I have some general questions to Mr Tomasz. Sort of interview kind of questions:

1. What external tool did you use to create FASM?

2. You are a Math grad. Did math help you a lot in crafting FASM and its macro language? Any particular field of math?

3. What's the real use of FASMG and how is it different than FASM's macro language? In what field FASMG should be most useful and helpful?

4. What's the future direction of FASM and its by-product like FASMG?

5. What HLL language do you prefer the most to accomplish most of your programming tasks?

6. In your opinion, which one or two of current HLL languages available today that could be redone by using FASM/G and its macrolanguage so that it can perform much better and / or with cleaner design?

Take your time Tomasz. Don't have to answer them all if you don't feel comfortable with some of it

My environment at the time was DOS Navigator (it was a shareware back then) and a ton of command-line tools. I already had some programming tools I created myself, and fasm was not my first assembler. But in the very beginning I used Turbo Assembler, since I learned the assembly language working with the Borland tools. However, I did not own a personal copy of TASM, and thus creation of a self-compiling suite of tools was in fact crucial for me to become independent. When I started writing fasm I also already had a copy of NASM, but for some reason I never used it (I don't remember why I didn't like it, in fact the syntax of fasm shows an influence of NASM in a few places).

I specialized in the abstract algebra, and there is in fact a strong connection between the theory of programming languages and this field. But please keep in mind that I was at high school when designing fasm and I gained such theoretical knowledge much later. In fact, I chose to study mathematics because I felt already very confident in the field of programming and designing algorithms and data structures, and I thought that the math studies may be more valuable to me, by giving me more new knowledge. I do not regret this choice.

I initially had a simple plan that this would just be a base for fasm 2. But as I saw many examples of using pure macro language of fasm 1 for purposes other than x86 assembly, it inspired me to release this basic engine on its own, with examples demonstrating that it is even more capable than fasm 1 in this area. If one wants to create a set of macros to assemble for another architecture, fasm g may be a better choice than fasm 1.

I plan to continue adding new instruction sets into fasm 1, but I do not plan to add any other large features there. I really feel that I already pushed this architecture near the limits of its capabilities. As for the fasm g, it still may become a base for the fasm 2, but I would need some incentive, because for my personal purposes fasm 1 combined with fasm g are quite enough at the moment.

I really have no strong preference here, I use whatever is required at the moment (and when there are no requirements, I actually fall back to assembly language, I really find it very easy to express my thoughts in assembly). In recent years I very often had to use Java (usually combined with some scripting languages), but I do not really like it, I liked C# much more.

I'm not sure whether it is a good idea to implement a complex HLL language with macros. Maybe for educational purposes? I wouldn't try it myself, though.

thank you Tomasz Grysztar and fasmnewbie
nice mini-interview

7. As we are now staying at 1.71.39 version for quite some time with less and less bug reports coming in, can we safely say that FASM has actually reached its Mature phase in its development lifecycle? Is FASM 2 to be released as a fully-matured product?

I recommend this thread to anybody who wishes to formally ask Tomasz questions regarding anything, particularly on FASM and others in a more general and broader sense (opinions, directions, outlook, preferences, hobby, likes / dislikes, history, experience, insights etc).