Embottered

May. 11th, 2007 08:51 pm
peterbirks: (Default)
[personal profile] peterbirks
Well, the 2+2 debate rages on. The interesting thing here is that I mentioned the "faux bot" concept quite a few months ago, and now 2+2ers are struggling with the philosophical problems involved. I kind of like my "reverse Turing test", and, if it's true that these are bots with a human supervisor some of the time, then you are entering weird territory.

Clearly the defenders are claiming that it's a sweatshop, with loads of poorly paid humans following a template, with a "supervisor" coming in to add a higher level of play should the situation demand it. The reason the defenders are claiming this is that it is probably just on the right side of legal, if not moral. This is one reason to doubt this explanation. I still like the original bot system theory, with a supervisor moving in to play as a human when the situation demands.

One point here that separates it from the standard bot-script donkeys is that we probably have one or two players here who can really play, but who are earning more by running less than efficient bots over many many tables ("a more productive use of time", so to speak).

I've prejudices against sites that allow data-mining. But this creates a problem for me, because it seems clear that the only way to statistically prove that bots are playing is to use datamining. If you could only gather data when you were sitting at the table, you could just not gain the statistical proof.

However, this did lead me to look at the "overcards" site and to take note of what many 2+2ers are probably doing, and I wonder how far this pushes things over the line. Autohotkey, automatically sitting out when it gets to a certain number of players, and various other "automating" factors make it possible to play 8 to 12 tables, and 2+2ers see nothing wrong with this, even though a number of their decisions are taken "automatically".

So, you can see where I am heading. Although I have my own "line in the sand" about what I think is acceptable, and another (slightly further out) line in the sand about what facilities I will use (for example, at higher stakes, I would use datamining if it was available, while at lower stakes I might stretch the line to certain scripts such as autohotkey, even though I think that both are against the spirit of online poker, and I would prefer it if they were banned).

All of it is very difficult. I think that pokertracker is fine, but perhaps this is because I have used it for so long. But is a stat overlay on the table "fine"? Here I am less sure, although I use GameTime all the time. Where do I draw my own line and where do I draw the bot line?

In a way, it's not hard. Any script that makes an "action" during a hand is cheating, I feel, even if that script is "fold 7-2".

I also feel that any script that automatically decides whether or not you are dealt in is cheating. To play or not to play is something which should be decided by a human, and to argue that "I never play with five players or fewer, so automating it is fine" is so close to "I never play 7-2, so automating it is fine" as to make little difference. You are, as it were, sitting at a table, and saying "deal me out". That constitutes part of playing poker, I feel, so to automate it is beyond the pale.

How about automated waiting list requests? Well, that's a tough one. I really can't decide.

+++++++++++

Some bad times with the work computer today. I was happily typing away, ignoring the Windows Update notification, when an urgent e-mail was sent out from work stating "DO NOT INSTALL THE UPDATE".

Apparently it was causing a conflict with some of the company's wide-ranging software shit, installed by overkeen IT geeks who are not happy unless they can fuck up your life. And this was causing all Microsoft Office programs to freeze.


Anyhoo, I thought no more about this, apart from congratulating myself for not installing the update, and, after sending out the newsletters, I went shopping.

Upon my return, I discovered that the updates had been installed automatically. Jesus shit, my computer never used to do that. And, guess what, Excel promptly froze on me. Indeed, one of my poker files was corrupted almost beyond repair.

So, I rebuilt the file, and then it froze on me again, although not corrupting itself.

Hmm, what to do?

Well, "System Restore" is a favourite of IT geeks, and, impersonating said geeks, that's what I went for.

The problem here was that the updates still installed themselves automatically. Fuck.

So, I went to the updates part of the control panel, only to find it all greyed out.

Yes, the IT people had set it so that we mere users could not alter the update settings. Great.

So, this was a serious problem if I wanted to do any work in Excel or Word over the weekend on the office machine (i.e., the machine I have at home that connects to the Virtual Private Network).

Serious times call for serious measures, I decided, and off I trotted into RegEdit. There's a couple of settings in here that dictate whether the Windows Update is greyed out or not, that allow you to bypass the administrator requirements. A quick alteration, and back I was in the main menu.

Smack in System Restore again to get rid of the updates, and I should be fine.

I think you can see the flaw in this line. Not only did the Restore get rid of the Windows Update, but it also got rid of my change to the Registry Settings. Whoops.

So, what I had to do was, System Restore back to before the Updates, then change the Registry settings before the update kicked in automatically, then turn the machine off again, reboot, and, voila, I'm back in pre-fuck-up state, without automatic updating enabled.

I actually had a lot of fun looking at some other "Run" commands that change this whole updating scenario from within the Registry. Windows Updates are the spawn of the devil, but the problem would never have arisen if our own IT people hadn't been so keen on controlling everything that users do, down to the air we breathe. And, on top of that, IF we had up-to-date versions of Office, rather than Excel-fucking-2002, the situation would not have arisen either.

Right, let's do some ironing.

Turing or not Turing?

Date: 2007-05-12 08:55 am (UTC)
From: [identity profile] real-aardvark.livejournal.com
I don't know what "the spirit of on-line poker" might be. I'd always thought it was a cruise ship off the Bahamas for elderly American Jews, but perhaps I've been misled. (Or drunk too much coffee in this Internet cafe. I drink a of coffee when it's free.)

In theory, there is no moral case to be made against at least a minimalist Turing approach to poker bots. I'm not sure there's a moral case to be made against any form of bottism, although it obviously tends to take the fun out of things.

However, if I'm a highly experienced player who is good enough to make, say, $10 an hour on twelve tables, then presumably the only limit to my earnings is my sheer inability to move up to a higher number of tables. Since, presumably, there is a high percentage of positions where my actions are automatic, I don't see a moral issue with actually automating them. All this is doing is taking away the drudge work; whether I'm physically there or represented by my little bot-pal makes no difference. Once a table hits a position that isn't covered in the "scripts" (and they're presumably more sophisticated than mere templates), it pops up inn front of me and I make an appropriate decision. The limit of tables I can play is thus increased, but they're still being played in exactly the same way as they would be, were I to play without bot-wizards.

One might object that I am only playing more tables because of artificial assistance, and that not everybody has this facility. I would counter-argue that there are players out there who can play that many tables simply because of better focus, better multi-tasking memory, or just sheer mania. Why should they benefit from these (arguably non-poker related) abilities?

In the extreme case, I would just make my bot software Open Source, and then everybody plays the same way, if they wish. Since I make $10 per hour per table, this is probably even more beneficial to me -- more players playing more tables using my bot, and my own multiplier is still there ...

-------------------------------------

Loved the end-around on MS sysadmins by fiddling with the Registry. They're not geeks, btw, they're idiots. Why the Registry wasn't locked down is beyond me. These people do not have a Deep Understanding of how Windows works (although they clearly have a Deep Understanding of how to fuck it up).

The whole idea of removing Admin privileges on somebody's "personal" -- ho ho -- computer, on a mandatory basis, is anathema to me. It's clearly beneficial to a certain type of user. It's equally clearly non-beneficial to the likes of you or me. I can't even get the calendar up off the system tray date icon, because administrators in Stockholm have decreed that I'm not allowed to change the date. So I can't even read the bloody thing.

What should happen is one of two things:
(1) If you're not competent to use a computer, it should be removed from you. This is extreme, but would save companies money in the long run. No need for system administrators at all ...
(2) The default is a locked-down computer. However, anyone who asks is given full administrative privileges. Possibly, a certain amount of cock-ups a month would result in these being revoked, although I'm not convinced even this is necessary.

It's all to do with ownership, of course. I wish people would just do their damn job, nothing more and nothing less.

Maybe with the advent of virtual machines, we'll see all this changing. But I doubt it.

Re: Turing or not Turing?

Date: 2007-05-12 08:04 pm (UTC)
From: [identity profile] peterbirks.livejournal.com
As I wrote, it's an interesting philopsophical area. However, I don't think that it's unusual that computer people (particularly those not involved in poker of any kind) "don't see anything wrong with it". while poker people do.

You make a long list here, so I'll just pick up on a few points.

1) Since, presumably, there is a high percentage of positions where my actions are automatic, I don't see a moral issue with actually automating them.

The flaw in this line is that, even though the actions are "automatic", when you get tired, you often make mistakes. The computer does not make those mistakes. If you get a bad beat, you might tilt. The computer does not. It's like a Formula One driver saying "I always take this line, so there's nothing wrong with automating it". The fact is, knowing what you should do, and consistently doing it, are two very different things.

2) Why should they benefit from these (arguably non-poker related) abilities?

Non poker-players often get confused about "poker abilities" and what constitutes them. Being a winning poker player goes far beyond knowing the right thing to do in a hand and being able to read your opponent. "Focus" and "multi-tasking ability", as well as a cool head that doesn't get easily tilted, are very much poker-playing abilities. No-one questions that fitness and stamina are part of football-playing ability, so why should a sport not so obviously physical be seen differently?

Now, I have made my view clear. I don't mind playing bots. While they don't get tired or go on tilt, they also do not react to your style in the way that humans do. All I want is that it is clear when you are playing a computer and when you are playing a human.

PJ

Re: Turing or not Turing?

Date: 2007-05-13 07:18 pm (UTC)
From: (Anonymous)
Gee, nice to know that I'm a "computer person." Perhaps I should consider "computer dating." Only the latest multi-core, please. And I want the casing blonde, with a big nose.

Actually, I was trying to argue from your original premise: that it is an interesting philosophical area. Which it is. Obviously people -- not you -- are going to focus on the technology involved, which I think is a red herring (and an underdeveloped red herring at that -- wait 'til Google gets involved, or indeed any Bayesian algorithm). I would regard the technology as happenstance -- perhaps 100% accurate, perhaps 95% accurate, and who knows at what point it distorts the market, at which point we get into economics and finance. Bugger all that. This is truly an interesting philosophical question, and it ranges far beyond poker, online or otherwise.

I think I agree, substansively, with your point number (1). (O Lord, now I'm turning into a "Powerpoint person.") Essentially, I was arguing as the Devil's Advocate, here. I'm well aware that human factors, such as (for example) tiredness, lapses in concentration, or indeed sheer bloody-mindedness, are a part of playing poker. On-line poker may well be the "better," whatever that means, for incorporating these factors. I would argue, equally, that if you want these factors to play a part, you should only play face to face. (And probably in a Faraday Cage, these days.) This is on-line poker. On-line means "with computers." It's an entirely different,er, ball-park.

But, having said that, I basically agree with point (1). (Aargh! Mummy! Powerpoint-San is coming to get me!) Unfortunately, I don't think there's a hell of a lot you can do about it. (Although I'd be interested to hear theories...)

I don't really accept your point (2), I'm afraid. It would seem fairly clear that there is a difference between playing poker online and playing soccer at, say, Old Trafford. In fact, having put it that way, I can think of five major differences immediately. Although I'd be deeply interested to watch a penalty shoot-out between the cream of European footballers and five random online poker players, all of whom have stayed up for 60 hours straight ...

Damn, that's six major differences, and poker players win hands down on the last one.

The byte stops here

Date: 2007-05-12 09:39 am (UTC)
From: (Anonymous)
Working from home has the advantage that I'm in charge of my own computer -- though it also means that I had to buy it with my own money.

I don't have a completely free choice of software: there are some programs I need to install in order to do my job. But at least the system settings are my own business. And so far I obstinately continue to use Mozilla Thunderbird for e-mail, even though most of the people I work with are using MS Outlook.

I don't have MS Office installed on this computer, though I still have Office 2003 on my previous computer in case of need.

I find that OpenOffice seems to do a reasonably good job of handling the Excel spreadsheets I come across (though I hate spreadsheets and wish people wouldn't use them for text tables), but it can't be relied on to handle complicated Word documents correctly: I've noticed some problems with tables.

However, if you just want to read incoming documents, Microsoft provides free MS Office viewer programs.

-- Jonathan

Re: The byte stops here

Date: 2007-05-12 08:05 pm (UTC)
From: [identity profile] peterbirks.livejournal.com
I agree with you about the use of Excel for text tables. That's almost as horrible as the use of tables in Word (and, indeed, in HTML so that people can dictate the width of the columns -- donkeys.)

PJ

Re: The byte stops here

Date: 2007-05-13 07:44 pm (UTC)
From: (Anonymous)
Yes, all right, stop sniggering, she's multi-core with a heart of gold and nicely dyed auburn hair.

Nobody sensible has used tables in Word since around '98 or so, if not before then. I have periodically tried to make these things work, at least approximately, and then given up and written the relevant table in Excel and imported it via OLE2 (or whatever the invisible mechanism is these days). Not to be too much of a computer weenie, but for specific purposes, I'd rather play around with PDF, which is supported rather well in Python and therefore can be embedded into C, C++, Java or (I think) .NET. On the other hand, if the document is only aimed at <100 people, and will only go through <10 revisions, and there is no requirement to keep >10 copies of each version, then embedding Excel in Word is fine. You do at least get some control.

Dictating width, or anything else of consequence, in HTML is an absurd concept. I won't even mention CSS. (Oops.) It's a mark-up language, for God's sake. The damn viewer works out the details. It's also a bloody awful and incompetent mark-up language (vide comments on PDF, above), which is why people insist on grotty short-cuts to make it "look right." (Oops - CSS.)

Wait until XML and its bastard spawn really get going. At the moment, they're just the silver bullet for "interoperability." Which is absurd, and a demonstrable disaster. Wait until XML hits typesetting and presentation in a big way... which will almost certainly happen somewhere inside the shit-swill that is Web 2.0.

Me? I don't care. I'm a Philosopher, baby.

The tabular blues

Date: 2007-05-14 07:40 am (UTC)
From: (Anonymous)
Well, I wouldn't choose to use MS Office at all; but, if I have to use it, I'd rather do tables in Word than in Excel, and I wish other people would too. I haven't had any particular problem doing tables in Word, and they're somewhat easier to edit than in Excel.

My main gripe about Excel is that I keep getting sent Excel spreadsheets about twenty columns wide, each column containing so much text that I can see only a few columns on the screen. The horizontal scrolling drives me crazy, and there's no way I can print it out on A4 paper. The people at HP who make these things all have access to Designjet printers and can print on enormous rolls of paper five feet wide if they want to. No consideration for others.

Using Word rather than Excel at least forces you to give some thought to the idea of page width.

Now you mention it, an HTML table would be somewhat handier than a spreadsheet. At least the columns would adjust themselves to the page width available -- unless, as you say, some idiot has fixed the column width.

-- Jonathan

August 2023

S M T W T F S
  12345
6789101112
13 14151617 1819
20 212223242526
27282930 31  

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Feb. 24th, 2026 05:43 pm
Powered by Dreamwidth Studios