Guys!

Hello fk0 Offtopic gone! ;( Stanislav raised the topic correctly if there was such a violent reaction! From the very beginning, what I want to say is that the topic of a ready-made set of procedures is not zero point (although not very far from it) - the same example with BGE - there is almost the entire GUI was released in the form of libraries. Another question - quite serious - is that the vast majority of libraries poorly or not documented at all (for example: there is a library with a function call_addr_2_file - what does this mean???). If you open source Linux, Unix, etc., you will see that half or even two thirds of the sources are there spent on comments. Unfortunately, I have never seen this in the source code, posted for public use on the ZX (although I do not deny that they are possible yes). This is a rather serious neglect in the description of teams and their roles inside procedures/functions leads to actually an obscure way of launching, which, the author, of course, knows quite well, but I (not the author) can’t figure it out I want to, and as the forum members have already correctly noted, I won’t. And how natural the result is that such a set of libraries is supported/developed only by the author. Therefore, I think it is very important to properly document and explain code - both inside the code itself - and in the form of additional documentation MANDATORY with examples. It will be very easy to work with such libraries - with oneparties - coders who want to somehow improve the code - I can do it without any problems It will be very easy to deal with it and add something where needed. C the other side is a programmer who is not going to interfere with the code - just takes libraries - prints an example of their use - and uses them to the fullest reel. If you look at how functions/procedures - and their sets - are usually organized in other software/hardware systems, you can see that their names in general then they correlate quite well (I would say even more :D) with those functions which they actually carry out. That. suggested for example for working with memory (by the way, there seems to be a universal driver for all types of memory - however, where can I get it in the form of a ready-made toolkit?) call all procedures mem_* To work with files - disk subsystem - iofile_* Etc. That. I want from the very beginning when developing such a set of universal utilities (which by the way, I have never used :D) from the very beginning, focus on FORM, and not on content, which, it seems to me, with good form - design of codes and examples of use - will develop one way or another - There will be people who are willing, even me ;))). About the manager. I agree with CityAceE that without a project manager will get stuck. First of all, on “unprincipled” issues - at least take a looka little higher how people are almost ready to get into a fight over something that is basically unimportant questions. The role of a leader in this sense is that he will take on the burden choosing one or another solution/approach, and everyone else will follow it - and implement the approach he proposed. And it’s quite natural that some Some people won't like this choice. In a commercial project there is a principle - “you they pay - you work", but here such a principle is of little use (? not at all applicable), therefore the person who will manage this project must be able to on the one hand, it is clear and accessible to explain why this or that was chosen solution and on the other hand, people who work in a team must respect his authority significantly, so that at the first conflict of this kind interests cannot be escaped by saying “they work and code here alone... I can’t work with them.” along the way." P.S. It seems to me that saving on source codes is stupid... one TR-DOS disk is enough for a very large set of libraries, but if you use pseudo-object code, then even more. The vast majority of more or less modern compilers support compiling/building from disk (even my favorite GENS). At the very least it is possible after all, spread over several disks, so I consider the size of the source files parameter, the importance of which I evaluate as "1" only from the end ;)