I'm 39 years into a coding career with no CS degree. But as a compulsive auto-didact (probably a common syndrome among HN readers) it hasn't been a barrier. Self-learning is a continual, daily requirement for coders. If you can continue your education as needed for this job you probably also have the skills and interest to learn from scratch. I studied the same books and did the same exercises as my brother the CS ma…
38 years here, I had undergrad in Chem and part of a masters. Had no college classes or experience when I got my first programming job. Always worked at the leading edge of whatever I did and always had success. Was never an issue. Today I wonder if I would have even gotten an interview.
So you think it is harder for no CS degree people now?
I have a religious studies degree. I wonder if it’s prepared me better than a CS degree for the actual work I do. What did I learn with my RELS degree? Hermeneutics— understanding ancient texts written by lots of different authors in their original historical contexts, often which has been deeply redacted over time. Anyone who’s worked on any production code can see the benefit of being trained in that kind of thinki…
I often used to do this to save myself from sinking feeling of how I made a bad choice and wasted precious time that I could never get back. But let's face it... You can make this argument pretty much with any degree, let it be art, humanities, literature or whatever. But, no it didn't prepare you better than actual CS degree. All those things you described could be perfectly done as part of CS degree as well. Most g…
I would say it prepared you differently. The argument is between specialisation and generalisation. It could be argued that a large proportion of a CS degree is useless for one particular application. Who cares about assembly when you do python all day? Or how does that computer graphics module help you with a back end server? Especially when most people have forgotten it when they did it 5,10, 20 years ago.
Ah but one might argue there is transferable knowledge in these areas! Your knowledge of assembly helps you to know what’s actually happening to the computer and why that bug happened; or that algorithm for graphics is very similar to this one for a server. These are very valid arguments, and a similar thing is true for an even broader education. Deciphering ancient religious texts gave him a similar skill set as needed for programming, and potentially a different one to most other programmers. So much of life is learning how to reason in different ways, often they come in handy, and different experience leads to a diversity in thoughts, which gives an edge in solving new problems.
I am very much a proponent of generalisation and then specialisation (along with variety in this path) for optimal performance.
I have a religious studies degree. I wonder if it’s prepared me better than a CS degree for the actual work I do. What did I learn with my RELS degree? Hermeneutics— understanding ancient texts written by lots of different authors in their original historical contexts, often which has been deeply redacted over time. Anyone who’s worked on any production code can see the benefit of being trained in that kind of thinki…
I just posted some rambling thoughts in this thread about not having a CS degree (or any real degree) and being successful as a self-taught software engineer. I briefly mentioned how my non-formal studies of religion and philosophy have helped to inform me - mainly in the context of "machine learning". Your thoughts about how such studies apply to CS in general, though, are interesting. I noted that there are extreme…
Yes. But cs degrees are not about learning coding. We have coding schools for that. It’s about learning the intersection of computing and math. But having science in the name sounds cooler so people take that instead.
> Don't forget that you've probably also studied holy wars in considerable depth, too. This prepared him/her to understand the great conflict of vim and emacs users no doubt.
tabs v spaces may be an even greater conflict.
It’s not a conflict. A huge portion of the population is just still in denial.
It's fun when people assume a CS Degree means you are a developer. Practically all development is self-taught, a CS degree is only handy for those moments where you dig into some abstract theory, which in most development jobs hardly ever happens. You mostly spend (or waste) time on tools, frameworks and interaction (both with humans and other systems). Lots of common development is rather mundane compared to all the…
CS is more than abstract theory. It gives you a foundation that you otherwise are unlikely to get. Self-taught means learning things useful for your existing job or things you find interesting. I'm work almost exclusively on the frontend, but my CS degree gave me enough foundation that I'm fairly familiar with compilers (I wrote one in college for a required course). I don't necessarily enjoy writing compilers. But t…
In a cs degree building a compiler would not give you any points what so ever. Writing about Operational semantics would:
Writing a compiler is just the side effect of showing it could be useful. It’s not hard programming equals cs. It’s the science part of computer science that matter.
Well, my attempt to summarize OOP: OOP is a concept. It doesn't care about the implementation and neither should you. So if you are talking about integers, don't care about the implementation which might have limitations like MAX_INT, care about the idea of having a number which you can increase and decrease. That is a concept and it can be implemented in different languages and with different limitations. Next, unde…
Frankly and in retrospect, I consider this difficulty a sign of the unsoundness if OOP. While with typed FP you can get entangled into some seriously complex mathematical models, in OOP it’s just a bunch of metaphors and mental models you have to convince yourself are true. The effort poured into memorizing one rarely transfers to unrelated frameworks or different languages, or at least does so to such a high level that it doesn’t help that much anyway.
OOP is easier to use when describing a mental model as-is, without the analytical deconstruction that (typed) FP would demand.