i began to write rogualike game (can not count which time maybe like 6
or 7) and i noticed yet what i noticed before

if you work on some codes when its small its liek linear mostly and this
makes fast prograss but later, esp with some types of codes/programs
when you add some functionality its like this functionality,
though logically it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in my
last case) and

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you ned
to jump to all this places (bookmarks may maybe partially help but only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not readinch
much (any) of programming literature so i dont know)

its quite unfortunate situation

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 04/09/2026 18:59, fir wrote:

i began to write rogualike game (can not count which time maybe like 6
or 7) and i noticed yet what i noticed before

if you work on some codes when its small its liek linear mostly and this makes fast prograss but later, esp with some types of codes/programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in my
last case) and

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you ned
to jump to all this places (bookmarks may maybe partially help but only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not readinch
much (any) of programming literature so i dont know)

its quite unfortunate situation

You can't spell. You can't work the shift key. You can't use punctuation.

I can't begin to imagine that any of your code actually compiles.



--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Richard Harnden wrote:

On 04/09/2026 18:59, fir wrote:

i began to write rogualike game (can not count which time maybe like 6
or 7) and i noticed yet what i noticed before

if you work on some codes when its small its liek linear mostly and
this makes fast prograss but later, esp with some types of codes/programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in
my last case) and

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you
ned to jump to all this places (bookmarks may maybe partially help but
only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not
readinch much (any) of programming literature so i dont know)

its quite unfortunate situation

You can't spell. You can't work the shift key. You can't use punctuation.

I can't begin to imagine that any of your code actually compiles.

LOL!

Maybe he uses Spanish C.

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Richard Harnden pisze:

On 04/09/2026 18:59, fir wrote:

i began to write rogualike game (can not count which time maybe like 6
or 7) and i noticed yet what i noticed before

if you work on some codes when its small its liek linear mostly and
this makes fast prograss but later, esp with some types of codes/programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in
my last case) and

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you
ned to jump to all this places (bookmarks may maybe partially help but
only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not
readinch much (any) of programming literature so i dont know)

its quite unfortunate situation

You can't spell. You can't work the shift key. You can't use punctuation.

I can't begin to imagine that any of your code actually compiles.

very boring and stupid comment... i met a few of such comments in my
life (probably few (like 5 or 7) once a year - and its not big amount, regarding a write quite number of messages in net) but such kind of
people imo are usualli intelectual lowlifes - so it is quite fortunate
they are unable to read it and im quite happy with that..i met also a
lot of people who can answer and see no problem - they just focus on
right things


--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 04/09/2026 18:59, fir wrote:

(It's not quite clear [to me] what some of your formulations mean.)

i began to write rogualike game (can not count which time maybe like 6
or 7) and i noticed yet what i noticed before

Are you saying that you wrote that many roguelikes already? - And
are still suffering from the below outlined situation?

if you work on some codes when its small its liek linear mostly and
this makes fast prograss but later, esp with some types of codes/programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in
my last case) and

It appears to me that you seem to be omitting a design phase in your
software writing - I'm deliberately not saying software engineering -
and you might also not use appropriate paradigms (if not the "wrong"
language, in the first place, given that you post in a C-newsgroup)
for a project like the one you currently do.

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you
ned to jump to all this places (bookmarks may maybe partially help but
only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not
readinch much (any) of programming literature so i dont know)

It seems the word you are searching for is alike "spaghetti-code"?

Or are you referring to the fact that complex systems that interact
in many ways need a better structured, more robust design?

I made good experiences with using object oriented designs and some
language that supports it; given that you post in a C-newsgroup use
of C++ might be a more appropriate language choice. (Of course you
should be, or make yourself familiar with OO-design before starting
to program [just C-alike] in C++.)

Note also that it's not uncommon to refactor your code at the moment
you notice unwanted proliferation ("rank growth"); modularize your
code, with low-level functional building blocks, high-level semantic
classes, etc. (I may be telling you just the obvious, you may know
that already, I don't know, but yet you are still having such basic
problems? That's somewhat strange.)

its quite unfortunate situation

I don't think that situation isn't under control of the individual
software writer; you could change that situation.

I'm not quite sure what you expect from your post. But you may find
some basic hints above to tackle your reported fundamental problems.

Janis

PS: I notice that for some reason "fir" is in my killfile, so don't
expect a reply unless your post is quoted by others (as was the case
here).


--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Janis Papanagnou wrote:

On 04/09/2026 18:59, fir wrote:

(It's not quite clear [to me] what some of your formulations mean.)

i began to write rogualike game (can not count which time maybe like
6 or 7) and i noticed yet what i noticed before

Are you saying that you wrote that many roguelikes already? - And
are still suffering from the below outlined situation?

if you work on some codes when its small its liek linear mostly and
this makes fast prograss but later, esp with some types of
codes/programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in
my last case) and

It appears to me that you seem to be omitting a design phase in your
software writing - I'm deliberately not saying software engineering -
and you might also not use appropriate paradigms (if not the "wrong" language, in the first place, given that you post in a C-newsgroup)
for a project like the one you currently do.

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you
ned to jump to all this places (bookmarks may maybe partially help
but only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not
readinch much (any) of programming literature so i dont know)

It seems the word you are searching for is alike "spaghetti-code"?

Or are you referring to the fact that complex systems that interact
in many ways need a better structured, more robust design?

I made good experiences with using object oriented designs and some
language that supports it; given that you post in a C-newsgroup use
of C++ might be a more appropriate language choice. (Of course you
should be, or make yourself familiar with OO-design before starting
to program [just C-alike] in C++.)

Note also that it's not uncommon to refactor your code at the moment
you notice unwanted proliferation ("rank growth"); modularize your
code, with low-level functional building blocks, high-level semantic
classes, etc. (I may be telling you just the obvious, you may know
that already, I don't know, but yet you are still having such basic
problems? That's somewhat strange.)

Yes, there's refactoring and creating modular code, and also something I
find effective from time to time is to completely start over from the beginning.

After everything I've learned from the previous go, this can sometimes
get put together quickly and is always more readable straightforward and correct. It's less of a hassle than it may sound like, but my programs
are usually 30K or less.

I do go in for modular code greatly, and you mentioned 6 or 7 of the
same thing, why couldn't that be made into a procedure?


--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Janis Papanagnou pisze:

On 04/09/2026 18:59, fir wrote:

(It's not quite clear [to me] what some of your formulations mean.)

i began to write rogualike game (can not count which time maybe like
6 or 7) and i noticed yet what i noticed before

Are you saying that you wrote that many roguelikes already? - And
are still suffering from the below outlined situation?

if you work on some codes when its small its liek linear mostly and
this makes fast prograss but later, esp with some types of
codes/programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in
my last case) and

It appears to me that you seem to be omitting a design phase in your
software writing - I'm deliberately not saying software engineering -
and you might also not use appropriate paradigms (if not the "wrong" language, in the first place, given that you post in a C-newsgroup)
for a project like the one you currently do.

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you
ned to jump to all this places (bookmarks may maybe partially help
but only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not
readinch much (any) of programming literature so i dont know)

It seems the word you are searching for is alike "spaghetti-code"?

Or are you referring to the fact that complex systems that interact
in many ways need a better structured, more robust design?

I made good experiences with using object oriented designs and some
language that supports it; given that you post in a C-newsgroup use
of C++ might be a more appropriate language choice. (Of course you
should be, or make yourself familiar with OO-design before starting
to program [just C-alike] in C++.)

Note also that it's not uncommon to refactor your code at the moment
you notice unwanted proliferation ("rank growth"); modularize your
code, with low-level functional building blocks, high-level semantic
classes, etc. (I may be telling you just the obvious, you may know
that already, I don't know, but yet you are still having such basic
problems? That's somewhat strange.)

its quite unfortunate situation

I don't think that situation isn't under control of the individual
software writer; you could change that situation.

I'm not quite sure what you expect from your post. But you may find
some basic hints above to tackle your reported fundamental problems.

Janis

PS: I notice that for some reason "fir" is in my killfile, so don't
expect a reply unless your post is quoted by others (as was the case
here).

no problem - its your funny problem..if you want to see an answer and
you killfiled it (problems of your type fellows ;c )

no i started some but never finished them - i always have big ambitions
in rogualike topic (want some like say 5 times more complex as adom at
least in core mechanics, in that aspect maybe more complex than few
times)

im not even expect to finish it its just kinde interesting (for some
reasons)


what i say its not my problem - i may jump many times and code its more
like general problem of programming..the problem which is rarely phrased
if any... (i know its not commonly phrased problem so in fact i dont to
much expect insightfull answer tbh, but some may comment on this_

design probably will not help as design of the game is quite good

i may even gave an example:

1) you have hero (which is commended by keypresses)
2) you have npcs/bots which are comended by simple alghoritm

i got code for punching - like complex routine where hero punches bots ,
other complex routine where bot punches fellow..
this punching code is working


now i wanted to add kicking

i must go to bot routine , then fellow routine (two different places)
i need to put fields on kicking in structures

like

float kick_rate; // 0.3 means 30% kick 70% punch
float kick_strength; //relative to punch
float kick_accuracy; //relative to punch
float kick_time; //relative to punch

int kick_counter; //used to level up kicking
int punch_counter;

unsigned last_action; //denote kick to not allow kick after ick etc

then i need to feel it up in initalisations/initialisers

then yet i have HUD code which shows attack precision ad damege and
after kicking i got two dameges and precisions so i need also jump there


if i not forget something (and i could - in fact i got yet two other
routines which are childs of the main ones - like those who update
damege and check for death)

so it shows there is 8 quite distant places i need to jump to add
kicking - i think design will not resolve this..this is just coding
reality that is wort to think about imo

(in fact some may say i coud write function of hiting symmetric to this
extent that hero and bots are the same structores so i could use one
code for two just put like hero at 0 index in array and all other
starting form 1 and above - but i decided to do it separatelly which im
not sure is good decison but its probably ok becouse if not i probably
would have problems becouse hero and npc's are not quite te same things

so this is a kind of theoretical 'problem' (topic) here


i mnot sure if it is 'resolvable' becouse it would need some more
radical changes in approch to coding maybe

this is gearal problem asi imo linear codes are good codes and when yu
got such 'knots' or how to call it you are both much more worse to write
it (jumping is terribly distracting) yet also probably analising it

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

fir pisze:

(in fact some may say i coud write function of hiting symmetric to this extent that hero and bots are the same structores so i could use one
code for two just put like hero at 0 index in array and all other
starting form 1 and above - but i decided to do it separatelly which im
not sure is good decison but its probably ok becouse if not i probably
would have problems becouse hero and npc's are not quite te same things

as to this in fact what i wrote to this moment is
in fact quite symmetric and if i use one code terating this hero fellow
same as npc i would have like close to 1/2 of typing

but im unable to say right now if this decision was wrong it is really
hard to say when i will meat more non symmetric parts (and there are non symmetric ones too, then i wuld need to use ifs etc )

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On Fri, 4 Sep 2026 19:59:34 +0200, fir wrote:

i began to write rogualike game (can not count which time maybe like
6 or 7) and i noticed yet what i noticed before

if you work on some codes when its small its liek linear mostly and
this makes fast prograss but later, esp with some types of
codes/programs when you add some functionality its like this
functionality, though logically it is one and 'cohesive' (? not sure
for this word) it is "spread" on many places in code (literally like
about seven in my last case) and

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you
ned to jump to all this places (bookmarks may maybe partially help
but only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be
linear but becomes like web or knot (?) (maybe it has name but im
not readinch much (any) of programming literature so i dont know)

?Spaghetti code?.

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 05/09/2026 1:59 AM, fir wrote:

i began to write rogualike game (can not count which time maybe like 6
or 7) and i noticed yet what i noticed before

if you work on some codes when its small its liek linear mostly and this makes fast prograss but later, esp with some types of codes/programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in my
last case) and

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you ned
to jump to all this places (bookmarks may maybe partially help but only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be linear
but becomes like web or knot (?) (maybe it has name but im not readinch
much (any) of programming literature so i dont know)

its quite unfortunate situation

Dear fir,

I have yet to read all the other replies, so I will just congratulate
you on reaching the stage of combinatorical and graph theoretic trou-
bles. This is well known in the field of /interactive fiction/, for
which I recommend the T.A.D.S. III programming language and environ-
ment. For a very, very good introduction to this field is also the
literary masterpiece /Leather Goddesses of Phobos/.

No self respecting programmer can go for long without a reference to
this masterpiece of interactive literature!

I have added rec.arts.int-fiction to this discussion.


Happy game making!
--
Johann | email: invalid -> com | http://www.myrkraverk.com/blog/
I'm not from the Internet, I just work there. | via Easynews.com https://bsky.app/profile/myrkraverk.bsky.social | for ( ;; ) _:;

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 05/09/2026 5:33 AM, fir wrote:

Janis Papanagnou pisze:

On 04/09/2026 18:59, fir wrote:

(It's not quite clear [to me] what some of your formulations mean.)

i began to write rogualike game (can not count which time maybe like
6 or 7) and i noticed yet what i noticed before

Are you saying that you wrote that many roguelikes already? - And
are still suffering from the below outlined situation?

if you work on some codes when its small its liek linear mostly and
this makes fast prograss but later, esp with some types of codes/
programs
when you add some functionality its like this functionality,
though logicallyÿ it is one and 'cohesive' (? not sure for this word)
it is "spread" on many places in code (literally like about seven in
my last case) and

It appears to me that you seem to be omitting a design phase in your
software writing - I'm deliberately not saying software engineering -
and you might also not use appropriate paradigms (if not the "wrong"
language, in the first place, given that you post in a C-newsgroup)
for a project like the one you currently do.

1) it is no way to denote it in code (though logically it should

2) the fact that you need to find that placeswhere to add it and you
ned to jump to all this places (bookmarks may maybe partially help
but only partially) is extremally wearing and extremally distracting

3) the situation has no name - that code as i said stops to be
linear but becomes like web or knot (?) (maybe it has name but im
not readinch much (any) of programming literature so i dont know)

It seems the word you are searching for is alike "spaghetti-code"?

Or are you referring to the fact that complex systems that interact
in many ways need a better structured, more robust design?

I made good experiences with using object oriented designs and some
language that supports it; given that you post in a C-newsgroup use
of C++ might be a more appropriate language choice. (Of course you
should be, or make yourself familiar with OO-design before starting
to program [just C-alike] in C++.)

Note also that it's not uncommon to refactor your code at the moment
you notice unwanted proliferation ("rank growth"); modularize your
code, with low-level functional building blocks, high-level semantic
classes, etc. (I may be telling you just the obvious, you may know
that already, I don't know, but yet you are still having such basic
problems? That's somewhat strange.)

its quite unfortunate situation

I don't think that situation isn't under control of the individual
software writer; you could change that situation.

I'm not quite sure what you expect from your post. But you may find
some basic hints above to tackle your reported fundamental problems.

Janis

PS: I notice that for some reason "fir" is in my killfile, so don't
expect a reply unless your post is quoted by others (as was the case
here).

no problem - its your funny problem..if you want to see an answer and
you killfiled it (problems of your type fellows ;c )

no i started some but never finished them - i always have big ambitions
in rogualike topic (want some like say 5 times more complex as adom at
least in core mechanics, in that aspect maybe more complex than few
times)

im not even expect to finish it its just kinde interesting (for some reasons)

what i say its not my problem - i may jump many times and code its more
like general problem of programming..the problem which is rarely phrased
if any... (i know its not commonly phrased problem so in fact i dont to
much expect insightfull answer tbh, but some may comment on this_

design probably will not help as design of the game is quite good

i may even gave an example:

1) you have hero (which is commended by keypresses)
2) you have npcs/bots which are comended by simple alghoritm

i got code for punching - like complex routine where hero punches bots , other complex routine where bot punches fellow..
this punching code is working

now i wanted to add kicking

i must go to bot routine , then fellow routine (two different places)
i need to put fields on kicking in structures

like

ÿÿÿ float kick_rate;ÿ // 0.3 means 30% kick 70% punch
ÿÿÿ float kick_strength; //relative to punch
ÿÿÿ float kick_accuracy; //relative to punch
ÿÿÿ float kick_time; //relative to punch

ÿÿÿÿint kick_counter; //used to level up kicking
ÿÿÿÿint punch_counter;

ÿÿÿÿunsigned last_action; //denote kick to not allow kick after ick etc

then i need to feel it up in initalisations/initialisers

then yet i have HUD code which shows attack precision ad damege and
after kicking i got two dameges and precisions so i need also jump there

if i not forget something (and i could - in fact i got yet two other routines which are childs of the main ones - like those who update
damege and check for death)

so it shows there is 8 quite distant places i need to jump to add
kicking - i think design will not resolve this..this is just coding
reality that is wort to think about imo

(in fact some may say i coud write function of hiting symmetric to this extent that hero and bots are the same structores so i could use one
code for two just put like hero at 0 index in array and all other
starting form 1 and above - but i decided to do it separatelly which im
not sure is good decison but its probably ok becouse if not i probably
would have problems becouse hero and npc's are not quite te same things

so this is a kind of theoretical 'problem' (topic) here

i mnot sure if it is 'resolvable' becouse it would need some more
radical changes in approch to coding maybe

this is gearal problem asi imo linear codes are good codes and when yu
got such 'knots' or how to call it you are both much more worse to write
it (jumping is terribly distracting) yet also probably analising it

Dear fir,

It seems you have problems with the concept of code complexity, and
state machines. I am not going to waste kilobytes of text to explain
these topics, but instead just refer you to /Patterns in C/ by Adam
Tornhill, who is wise enough not to rise to trollish baites on Usenet.

https://www.adamtornhill.com/articles/patternsinc/patternsincnew.htm

That is the author's website. I am going to assume you have the budget
for the book, because it's an ebook for sale. I happen to have a print-
ed copy I made myself.

Then there's the book /Object Oriented Programming in C/, which I admit
I have not read yet, but have commenced in procuring myself a self-made hardcopy. You can find the book at various places, I will link some,
and then a toolkit.

https://www.cs.rit.edu/~ats/books/ooc.pdf

https://www.mclibre.org/descargar/docs/libros/ooc-ats.pdf

https://ooc-coding.sourceforge.net/

Note that although /Object Oriented Programming in C/ is readily avail-
able online, I did not recommend it first. It does appear you benefit
more from reading Adam Tornhill's book first, but I do not know your
budget.

Of course, every self respecting programmer also has a copy of Loudon's /Mastering Algorithms in C/, and Sedgewick's /Algorithms/. My copy of
the latter is the 4th edition, which is only available in Java, but
everything I've read so far of it is readily translatable to C, parti-
cularly after reading the second book I linked. I admit to never hav-
ing read his /Algorithms in C/ book, but I presume he uses the very
same techniques as David P. Hanson in his /C Interfaces and Implemen-
tations/. That book seems slightly hard to get now, so I'm not sure
you can get a copy other than by happenstance in your nearest used
book store.

Now, as a side note on the people that bullied you before in this con- versation, none of them recommended any book for you, nor gave you any
tips on how to advance your own adventures in C. As I've said repeated-
ly, none of them show any sign of every having done anything in C, other
than perhaps thump the standard here in comp.lang.c.

Then, as a courtesy to people who like roguelikes on character cell dis-
plays, I want to recommend /Advanced C Programming for Displays/, by the
great Marc. J. Rochkind. That book is problematic to find, so best try
your luck in your neared used book store as well.

I added

o rec.games.roguelike,
o 3dfx.games.discussion, and
o alt.game.reinventing.misc

to this discussion.


Best wishes, and happy game coding!
--
Johann | email: invalid -> com | http://www.myrkraverk.com/blog/
I'm not from the Internet, I just work there. | via Easynews.com https://bsky.app/profile/myrkraverk.bsky.social | for ( ;; ) _:;

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Johann 'Myrkraverk' Oskarsson pisze:

/Object Oriented Programming in C/, which I admit
I have not read yet, but have commenced in procuring myself a self-made hardcopy.ÿ You can find the book at various places, I will link some,
and then a toolkit.

its probably (this punching/kicking) to much convoluted and not suitable
to make object with this..as this code of such general is making
something like web (or even web of webs) here not linear

i more think maybe possible approaches to this are
1) mental reckognition (of such tape of situations)
2) editor support
3) maybe some help from language like aspect aspects of coding

it also may show that many files form of code is not natural
(becouse if i wuld add herro fields i eventually could think not to hold global definition of a structure bt use its fields ad hoc
then the structure would not be present in code in one place
thus i would not need to jump there

generally i think languages could probably be designed this way to
minimalise this wearines when coding and this code jumpings atc it major source of weariness

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Johann 'Myrkraverk' Oskarsson pisze:

the stage of combinatorical and graph theoretic trou-
bles.ÿ This is well known in the field of /interactive fiction/, for
which I recommend the T.A.D.S. III programming language and environ-
ment.ÿ For a very, very good introduction to this fie

welll thanx, im not sure is i consider myself self respecting as i feel
old and weary

i like just to consider some theoretical problems..overally when coding roguelike it needs imo some special care - you need to
projecst those formulas with high care (its imposible to do it careless
it seems) and this is not easy piece of coding

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 2026-09-04 22:16, Lane W wrote:

Janis Papanagnou wrote:

On 04/09/2026 18:59, fir wrote:

(It's not quite clear [to me] what some of your formulations mean.)

i began to write rogualike game (can not count which time maybe like
6 or 7) and i noticed yet what i noticed before

Are you saying that you wrote that many roguelikes already? - And
are still suffering from the below outlined situation?

[...]

Note also that it's not uncommon to refactor your code at the moment
you notice unwanted proliferation ("rank growth"); modularize your
code, with low-level functional building blocks, high-level semantic
classes, etc. (I may be telling you just the obvious, you may know
that already, I don't know, but yet you are still having such basic
problems? That's somewhat strange.)

Yes, there's refactoring and creating modular code, and also something I find effective from time to time is to completely start over from the beginning.

A separate prototyping step may also be helpful; i.e. being aware
that the first version will probably anyway not be what's wanted,
or that you need "dynamic adjustments" (a'la Scrum processes etc.).

I've also the habit of testing the building blocks ("unit tests")
so that - with sensible interfaces defined - you may just compose
your system in reliable steps. But the high-level design is anyway
important.

After everything I've learned from the previous go, this can sometimes
get put together quickly and is always more readable straightforward and correct. It's less of a hassle than it may sound like, but my programs
are usually 30K or less.

With larger systems you have and need yet more things to structure;
cf. for example systems with multiple processes, external interfaces, decentralized servers, redundant systems and sites, etc. - Structure
is important on all levels.

I do go in for modular code greatly, and you mentioned 6 or 7 of the
same thing, why couldn't that be made into a procedure?

Frankly, I'm not sure what "fir" meant here with these numbers.

Rethinking about it, he might have meant that he needed a couple
attempts until he succeeded, or that he haven't succeeded even
after these many attempts, to create a roguelike?

Janis


--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)