OSDev.org

The thread title is deliberately exaggerated. Have you ever felt that you have wasted a lot of time doing something that does not lead to anything? I have used computers more or less for my whole life but I started programming just a few years ago. It seems that the pre-programming period was totally wasted time when it comes to using computers. It is a good question whether using computers in itself is a huge waste of time. If those time-consuming machines are used, it would be best to do something productive. Programming, writing books, creating graphics, and composing music are good examples of productive work. I chose programming but a bit too late (bad pun intended). I should have started programming much earlier. It definitely was waste of time to use computers unproductively (e.g. playing games).

There is one problem still left because programming may not be very productive either. It is especially unproductive when you choose to do something that requires a lot of study and research before you can do anything. Have you heard of the operating system development? You may waste many years and realize that the only thing you have is a hard drive full of unfinished prototypes and tests. An outsider may call this as a waste of time. However, there is one product: know-how. Unfortunately, know-how has some problems. It may not be something you can think back when you are an old fellow. With bad luck, you may not even remember anything anymore. It is not something you can put on your shelf in your living room and show to your old friends. Know-how disappears.

To solve the wasted years problem, I think it is important to create something that is finished. It does not matter whether it is a music album or a computer game, or an operating system. I know that it is quite fuzzy to say when an OS is finished. Perhaps I should say that it is published as a stable version, have documentation, and has real users (albeit very few). I have decided to reach the goal of having an OS finished like this. It would be a concrete result of all study and research. Perhaps it will be a proof of not-totally-wasted years. I try to avoid endless feature creep that could prevent the OS being finished as defined above.

Please note that I am only talking about this as a spare time hobby because I am not programming all day long. Paid jobs are a different topic. An OS could be one part of the Life CV. Of course, it could be a profession too. For me, it is not going to be that because I do not have potential and will. It requires that I would fully concentrate on programming and forget all other things. Too risky.

No no no no.

Anyway, having some short term and achievable targets along with long term target (i.e. OS dev) is good idea. You just need proper time management.
Note that by targets I meant anything, not only programming.

Some would argue that programming is a waste of years.

A waste of years of work that could be done by hand

There is one thing I overlooked. Maybe it is not so bad that some spare time is spent doing nothing.

Indeed. We may call that "resting". : ) But like with everything, there must be a balance between work and rest.

Also, creatively, I try to keep a balance between producing and consuming.

As for the original question. I felt like wasting my time, while dealing with x86's legacy crud. I'd rather build something more modest from scratch than deal with machines complexity to introduce more complexity to it's software.

Right now I'm more focused on making music, but my on-hold programming project is a virtual machine, which I would like some day to build physically.

Learn to program or program to learn?

This is a list of books that fall into the "program to learn" category, i.e they use programming as a means to gaining insight. I've got them on my bucket list but there's just not enough time....

Turtle Geometry: The Computer as a Medium for Exploring Mathematics
Handbook of Practical Logic and Automated Reasoning
Computational Category Theory
Explorations in Monte Carlo Methods
Structure and Interpretation of Classical Mechanics
The Haskell Road to Logic, Maths and Programming

Also any book on writing simulations (discrete event or system dynamics, or, whatever) would lead to endless hours of playing around and writing programs trying to understand the world a bit better.

I have been thinking this more and I have another question. Do you think you could make progress in your OS project significantly better if you had a chance to do it like a full-time job? I do not think my OS project would progress very fast. At least not that fast it would justify the time spent on it.

@bwat: I like your approach to computer science. Especially the interest in compilers and programming languages.

I've worked as an OS designer but not on my own OS. Basically, you do a lot of bug fixing and document writing. New code is written when new hardware comes along. It is the customers who direct your effort - the focus goes where the money flows. Your personal projects should be directed by yourself and you shouldn't have to minimise the costs. The hobbyist has a great deal more freedom.


I'm just a practical programmer who likes a bit of DIY which also describes pretty much everybody that reads or posts here.

There are few things that make projects/users interesting to me. I do not mean to reject others but these are the things I pay attention to.

- Lack of free software/open source fanaticism.
- Interest in non-Unix-like operating systems.
- Interest in creating a programming language and a compiler.
- Everything is created from scratch (e.g. no ported programs).

I think the "free software/open source" clause might be the most controversial and flammable topic. There are users (like you I guess?) who mostly fill up the list above and it is like a shared basis. Even though there might/will be controversies, the vision does not fundamentally differ too much.