News:

Simutrans Tools
Know our tools that can help you to create add-ons, install and customize Simutrans.

About that splited topic

Started by Iluvalar, January 14, 2016, 11:41:43 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Iluvalar

Just answering to that post from Isaac
http://forum.simutrans.com/index.php?topic=15095.msg149488#msg149488


No, I don't think there was anything else to say.


Jamespetts do not understand what I'm trying to explain, it's a bit frustrating for sure, but that's life. He gonna waste what I assume look like 3 days of codings and 1 week of playtesting before he can see for himself what I meant. But that's also life, he will learn something. Hopefully, my conversation with him will cut his playtesting time a little at least as he will start to notice the problem earlier. So it doesn't go in waste.


I was in the impression, from the length of his answers, that he was somewhat interested in that conversation, If I've annoyed him, as I said before I'm sorry. I didn't meant to.


I'm also aware that it was going a bit offtopic, I don't mind the split, but at the same time the nature of his original post lead to a lot of off topic.  So I didn't felt the need to split myself. I might have been wrong, I still don't see the point in splitting.


Also, I'm fully aware I was using negative words, like failure, chaos and bankrupt. It might have gave to the whole post a negative tone that I didn't meant to create.


It is all there is to say about it. Unless you want to explain something else that I missed ? All this is just a cultural bias I guess.

IgorEliezer

#1
Edited post: Isaac explained me what happened in dept.

Sorry for the inconvenience. :)

Isaac Eiland-Hall

Igor - I invited this topic, so I'm unlocking for now. It may lock again; not trying to overrule you at all.

Iluvalar: The main reason I gave the warning in that topic when I did, and then subsequently split and locked: You were going on and on to the point where you were causing frustration for people trying to reply. Misunderstandings happen, but doggedly going on about something you'd like to see happen when it's already said that it won't; you were asked to try the game, and you said you would, but still went on and on after clearly not. It crossed the line into way too much. That's why I had to put a stop to it.

Debate and discussion are fine. You can see that all over the forum. It's when it becomes disruptive that it has to stop.

It's not about the lack of communication, it's about going after the same question again and again and again. That was the problem.

IgorEliezer

#3
Quote from: Isaac.Eiland-Hall on January 15, 2016, 04:40:59 AM
Igor - I invited this topic, so I'm unlocking for now. It may lock again; not trying to overrule you at all.
No problem. :)

An_dz

We stop discussions that won't go anywhere but in people losing their minds.

James got clearly irritated when you continued to tell him your idea was possible and would not cost anything or be simple. He knows the code, he is graduated on this area, he is the maintainer of the game and he plays the game. Are you all of these? I'm not, and so I try to understand why my assumption was not in accordance with what a coder said. ;)

And about this:
Quote from: Iluvalar on January 14, 2016, 04:22:56 PM
the posts you censored.
Don't flame :| , the thread was just locked and it's still available for anyone. When people do this they are just saying: "Hey, you are getting really annoying, seriously, stop."

I know you want to help, but it's not by repeating the same assumption over and over that you'll make anything. ;)

See Ves reply, he did a great request without pushing the idea like absolute truth.

I hope this can more clearly elaborate our point of the view of the problem.

Iluvalar

Quote from: An_dz on January 15, 2016, 06:39:05 PM
He knows the code
This have no impact since most of the suggestion is a new simulation. Which he could then use the data from to run his code.

Quote from: An_dz on January 15, 2016, 06:39:05 PM
he is graduated on this area,
I code in PHP since 15 years and worked in the domain for a while on top of being a benevolant technical support for a while.

Quote from: An_dz on January 15, 2016, 06:39:05 PM
he is the maintainer of the game
I once forked and improved the A* algorithm in another game called warzone2100. Which was writen in C.

So yeah, I'd say yeah I'm kinda all of these. Actually, he openly said he's not very familiar with the A* algorithm, so I'm probably ahead of him in that particular topic. :)

So yeah I DO consider the idea to fork the project and try for myself seriously.

An_dz

You still sounded annoying, like this post you just made.

I hope to see any progresses you make, seems interesting.

HarrierST

Quote from: Iluvalar on January 15, 2016, 09:19:54 PM
I once forked and improved the A* algorithm in another game called warzone2100. Which was writen in C.

I used to write C programs commercially and also teach people  how to program. But this is C++. A big difference.

C is the old mind set of programming (which includes me).  I did start to learn C++ in the early 90's. But my career path changed. So I dropped it. In those days you could not download free versions so I did not continue it as a hobby.

C++ requires a different way of thinking, to design and then to program.   But once mastered makes some changes/additions quicker.

So unless you can design and program in C++, you can't really say it is easy to implement somethink.

Just my opinion.


Isaac Eiland-Hall

Quote from: Iluvalar on January 15, 2016, 09:19:54 PM
So yeah I DO consider the idea to fork the project and try for myself seriously.

That would be quite a good solution.

So going back to what went wrong in that other thread: It doesn't matter what you know. The problem was that you repeatedly pestered a Simutrans volunteer. Discussion is one thing, but you came back again and again and wouldn't listen to explanations given.

So if you end up with a fork that does some things better, and if the devs want to use your code, and if Simutrans is improved because of it all, that would be fantastic.

Meanwhile, this thread is not about possible improvements to the code, it was about the thread that went wrong.

Does this make more sense? I should have shut down the other thread sooner, so that's a failure on my part.

Also, to clarify: Your welcome here is not affected by this in the slightest. It was a problem and it got fixed. :)

I won't lock this thread for now, but let's keep it on topic (and hopefully that topic has been addressed) - programming is not a part of this discussion.

jamespetts

Iluvalar: I very much appreciate input from experienced programmers. I hope that you are not discouraged from contributing in general to Simutrans, either Standard, Experimental, or any other fork that you might wish to create.

However, in order to have a constructive dialogue, it is important (1) to be moderate in your tone and respectful in your content, (2) to engage in the substance of the responses made to you if you wish to disagree with somebody, (3) explain what you mean with sufficient precision when asked, (4) ensure that the degree of confidence with which you make any claim is commensurate with the degree of your understanding of the thing about which you are making the claim and your ability to communicate that understanding with commensurate precision, and (5) not make accusations that others are not acting in good faith without a seriously overwhelming basis (in which case, you should consider a report to a moderator in any event).

The reason that the discussion in this thread was not constructive was because the above was not always observed.

For example, your references in this post to "hiding the answer" or "refusing" to tell you the number of trips per month generated by a typical building (as you described it) infers bad faith without any basis for doing so. You had not, in fact, asked that specific question before. See (1) and (5) above.

In this post, you did not engage in the substance of my response to what you wrote about what you believed would be a "sawtooth effect" in this post: you repeated what you had earlier written on the subject without acknowledging my response to it (see (2)) or explaining how what you wrote related to the way that I described how the system actually worked ((3) and (4)).

I had asked in this post whether you had tried testing the passenger generation system about which you were commenting yourself: you did not reply to that question (see (2)), but it later became apparent that you had not, in fact, tested it, but that did not seem to diminish the stridency of your claims about how you thought that it would work (see (4)).

I pointed out in this post that your understanding of the consequences of job slots not being entirely fulfilled for a short period (set out here) was incorrect (and therefore that your claims about the adverse game-play consequences of what you thought would be a "sawtooth" effect were also not correct). Without acknowledging that (see (2)), you again wrote in this post of businesses going bankrupt. When I pointed out that I had already noted that there is and is proposed to be no mechanism for businesses to become bankrupt at all (let alone in consequence of a short-term shortage of workers of the sort that you appeared to envisage), you responded in this post by setting out a proposed way of simulating such bankruptcies without explaining why such a system was necessary nor, more importantly, acknowledging that the lack of such a system at present means that the conclusions that you had already reached were not well founded (4).

I asked you in this post to describe how your proposed system might work and fit in with the existing system and fulfil the design goals (already fulfilled by the present system) set out in the ante-penultimate paragraph of this post, but you were unable to do so, (3), which is inconsistent with the stridency of your claims that the system that you propose would be better than what already exists (4).

In this post, you wrote, "I gonna try to explain the statistical problem again. slowly," which is not respectful (see (1)) for reasons which I hope will be obvious, and likewise in this post your statement that, "I feel just like Malcom in Jurassic park trying to explain chaos theory, statistical certainty and butterfly effects to people who don't understand such things. lol.". The implication that the problem was others' understanding rather than either your misunderstandings (and see also (4)) or incomplete explanations (3) is the problem here (you could perfectly well have been neutral, for example).

The overall impression created by the combination of the stridency of your statements, the sometimes disrespectful tone, the inability to explain with precision what you envisage, the failure to engage with responses in a number of instances and the somewhat odd sequence of events relating to business bankruptcies described above was that, on one interpretation, it appeared that you wished to persuade me and/or others that there were flaws in the existing system whether or not that was in fact the case in order to promote the alternative system that you were suggesting. I am not sure whether this impression is correct, and will give you the benefit of the doubt, but you should be aware of the sort of effect that the way in which you conducted the discussion is likely to have.

I have taken the time and trouble to set out - in some detail - the prerequisites to constructive discussion, the ways in which the problematic thread did not fulfil those prerequisites, with the result that the discussion there was not constructive, and the consequences thereof with the aim that you will be able in the future to engage in a thorough discussion of any topic without these problems recurring. I have done that because it is worthwhile to have people engaging in that sort of constructive discussion relating to features and development, as it helps to improve the overall result, and that is something that benefits all people who play Simutrans, whether Standard or Experimental. By contrast, unconstructive discussion has the opposite effect, as it takes time and effort away from development without contributing anything worthwhile in return.

I shall look forward to constructive dialogue with you in the future.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

Iluvalar

#10
Argh! I accidentally hit "modify" instead of "quote" and didn't realize until I completely overwrote the comment. I'm very sorry. In my defense, I've only managed to do this a couple of times in the history of the forum. d'oh! -Isaac Eiland-Hall

Vladki

#11
Please forgive the mod note, but this is an example of a reply that doesn't belong in this thread. I'm leaving it so it might be referenced elsewhere if a discussion of the topic might continue, because it's not that the topic itself is forbidden; just the way it was done. -Isaac Eiland-Hall

Ah, now I think I understand your statistic simulation...

Let's pretend that we simulate 300 jobs by 3 jobs (I'm not sure if it is correct but that does not matter now).
You say that if only 2 out of 3 simulated jobs are filled, you can also say that 1 out of 3 was missed. And when you scale up to 300 jobs it can be anything between 2 jobs filled (298 missed) and 1 missed (299 filled). I think the simutrans simulation does not work this way. It just simply scales up, so 2 out of 3 simulated jobs would mean 200 out of 300 jobs. No more no less.

I hope James will correct me - I have not studied the code either, that's just how I feel the scale is working by playing the game. But when I play, I do not think about the scale. I just play and think about all passengers and houses as real.

jamespetts

Iluvalar - please be careful to discuss the correct topic in the correct thread: this thread is about the reasons for the other topic being split and locked and the way in which the issue was discussed. The substance of the points are for elsewhere.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

isidoro

Just to be constructive and watch the topic from both sides:

       
  • Iluvalar, although you claim that there may be cultural differences, you have to realize that English isn't the native language of many of us, including you.  It's difficult for us, non-native speakers, to use words just as a proficient surgeon uses the scalpel.  And that can lead to misunderstandings.  Therefore, we must be extremely careful at that.  And we should avoid conversations/topics "in the border" and keep them "in a safe zone"(1).
  • Nontheless, I know that it's very frustrating when you clearly see an idea and somebody doesn't understand it, no matter how hard you try to explain it.  I have felt the same myself long time ago here.  But that is a problem of two: the one that explains, and the one that tries to understand.  No matter how difficult a topic is, with patience and skill, it can be explained to anyone.  Not to mention someone that knows about the subject.
  • Even considering both preceding points, your tone was quite harsh.  You looked like an elephant entering a pottery shop.  As posts passed by, you got more frustrated, you claimed you mastered the subject and asked to be believed, since you saw everything clear and were unable to make you understand.
  • Also consider that this is a project made by volunteers, devoting their time and effort for free.  If attitudes like the above can be seen in a teacher-pupil, boss-employee situation, they are certainly not appropriate in this case (I also believe that they aren't appropriate in the former case too, btw).
If I understood some of your posts, your point is that if James follows his plans, the simulation will have a possibly unwanted butterfly effect ("if a butterfly flaps its wings in Brazil, that can lead to a hurricane in the Pacific Ocean later").  That is, the simulation is very dependent on initial conditions.  A slight variation of those physical initial conditions (even rounding errors) can lead to very different results and I infer that you consider that unrealistic.  That's the technical meaning of chaos, as you used it, isn't it?
__________________
(1) Often I'd desire that English weren't the lingua franca of these days.  But that's how things work for now.  On the back side, English just because of that is thoroughly abused by non-native speakers like us (and some native, I guess  ;D ).  Just if Shakespeare came back to life, and rose his head, and saw what we do with his beloved language, he would surely die again of disappointment!

Isaac Eiland-Hall

Quote from: jamespetts on January 16, 2016, 06:31:56 PM
Or code that **** thing and take a look by yourself. At this point that solution will be the less frustrating for both of us.

THIS is the heart of the problem. Do not tell anyone what to do here. Ever.

Suggestions are welcome, but if suggestions are rejected, then consider dropping it. Perhaps take another try, but if the second try doesn't work, then accept that it's not going to happen.

The game is open-source. If you wish to make a change, do it yourself. That has always been the culture of Simutrans. And even with that, it is amazing to me how many extension requests get done — because fresh ideas and perspectives are welcome. Brow-beating someone who is contributing to Simutrans is not.

The topic of the discussion was not the problem; the method and attitude were. If you cannot understand that, then you will have a problem here, and you will have a problem with me.

Absolutely ZERO of the problem was the topic of the discussion, but rather how you kept pushing it - and indeed are pushing it with the bit I quoted above. THAT is the thing that is not welcome.

Other than that problem, you are most welcome here. :)

Iluvalar

Vladiki, yes but the simulation wont be able to assess if it's exceptional or completely random. In a small city, where the network should only cover 20/300 jobs, some of the building will be simulated as if they were receiving 200/300 jobs just because a few of them landed on the same tile. Even if it would never happen in a larger scale simulation.

Quote from: jamespetts on January 17, 2016, 12:34:08 AM
The substance of the points are for elsewhere.
No, there is no need to hit the same nail again. The moderators said we fought enough lol. What you suggested is easy enough to code and try, I guess that's you next step. Just hope that I'm wrong :) .

Quote from: isidoro on January 17, 2016, 01:32:51 AM
That is, the simulation is very dependent on initial conditions.  A slight variation of those physical initial conditions (even rounding errors) can lead to very different results and I infer that you consider that unrealistic.  That's the technical meaning of chaos, as you used it, isn't it?

Yes, I used the term chaos in the mathematical sens. It's not a matter of considering it realistic or not. But rather if it's stable or not. A player would know perfectly that the mean trip per month on his line should be 80, but would be unable to figure if the trip in the next few month would be 20 or 200. That would cause both gameplay and balance issues.


-------------
Isaac, you got me wrong again. I'm telling James to code what HE suggested and what HE was willing to code before I posted anything.


So you're basically telling me to not tell people to do whatever they want to do no matter what I told them. :)

Vladki

Isaac, sorry for getting off topic. I just could not resist that feeling "ah now I understand what Iluvalar was talking about". Though it is more complicated than I imagine, I'm not going to talk about it here any more.

Iluvalar:  your suggestion really sounded like you tell James to code your proposal.

Anyway I know how you may feel about the mutual misunderstanding. I had similar problem when I proposed this change:
http://forum.simutrans.com/index.php?topic=13097
I hope I did not sound annoying at that time. Maybe the problem was, that I could not explain my thoughts clearly enough. Anyway, I decided that the best way would be to just code it myself. At that time I did not know anything about the code, so I knew it will take me long time, and I had lots of other interests. Then about half a year after that discussion, DrSuperGood came with almost the same proposal, and made it a reality: JIT2
http://forum.simutrans.com/index.php?topic=13789.0

So if you feel that you know better how it should work, and the others seem to not understand you, start coding.

jamespetts

Quote from: Isaac.Eiland-Hall on January 17, 2016, 03:52:59 AM
Quote from: jamespetts on Yesterday at 18:31:56

    Or code that **** thing and take a look by yourself. At this point that solution will be the less frustrating for both of us.


THIS is the heart of the problem. Do not tell anyone what to do here. Ever.

That wasn't a quote from me, I don't think...?
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

Isaac Eiland-Hall

Nope, James, wasn't a quote from you. I accidentally hit MODIFY instead of QUOTE and didn't realize I was doing so; and didn't catch that he'd quoted you until it was too late, and didn't think to change that once I'd realized my screwup.

Either way, this topic is now done. I hope it has accomplished something.