KoA chaos

greenspun.com : LUSENET : MARP Editors : One Thread

As some of you may have noticed, Chris Parsley is blundering through the organisation of the KoA tournament. And it would have been even worse if Gameboy hadn't helped him a little and if I hadn't made some remarks on IRC to Chris every now and then.

If I hadn't mentioned Cicca's Aurail recording being slow, he would never have known. How the hell can he enforce a 97% average speed rule if he has no way of knowing? Anyway... now he's begging me for a copy of the all-in-one Analinp. I think the rule is (or should be!) that only Editors should have access to Analinp and other such tools, so if we were to deviate from that rule, I told him that at the very least, this should be a joint decision by Chad and me, since most of the code in it was written by us.

But I want to know what the other Editors think about it too. At this point, my inclination is that if we were to give him a copy, then it should be a unanimous decision by all Editors.

So... what do you guys think? Or what kind of alternatives are there?

Thanks. Cheers, Ben Jos.

-- Anonymous, September 02, 2000


No, no and no. This KOA tournament had disaster written on it from the start, which is why a number of us chose not to play in it. There is no accountability, nor structure, nor rules, nor anything else needed to run a tournament, partiicularly one handing out prize money. Parsley had a wealth of experience and talent he could have drawn from the MARP pool; instead, he chose the glory route and decided to go it alone on this tourney. So he can damn well do without our software if he doesn't want our help.

As for Analinp, we've all invested some time into it, even if it was just minimal like pointing out games that didn't test properly or submitting sample .inps. I don't claim to have ay ownership over it whatsoever, but I think it is somewhat the property of the Editor's board. And as such, I can't see allowing anyone other than an Editor the right to use it.

If Parsley wants to apply for Editorship, fine, but I think I know how that vote will go already.


-- Anonymous, September 02, 2000

analinp doesn't belong to the editors, it belongs to anyone who honestly wants to analyze what is going on in a game's recording. I don't see any objection to giving it to parsley, other than not giving it to someone who wants to disallow recordings that are less than 97% slow with out using tgmame. That is NOT what analinp or InpHist can do. If we deny him marp tools, we'd be just what he's trying to be "elitist in having his own tournament." I have no major objections nor inclinations to giving analinp to the KOA.

Another minor objection, since none of the editros would be using it on KOA recordings a lot of experience would be neglected in determining what it's output means.

-- Anonymous, September 03, 2000

Welp... here's what I get to say about it:

In Robert's Rules Of Order... the motion belongs to the maker UNTIL the chair mentions the motion, then it belongs to the assembly.

But Robert's Rules Of Order doesn't apply to software - in software - that belongs to the people who programmed it - big time. To my observation - Skito before a certain point, then benjos joined him afterwards.

Now if Skito wanted to give Chris one of his earlier versions that he programmed himself - he could do that w/o the consent of anyone else. Afterwards, benjos would probably have to agree to this - since this software is not available to the public domain.

Now I'm not a lawyer - so that might not be right - but my vote would be summarized like this: Will Chris abuse this program and distribute it to someone else? High chances are probably not. Probably telling him not to distribute it is enough. I know this: He isn't like Stig for sure. Man... imagine Stig running the knockout tournament... Did any of you know that before Pat Laffaye took in the idea of judging coordination Stig wanted to do that? Imagine what that would do to the current tournament...

But in any case - Chris may be having a hard time running the tournament right now - but look back on the first couple tournaments I ran... yuck! I'm sure some people wanted to impeach me outta there after those were all said and done.

Will Chris improve the coordination? I think he will - and as a tournament coordinator myself - I take a responsibility to help him out.

Another point: 18 recordings confirmed in Aurail belonged to myself (which means some form of the software is being used - cause I sure as heck am gonna use that software for that tournament) - and there are ICQ logs that point out him wanting to "hire" me for judging coordination...

Therefore - if I had a vote on this issue - I would have to vote "aye".

Thanks for reading - GB9

-- Anonymous, September 03, 2000

I'm not getting into RRoO here, that's a waste of time. Our chair is rarely here as it is.

Reading gb9's post alarms me for some of what it takes for granted. Chris is a tournament coordinator? No he isn't? The MARP community never said: "Chris, we'd like you to coordinate this tournament." It wasn't even his idea. And in the process of building his tournament, he has shown accountability to no-one else except when he absolutely needed to--something none of us here have the luxury of doing.

If we give Chris analinp, then I urge the rest of my peers here to think through the consequences. Not only will we be releasing a program to someone with 'less than tactful' interpersonal skills (would YOU want an e-mail from C. Parsley that informed you of a slowdown DQ? I'm guessing not.) Chris has been less than an angel in following the actual rules at MARP, flaming me over both the TG issue on Draggonninja (where he had no basis for his argument) and the 1-credit issue (where he was in opposition to the MARP majority.

Finally, if we release analinp to Chris, then we have an obligation to release it to the rest of our confirmers as well--and I don't need to tell you some of the names on that list. Why, you ask? Well, what's the difference between a tourney director who appoints himself, and a confirmer who's been checking scores for the past year at absolutley no personal glory for themselves? Where's the fairness? I would rather release analinp to BBH or bubble than Chris, but if we release to Chris, then I will personally e-mail a copy to every confirmer, including Phil Lamat and lagavulin.

It would be the only honorable thing to do.

All told, I believe the vote stands at 2-2.

Sincerely, Q.T.Quazar

-- Anonymous, September 03, 2000

This is probably going to be a long one from me again, because I'm going to try to address the most important things raised by others, and add some thoughts of my own:

First Chad's remark about Analinp not belonging to the editors: While I agree with this to a certain point, there are also several objections I have:

(1) Knowledge is power. Remember this one, because I'm going to get back to it a few more times. If someone who wanted to cheat had access to the latest version of Analinp, he could start experimenting and try long enough to find a way of cheating that would not be detected by Analinp. He'd also find out about Analinp's limitations (for instance on detecting slowdown on analogue games, and on the inability of it detecting autofire in games that run below 50 fps). He could slow down his computer just enough so that it would not be detected in non-analogue, 60 fps, games. If tg3 is used, he couldn't slow it down too much, but he still can do it somewhat. If another version of MAME is used, there usually is no way of detecting if someone slows the game down by 15% or so, while that still gives a player a considerable advantage. Somebody who knows we have these tools and who doesn't have access to it, nor know about its limitations is more likely to not even try to experiment, because it would become a lot more tedious (with Analinp, he could check immediately, now he would have to wait and see if an Editor confirms or disqualifies it). This is also why I was very much against the fact that the judges in T3 received a copy of Analinp, especially since one of those judges happened to be someone everyone suspects of having used slowdown a lot. And this brings me to the next point:

(2) Numbers. Although the exact figures I am going to use in this example may not be realistic, the general idea still very much is true. Suppose that there is a 5% chance that anybody who owns a copy of Analinp gives it to someone else at some point. To the best of my knowledge, only 5 people own a copy of the latest, all-in-one, version. But it's simple mathematics (for me at least) to see that, now, there already is a 22.6% chance that at least one of us gives someone outside this group a copy. And with every single person who is added to the group, the chances of a new person getting a copy go up, and so also the chances of a copy ending up in the hands of someone who is going to use it to cheat. And that defeats the entire purpose of the tools in the first place. More concretely: OK, you may think that Chris Parsley is not going to give anyone else a copy, but aren't we, at this very moment, considering giving it to him? Why wouldn't he (or anyone of us for that matter) at some point consider giving it to someone else? When 13 people own a copy, the chances will be 50% that, at some point, there will be 14. With 27, those chances are 75%. Either we keep the number of people low, or it will snowball to the point where everybody can get their hands on a copy. Also remember this point about probabilities, because I'll get back to this one too.

Even these two points do not mean I would never send anyone else a copy. But that person will have to "earn" it...

I know that we do not have a hard rule that only Editors have access to the tools, but I think we should, for the exact reasons I gave above. It may sound elitist, but that's not the purpose. The purpose is to keep MARP clean and to detect unclean recordings. As I hope I've pointed out, the more people have access to the tools, the less useful the tools will become, and so we'll slowly get back to square one, where people can freely cheat (only on a more advanced level, though, admittedly) again.

OK. Next QT. For now I'll address only one thing in QT's reply: The argument about testing and submitting sample .inps does not really hold. This would be the same as saying that because you were a beta tester for some Microsoft product, you get to have a say in what Microsoft is going to do with that product. It doesn't work that way. The best you can hope for is that you'll receive a complimentary copy of the release version. Or, closer to home, in a way we are all beta testers for MAME. Does that mean we get to decide on the fate of MAME? Not really, and especially not with someone like Nicola Salmoria running the MAME team.

I'll get back to more of QT's remarks later. For now, I'm more or less giving my objections, not confirming other remarks that I agree with.

Gameboy. Yes, you are correct that if Chad decides to send Chris Parsley an old copy (one from before he used my InpReader, InpHisto, and InpSpeed code), he is free to do so. Well, I think that Zwax had some small contributions to the early versions as well. And, yes, if we are talking about sending Chris Parsley a copy of the latest version, then both Chad and I have veto power over it. But, of course, an old version is less useful. Cicca's slowdown would not have been caught with it, for instance... . Also, if he is to be sent an old version, please don't send a version that shows the additional m35tg3 speed output. Again, knowledge is power. Although I myself had already hacked tg3, seeing that output made me discover some things I didn't know yet. But even if I hadn't hacked m35tg3 yet, the information displayed there would have been enough for me to hack it.

And you may be wrong about him not distributing it. I've already tried to show this earlier. At some point, there may be someone he considers trustworthy and give it to that person. Etc.

OK. Most of the rest that was said, I agree with, and here are some additions of my own:

Yes, Parsley would not know what to do with the InpHisto output that seems so crucial in detecting slowdown. Yes, Parsley is on this ego-trip where he considers himself king of KoA, and where he thinks he can do it all himself. Yes, I silently predicted a lot of chaos before KoA even started, purely based on the fact who ran it, and the fact that neither I nor anyone else with some experience in enforcing rules regarding recordings or running tournaments or being a judge in tournaments was contacted by him for advice. If we don't send him a copy of Analinp, are we being as elitist as he is? I hardly think so. In a way, by his completely ignoring us, he abused us and now he's trying to abuse us again. Is running away from abuse bad? This has nothing to do with elitism, IMO.

Chris Parsley never seems to get the point on anything he disagrees with. I think the following part of a conversation I had with him on IRC is very illustrative:

[21:29:13]  Right, but what about future rounds???
[21:31:16]  Try to stick with games that run in 35tg3, and try 
to get help from an editor.
[21:38:43]  sometimes, they are not the easiest to work 
with, and the games are already pre-selected, going back now and 
re-editing that list will cause chaos.

They are pre-selected, and re-editing the list will cause chaos?!?!?!?! More chaos than you've already created yourself? Telling Cicca that the deadline for him (and only him) on Aurail has been extended, but that he shouldn't tell anyone? Then scolding Cicca in public for doing it anyway? Great move, Parsley. Oh, man, how can you trust someone who runs a tournament that way? Moreover, it shows that he still wants to do it all by himself, and does not want to diminish his "glory" in pulling this tournament off by officially enlisting anyone's help. Editors are not easy to deal with? Hasn't Gameboy already helped him by checking a lot of Aurail recordings? Wasn't I the one who pointed out that Cicca's recording was below regulation speed? (And, yes, I did check all the KoA Aurail submissions myself as well, but decided not to do any confirmations.) Haven't I pointed out on IRC (maybe not in the kindest words, but still...) to him about several things that were wrong? Aren't Gameboy and I editors?

Some more quotes from IRC (I took out some irrelevant parts):

[20:40:23]  What about Cicca? Did you have a look at his .inp, 
[20:40:44]  haven't had the time
[20:45:15]  but let's look at it now
[20:46:35]  yikes!
[20:47:20]  yikes?
[20:47:47]  yeah... cicca's gonna have to be dqed for 
slowdown... that sucks
[20:48:16]  wait just a second, it was close, but let me see 
again to be sure.
[20:48:27]  brb
[20:53:48]  Checked the recording again, where is the 
[20:54:54]  Funny to have a 97% average speed rule and not 
even being able to detect an average speed well below that...
[20:54:57]  I just ran it on a 333mhz, frame rate dropped 
down some times, but the speed remained at or very close to 100%
[20:55:15]  That rule is a guide, not a brick wall, BenJos.
[20:55:17]  benjos... can you please start class?
[20:55:28]  Sure. If I run the recording on my 1 GHz Athlon, 
I'll get 100% constantly. Big deal.
[20:55:36]  That's not what Cicca got.
[20:56:05]  The speed you see is the speed of your machine, 
not the machine it was recorded on.
[20:56:08]  Ok, then, what did Cicca get?
[20:56:34]  I'd guess he played at around 70% average speed.
[20:56:49]  And you know this because?
[20:57:05]  No... the question is: Why don't YOU know this?
[20:59:19]  His average length on the fire button is way below 
what the fastest pressers on MARP are known to do.
[21:00:06]  One and a half times faster than the fastest I've 
seen Hisa-Chan do, for instance.
[21:00:29]  There is no need for autofire on this one 
BenJos, as the game does it for you
[21:00:44]  I'm not saying he used autofire.
[21:01:10]  He still pressed the fire button 1424 times.
[21:03:28]  Since I have to verify recordings for the entire 
tourney, care to share?
[21:03:44]  that's up to benjos
[21:03:57]  and I also help you out as well :)
[21:04:02]  Because I do still have final say on recordings 
in KoA, and unless someone can show me some clear item, Cicca remains 
for now.
[21:04:38]  Heh. Then just let it remain. I don't care.

That's enough. Lots of things in here. Several examples of how he always fails to get the point completely. Such as that there is no need to use autofire, because you can just keep the firebutton down all the time in that game. What does that matter? I just told him that Cicca pressed it at an incredible rate. Or the speed thing, of course, saying it plays at almost 100% constantly on his own machine.

And he completely ignores Gameboy's offer to help. And he pretty much "threatens" that if he doesn't get a copy, that Cicca's recording will stand. Again, he will not take any help or advice, but instead run this all by himself. He clearly has given no thought to how to actually enforce a lot of his rules. He is not willing to change some things because others advice him to, but he has no problem with changing things if it was his idea to change them. He sets the 97% average speed rule because it is TG (wrong, Chris, it's NOT TG), then says it's not a brick-wall rule. He postponed the start of the tournament first, in the hope that Mark would create an m36tg3. When that didn't came by the date he had postponed the start to, he started the tournament anyway. He decided on MAME36 for Aurail, but both Mark and I told him that wasn't very smart because of the possibility of re-recording. So then he quickly changed it to win36 and literally stated on his page "to avoid re-recording". Very smart move. Give people a suggestion on how to cheat. So he quickly deleted it when I told him that. But he didn't stick with win36 anyway. He allowed MAME36 as well?!?!?! As we all know, changing rules during a tournament invites chaos. He's done it several times already.

Well, that about sums it up for me. Yes, I very much agree with QT. This tournament has had "disaster" written all over it from well before it started. And Analinp is not some tool that is going to make the entire tournament magically run smoothly all of a sudden. I don't think Analinp (even if he knew how to use it) is going to solve anything. Parsley is very stubborn and is not willing to listen to anything once he has made up his mind about something. He has a habit of missing the point completely. And a lot of what has gone on so far already is not related, or not directly related anyway, to Analinp. It's about his behaviour so far in dealing with things. That certainly is not going to change by sending him a copy. Sending him a copy of Analinp is NOT going to stop him from blundering a lot more. Pat and Gameboy have learnt a lot about how to run a tournament. They learnt it the hard way. But they were very much willing to ask others for help. Did Parsley even try to get advice from more experienced people? Despite all that, Gameboy has more than once said he was willing to help Parsley, and, hell, even I'll help a little. But that's as far as I go.

I wanted to hear some opinions and arguments before I gave mine, because they might even make me change my mind. Other than being nice and helping someone who doesn't know what he's doing, I haven't seen any good arguments. And Parsley doesn't want help. Or only on his terms. The only help he wants is a copy of Analinp. He is king and no one is going to touch him. Well, fine... fly high, fall far...

This has gotten too long, and I haven't addressed a few things that I meant to address. That's OK, because they weren't directly related to all this, so it's better if I do it in a separate thread anyway.

In a nutshell: If Parsley wants some help, I'll help him, but as for sending him a copy of a version of Analinp that contains any code written by me, I veto it. He hasn't earnt it.

Cheers, Ben Jos.

-- Anonymous, September 03, 2000

Hmm... my IRC quotes didn't come out too well. The identities of the speakers have disappeared. Must be because the angle brackets (for instance: <BenJos>) surrounding them. For anyone not receiving my previous post by e-mail, I'll be happy to forward a copy.

Cheers, Ben Jos.

-- Anonymous, September 03, 2000

Oops. Sorry. QT's 2nd post in this thread must have been submitted while I was writing mine and I didn't see it until I sent mine. And QT's post reminded me of something. Something I should apologise for. QT said he didn't mind if BBH or Bubble got a copy. Bubble actually has a copy. I sent it to him. Without consulting anyone else. That was wrong. I sent it to him to help him in improving his very useful MarpFE program, but it was still wrong for me to do so without at least talking to Chad about it. I apologise.

I guess I can't even start yelling now if someone decides to send Chris Parsley a copy without mentioning it, but I sincerely hope that it won't happen, and I also hope I haven't given people any ideas now. :-)

Again, my apologies.

Cheers, Ben Jos.

-- Anonymous, September 03, 2000

Hmm... just re-read my post again. Lots of spelling/grammar/whatever errors. Ah well, too tired to correct them, and you'll get the point anyway.

Cheers, Ben Jos.

-- Anonymous, September 03, 2000

From experience, I know what Parsley is going though and it's not pretty. However, I'd also say, when I've needed help on whatever it may be, all of you guys, come though in a pinch and really make it count. With Parsley he totally screwed me over in T1, he was full of false promises about confirmations, deadlines, etc. I ended up having to look at and confirm most of his INPs. Later we bickered over ssprint, where he was wrong, and then didn't even bother to play. I've had it with his shit and won't help him unless it's a trivial matter. I don't think we owe him anything, but if he is so damn persistent then I'd say give him an old version of InpSpeed and be done with it. I don't think it will work with Mame36, but if you want to make the mods that's up to you. There's some risk in distributing the EXE, but source code, NEVER!!!

-- Anonymous, September 03, 2000

I did say anyone who honestly wants to use analinp to see what is in a recording, that doesn't include someone who would use it to disguise their inputs to fool analinp. And obviously if we let it out into the wrong hands (or any hands since it is bound to mathematically leak at one point or another) it won't be used honestly.

I abstain from this vote, since i don't feel comfortable voting either way.

-- Anonymous, September 04, 2000

Moderation questions? read the FAQ