self-reflection, healing, and growth, but more importantly, guidance from rational analysis plays a critical role in its effectiveness, and that takes practices to do well.

3. Keeping motivated is hard: As we know, the most effective motivation, and arguably the only real motivation, to keep you doing something, is to have fun. Now having fun is actually a complicated thing. For instance, typically if you put more heart and soul into something, you tend to have more fun out of it. But that doesn't mean having fun is particularly difficult or tricky. Write about what you are truly interested in or passionate about. Learn to reward yourself even though no external reward exists. For instance, if you want to understand something, how it works, and write a solid piece to explain it in a pretty unique and effective way, then, this article can become tremendously useful if later on you need to refresh yourself on the subject, and you may find that your piece explained it really nice (it may sound ridiculous that you can forget something you once understood, but the fact is, we do not keep the full details in the working memory when they aren't needed for a long time; that's the point of keeping notes after all). Therefore, I think creating something useful, at least to yourself, can be pretty rewarding. The basic idea is, in order to get internally rewarded, you want to feel that you get more out of doing this than what you put in. This of course is "psychological math", and it is part of the trick. But to stay simple, let's just go back to the first solution for a start: do not over-commit in the beginning: the first batch of articles don't have to be super refined, because refined pieces are too costly to produce for beginners! You know, you don't have to prove your writing techniques to anyone, it's not like you're being interviewed for an editing job! The point of self-motivated writing is ultimately about self-growth. Even if you don't want to invest too much effort on writing full-fledged, publication-ready articles, just keeping a fun journal of your life is a good start. And that's what I've been doing since June 2020 on this blog.

Older Draft

I first started this article on July 5, 2020. And I over-committed. Talk about irony (see Problem 1 above). You know how much work it is to produce a good guide for writing? It's as challenging and complicated of a subject matter as writing goes! And somehow I decided that that should be my first article (as an introductory piece to the blog). I ended up putting it off after drafting the initial ideas, and further work became psychologically difficult to resume. It's just difficult to revise an existing, long piece, even more difficult than starting a new one! So let's step back, and take the iterative approach, one step at a time, using Git to keep track of revisions.

Anyway, here's the initial draft from July, just for the git record:

Years ago I thought it a clever analogy that programming is like writing, or rather, a sufficiently good programmer programs as fluently as writing in a natural language. Perhaps that is a typical fantasy of a non-programmer, or a noob, as people tend to like the idea of explaining away something special with common activities. Well I'm not going to argue for or against the quality of that analogy, but the analogy the other way around: general writing should take the mentality of programming, as it serves as a good framework to produce good result in a sustainable way, by which I mean, of course, writer happiness.

Programming can be perceived as dull, robotic, utterly unromantic, as the common myth is that people have to think like machines in order to tell machines what to do. That can be quite true in the early days of computer programming, but today it is a very different landscape, and I think it is only a natural process, an evolution of human behavior that is inevitable, when most of the foundational work, the hard, tedious, "thinking like a machine" part, has been done. The giant has liberated us all.

Now when we talk about programming, we focus on productivity, ergonomics, and in general ways to better organize a complex project, maintain readability and usability of code, improve and refactor code, test and benchmark code, user experience, and finally, even the social part, where the software authors interact with the users, or other interested programmers with the goal of making the software grow on a healthy path.

Now that is a very human activity, and it's not hard to see that this whole suite of mental techniques can apply to general writing, and it might be the fix for many problems in writing, or at least the kind of writing I personally experienced.

So what are some of the problems?

• Over investment on non-semantic constructs: choosing smart or beautiful words, finding the cool/chill ways of expressing a common idea, as if an easy to understand expression is a bug.

• Premature optimization: before an article is fully constructed and debugged, I start to obsess with the details so that it looks refined.

• Distraction by details. Articles can get long quite easily, and it becomes hard to go over the entire content over and over to see the "backbone" of the massive body. This is the maintainability problem. Maintainability requires that the author always have a clear grasp of the essence of the article, so that they can see the important defects or opportunity of significant improvements.

• Being afraid or ashamed of making mistakes or making subpar content. This is probably the most distinctive feature of programming: debugging is a certain, natural, integral component of the development process, and finding and fixing a bug is not regarded as a shameful failure, but rather a step forward. Of course, some nasty bugs had dire consequences, but the fact that they're discovered and eliminated is something absolutely positive.

Now based on the "axiom" that no analogy is perfect, I do not intend to use programming as a strict and literal guide to general writing, but I think it's worth doing the experiment, a trial if you will, to see if taking the programming mentality to the practice of writing can improve the overall sustainability of keeping a personal journal on the web.

I haven't been keeping such a habit mainly for two reasons:

1. It takes quite a lot of time and energy to produce something of my standard.
2. I feel that I have much more to learn and do than to say.

However, the worth of saying really depends on what kind of things you say. In general, communications are vital, at the very least, as a result of keeping a programming journal/notebook, I realized that it has become part of my thinking process, and I became dependent on it when facing non-trivial problems, or complexity in general, and I believe that is the power of communication, which probably lies in its explicitness, namely it prevents you from hand-waving, if you can't write it down, you probably don't have it in your mind (given of course your expression power is sufficient).

In addition, especially recently, when I feel grateful for those people in the past who have helped me, and the fact that I have come a long way through education (in the general sense of the word mostly), I start to feel more and more strongly and clearly that perhaps saying something is a responsibility, or, a pay-forward, to this one world we all live in. No pressure of course, I do not aim to guide people's souls, but I think keeping a public journal is part of being a social being, as we should all be, because we all have received so much from humanity.

The good part of having time passed is that you get to collect substantial amount of real data, and from that you get to see what works and what doesn't. As for writing, keeping a private journal, and more recently, keeping a programming journal (or whatever your profession is), are helpful.

But because they're private, I started to take the extreme approach of not giving any damn about presentation, and as a result, they are not presentable. I'm not talking about refinement and polish, but even the most basic quality like completeness. For instance, in my programming journal, I often write down a question that I don't know the answer of, so that it can guide me through the thinking process to solve it, but then when I found the answer or solution, I was eager to jump to the code and totally did not care to add the solution to the end of the journal, most of the time.

Plus, if the private journal serves as a mostly daily record, this upcoming web journal could serve at least as a monthly record of my journal, and this I believe is a process that should be an integral part of a person who are "mentally active".

So this journal, dubbed "JoT", is my new experiment. As the introductory piece I would like to list some guidelines, all tentative, just to help guide myself along, again via the power of explicitness.

1. Get to the point, be simple, straightforward, use understandable words and phrases, do not try to sound cool/chill unless the intention is to mock (which is itself not often useful). The important point is, if you think "beauty" just lies in choosing fancy words, then you don't have it at all. In practice, refrain from looking up stuff in dictionaries and thesauruses just to make your words sound special. Focus on the substance of the article, I guess another way to test it is, imagine that you have to translate the content into a different language, are you concerned that the other language might not be "expressive" enough to convey the same meanings and vibes without being too lossy? I do not intend to say that perfect translation is always achievable, I think some artistic words do face the problem that their substance depend much on their mother tongue, I guess that is not because the mother tongue have more language features or constructs, but rather, the tongue carries within it much of the cultural information in the background and that is the secret sauce that triggers readers' emotional resonance. Digress much? Well, here's a lesson: culture is beautiful, but often not as universal, and if I have to choose, universal is always what I go with, and I think whether one can appreciate the beauty in a universal work is a good measure of their aesthetic maturity.

2. Short pieces are OK, in fact, short is infinitely better than nothing. I think this is like physical exercises. The weird trick that I learned from Feeling Good is the "dare you do less/worse". If you feel not up to the game one day, go ahead and give a lousy performance, run slower and for a shorter distance for example, rather than choosing not to run, because your goal is not to compete professionally (I assume generally), but to stay healthy. The same goes for writing, the point is not to win awards, but to take the responsibility of social communication and at the same time to have fun and enjoy the process of materializing your thoughts and potentially sharing with fellow earthlings. Therefore, do not overdo it, this should be no "burn-out city" keeping a blog!

3. Do not obsess over the unimportant details when you get started on an article. Since I first started writing, I have been forming this extremely suboptimal perception of the workflow of producing a work, which is, just taking pains to write very sentence one after another, giving your best shot and even agonizing over every choice of word, and after all this labor, you expect the highest quality of work that you can create. And it has been giving me the illusion that it does get the result. But the very fact that I ended up not writing at all for many years is a solid proof that at least this approach is not sustainable at all. You write every article like your last piece, way too much energy spent on things not essential, and you start to enjoy the process less and less. I'm talking about writing a blog without any kind of marketing, e.g. via social media, just producing content that is useful or fun, and it has to not burn you out. And I find the programming mentality a good candidate for an alternative approach to writing. When I program, I have in my mind the evolution of the code, that's why I invest in keeping a decent Git history, and always start out to build the basic functionalities first. And because of the constraints of computer programs, I simply can't spend much time doing things non-essential to the program, because I get the immediate feedback that I'm not making use of my time and energy. But natural language writing is often highly unconstrained, and with the erroneous education of my grade school I have the lousy value built into me that the fanciness of wording and expression determines the quality and value of the piece, not the meaning conveyed via them, which mislead me into over-investing in the extraneous show-off mentality that is hurting my writing experience badly. It's time to change. I'm not telling myself to write robotically, stripping all the humor away, no, but everything should serve the essence of the article, which is the meaning that it's trying to carry.

Things to experiment with:

• 1 hour session each time
• plan the main structure first