Saturated Fat Controversy

[FlatAssembler]

When one can write machine code directly in native binary on multiple platforms, one learns that compilers are for the weak.

Do you still do this manually? Sounds like a lot of effort to do. That's like someone still doing trig tables manually to this day.

It's not just about time. It's perhaps even more about the fact that programs written in Assembly (or, even worse, machine code) tend to be very buggy. If a Haskell program compiles, it probably works. If an Assembly program compiles, chances are, it crashes before even starting to do what it's supposed to. And Abaddon_ire claims that he or she is writing encryption software in Assembly and machine code. Well, chances are, if he or she is really doing that, those programs are very insecure both because they are buggy and because they are almost impossible to review by cryptography experts. Abaddon_ire has given me every reason to think he or she doesn't know what he or she is talking about. Why would a program that can translate 200 lines of code into Assembly, implementing QuickSort or the Permutations algorithm, be "not a compiler in any shape or form"? OK, maybe I went too far when I said you should show me a compiler you made or contributed to gain credibility to claim my compiler is worthless. But claiming that you are making encryption software in Assembly and machine code certainly isn't helping your credibility.

---

One of the first things we were told in our Digital Electronics classes was that AND, OR, NOT and XOR were considered basic ("osnovni") gates because they are implemented with 6 transistors, except for NOT which is implemented with 2 transistors. And, frankly, I am not interested in going much deeper than that, it's quite unlikely I'll have to do something that low-level.
That's not really the same thing. If you mess something up with UX, you will probably not immediately notice that, because UX is a softer part of computer science. Making a compiler or a PacMan game playable on smartphones is a more rigorous part of computer science: what you want to accomplish is quite well-defined and you can't reject it because you subjectively don't like the way it looks. Fine, maybe you can say you don't like GCC and that you like TCC more because GCC does many optimisations which slow down the compilation and that don't make the Assembly it produces significantly faster (hardly any programmer would agree with you, but it's not outright false), but you can't deny that GCC works just because you don't like something about it. And you can't deny that my compiler works, since there is evidence that it works (it can compile QuickSort written in my own programming language). Is it low-quality? Sure it is, the Assembly it produces is a few times slower than one produced by GCC for an equivalent C code, and the compiler itself runs slowly (running on top of the Duktape framework). But, frankly, I don't think you know enough about that to even evaluate those statements. Show me a compiler you've made, and then we can talk about your criticism of my compiler. Otherwise, you have no credibility.

When one can write machine code directly in native binary on multiple platforms, one learns that compilers are for the weak.

I think that writing some program in Assembly or machine code takes almost entirely different set of skills than writing a compiler. To write some program in Assembly or machine code, you most likely don't need to know anything about tokenizing, anything about parsing or anything about representing trees in memory and writing algorithms to convert ASTs into Assembly code (a naive left-child-right-child-parent algorithm leads to stack overflow even for relatively short expressions), and so on.

---

To give you some context about why that "I can write programs in Assembly, so of course I could make a good compiler." sounds silly, I was able to write a 800-lines-long program in Assembly back when I was 14.

---

Coding is your safe space, huh? The place you scurry off to whenever you feel like you just might have made an ass of yourself. It's cool bro, we all have them.

lol

Well, you claimed I have probably never done any substantial amount of research by myself. And, since you know something about computer science, you can see I have if I show you the compiler I've made. If you know something about linguistics, I can also bring up my alternative interpretation of what the names of places in Croatia mean.