A few of us are talking about using Zed Shaw's "Learn C the hard way" to (re-)learn C. The basic goal is to spend 90 days seeing how far you can get. Anyone have words of wisdom on how to keep such a group enthused and successful?
A few of us are talking about using Zed Shaw's "Learn C the hard way" to (re-)learn C. The basic goal is to spend 90 days seeing how far you can get. Anyone have words of wisdom on how to keep such a group enthused and successful?
Beer or weed.
Self-teaching? group? Sounds like an oxymoron to me. Seriously though, how about creating a favorite project of yours written in C rather than following a book as a group (which you can do very well individually). The advantage of such an enterprise would be you would have learned how to work as a team by the time you finish the project.
What happens after the 90 days?Originally Posted by Leam
Originally Posted by brewbuck:
Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.
Personally I passionately dislike that book (well, the online version of it... I don't know if the print version is any better). If I wanted to polish up my C programming skills I'd choose something a bit more exciting as well.
The self-teaching part is that each of us goes through the book on our own. No one standing up in front giving a lecture. From what I understand of learning theory, all good learning is self-learning. The group aspect is that we're in the project together and can encourage each other to continue. It's easy to let go a bit when no one is asking how you are doing. After a while you realize it's been months since you did anything. I've done this more than once myself and am trying to avoid it.
So far, no one seems to have ever learned C except me. Mine was a college class a couple decades ago. Working on this for a while to make the entry barriers as low as possible should open the door to a larger project down the road.
On-line group, so we're just gonna have to smoke some electrons...
Assess. I'm thinking of picking another book, either K&R or "Understanding and Using C pointers". I have this wild fantasy of doing Linux kernel work down the road. Whether I get there before I die of old age is the question. Another possibility is to take the meta-learning about this sort of thing and do another group on the same book, led by one of the people who just went through it. Sort of a multiplication effort.
Well, I have to admit to some reverse bias. I've chatted with Zed on-line in years past, and read some of his other stuff. While K&R might be closer to the source, dealing with the IRC community's attitude turned me off on recommending that one. My original C book was Prata back in the mid-90s. Haven't looked at a newer version so I don't know what it's like. I have several C books and none of them have really helped me get back in the groove. Zed's LCTHW has a set of videos so it's multi-modal. Each chapter is pretty short so it's easier for a professional with limited time to digest and make progress. It seems to be decently direct so not a lot of fluffy reading.
I won't say it's the best ever written. However, if it get's me to where I want to go, then it's good. If someone else can reach their goals with K&R, or "Head First C", I'm happy for them.
Funny you should mention the IRC community. In my experience the IRC community has similar protests/aversions towards LCTHW.
That's what I meant; there's an "if it's not K&R it's crap" mentality. I have K&R. I've tried at least twice to learn C from the 2nd edition and I think once from the first. It may be a great book for some, but it hasn't gotten me where I want to go. Being beaten up for that frustrates me, and pushes me away from using the book.
The multi-modal thing is big, to me. I learn best from a book. However, the learning theory folks say you only get 17% the first time you are exposed to something and 44% the next. By watching the videos first I'm letting my brain get going and then maximizing the book. Since each chapter is short I can "fail early, fail fast" if I don't get the material right. There's only a few lines of code that can be wrong so I don't have to wade much to find a learning moment.
Again, whatever helps people make progress is good.
Last edited by Leam; 11-28-2015 at 05:45 AM. Reason: spellin'
It certainly is banded that way by more than a few. But it is very likely incorrect from a conceptual point of view.
Self-learning is a lengthy and chaotic process if it lacks any sort of orientation. It may end up giving the same results, but will take a lot more time and force the learner to go through a lot of obstacles that may in fact unnecessarily test the limits of his motivation. Many self-learners give up in frustration, considering certain subtopics "too hard", when in fact what they lacked was just some sort of adequate tutoring to reveal how simple those same topics really were had them been presented in a different or more organized way.
Self-learning is a better tool if the learner already has a good grasp of the field in question. This may seem redundant, but it really isn't. An astrophysics will go through a better self-learning experience if he wants to learn particle physics, for the same reason a C developer will have little trouble learning Java without any formal tutoring. This is so, because the "self-learner" already has enough knowledge of the field to be able to define a more rewarding plan of study, go through it at a faster pace, know how to research new knowledge and understand enough of his field nuances to correctly identify its idiosyncrasies on sight and not perceive them as mounting obstacles to his learning process.
Learning Theory should be taken with a large amount of salt. Much has been written and little has been actually said. And getting too hung up on some of its ideas -- especially when these ideas are taken from the headlines of poorly researched articles, on mass media news outlets written by non specialist journalists, cherry-picking conclusions from a scientific study -- is a sure way to give the completely wrong idea that there is some kind of Truth to Learning Theory. The learning process is an highly individual experience and it is even greatly dependent on the individual psyche at any given time. What is right for me, may not be for you. And what is right for you today, may not be tomorrow.
But I need to stress that this is not to say that self-learning isn't an option. It greatly depends on what do you already know of programming and what type of programming languages you were using before (moving from Haskel to C, or vice-versa is not so easy, for instance). But if your knowledge is only minimal, be prepared for a rough road that only a very few have actually braved to the end. To become a good, competent and desirable developer in a modern programming language solely on self-learning will take you a lot of time and be entirely dependent on whether you actually have the talent and the mental stamina.
Originally Posted by brewbuck:
Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.
Seriously though, how about creating a favorite project of yours written in C rather than following a book as a group
Polly, want a cracker? Spam post incoming @ Post #3,2,1..?
He's on my ignore list since post #2. Can't be bothered with idiots these days.
Originally Posted by brewbuck:
Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.
