One last time
Yes, before you start, this is all unbelievably tedious. Again I'm posting up comments that should have remained just that, but as it happens there is a liberal (cough) loose with a penchant for redacting anything that calls other people's attention to her haughtiness.
As a side story, a musicologist with a loose pen and sloppy thinking then starts going off on one thinking I am someone else. The back story isn't really worth recounting. Fill in the pieces for yourself.
All that said, I take offence to people taking offence and deleting my comments without even trying to engage (at any level) in a conversation. So here is the comment thread from another post lovingly reconstructed for your delectation.
Comments
Eva, you often repeat that you distrust the Hungarian branch of Gallup ("Gallup is occasionally off by ten points."), but you have to learn to revise your opinions based on evidence.
You last repeated this statement in the run up to the referendum when you commented:
"Médián is talking about 46%, Gallup 54%, Marketing Central 33% and Publicus, whoever they are, 68%. (At every election we have, I suspect, a bogus polling company. This time it seems to be called Publicus. I have an idea whose brainchild this was.) As for Gallup, Gallup of Hungary is usually so far off that it is not worth taking them seriously. It's a shame that Hungarian Gallup is tarnishing the name of this oldest American polling company." (your post from March 7th, 2008)
The final turnout was 50.48%, which made the Gallup prediction the closest.
You don't seem to have revised your opinion or knowledge base based on plain stark fact.
Posted by: dinayekapelye | May 21, 2008 at 12:46 AM
Dinayekapelye,
You can't refute "occasionally" with one instance where Gallup were on the money.
Do have evidence that Gallup are never or rarely 10% off?
Or more interestingly, that Eva's preferred pollsters "(Századvég, Médián, Marketing Centrum, Tárki, Szonda Ipsos)" have similar margins of error?
Unless you have the evidence or the argument to support your criticisms, please don't waste our time.
Posted by: Adrian | May 21, 2008 at 01:57 AM
Adrian,
Your understanding of scientific knowledge and progress is unfortunately based on the very outdated (but widely used) model of monotonic logic and Popper's empirical falsificationist approach.
Having said that Eva's reasoning is based on the approach Popper worked so hard to prove misguided, that of the justificationist approach (dependent on either panrationalism which Eva favours or irrationalism).
Your logic is neither here nor there. I merely pointed out that one needs to revise one's opinions based on evidence or at least be more cautious, or better still clearly acknowledge one's bias.
And to answer your next question the model of reasoning you should be using if that of non-monotonic models such as Bayesian logic/probability.
For the record, I didn't refute or attempt to refute anything, I merely drew attention of an existing dataset that could help in revising opinions based on probabilities.
That said, it would be great to have the data and statistics as you outlined. My feeling is (to revert to abductive reasoning) that there would be little or no statistical difference between the main polling companies.
Posted by: dinayekapelye | May 21, 2008 at 02:35 AM
[Redacted here: comments on another discussion thread removed]
Water polo, anyone?
Posted by: Dumneazu |May 21, 2008 at 03:00 AM
Dumneazu,
As you appear interested in exploring this issue further I can recommend "I was a Teenage Logical Positivist (Now a Septuagenarian Radical Probabilist)"
Dick Jeffrey's PSA Presidential Address (1998) link http://www.princeton.edu/~bayesway/KC.tex.pdf
Though in your case that would be "... quinquagenarian..." - well, perhaps not. I think the teenage logical positivist is still dominant.
Posted by: dinayekapelye | May 21, 2008 at 04:21 AM
Hey, [Redacted], I'm not saying you are stupid, just extremely unethical, as befits a libertarian. Heck, a 3.50 in Economics is nothing to sniff at, is it?
But to the bemused spectators out there, V-who-can't-be-named seems to think it is fun to post his latest comments using the name of my band (the Hungarian klezmer band Di Naye Kapelye) with a hyper link to my photo from my band's web page. (I guess he gets a kick out of "outing" a Yiddish folklorist. Who would have thunk!)But then, all publicity is good publicity! Buy the CD, [Redacted]!
If anybody is interested in who V/[Redacted] is in the real world, I'm happy to divulge.
Posted by: Dumneazu | May 21, 2008 at 04:47 AM
Dinayekapelye,
"I merely pointed out that one needs to revise one's opinions based on evidence or at least be more cautious, or better still clearly acknowledge one's bias."
It's hard to imagine how Eva could be more cautious than by qualifying her point about Gallup than with "occasionally".
What do you mean by 'bias' here? The statistical meaning - she's a historian, not a statistician or a philospoher of science: or the ordinary one, in which case her bias, after the many thousands of words she has written is plain enough to anybody who has bothered to read them, and doesn't need any further comment. Why this insistance that she be bias free?
The rest I didn't understand, but I don't think you wanted me to understand, I think you just wanted to awe me with a display of technical langauge. Sad.
Posted by: Adrian | May 21, 2008 at 06:46 AM
[Redacted here: comments on another discussion thread removed]
Adrian,
It is not my fault you cannot follow a conversation. I have pitched my argument at the level of a reasonably intelligent postgraduate. This is not a blog for dullards. I make no excuses for your educational level, that is not my problem. You are simply wrong to assert that I don't want you to understand, I would love nothing more than to have a conversation reciprocated in kind. If you insist on assuming that I or anyone else has not got passed a 101 course in statistics and probabilities (which your reply inferred) then what do you expect?
The words/concepts/phrases you probably had difficulty with were: monotonic logic, Popper, empirical falsificationist, justificationist, panrationalism, irrationalism, non-monotonic, Bayesian logic/probability, abductive reasoning. You can simply look these up, if you don't know what they are. It would take chapters if not books to write my replies if I couldn't assume (which I do) that we are all ridiculously over-educated.
You are quite right Eva uses the hedging statement "occasionally", she also adds "in my opinion", "I personally" another hedging statement ("...some are better than others. I personally find Gallup the least reliable"). The statement from the 7th March goes further though and is marked with fewer hedging statements ("Gallup of Hungary is usually so far off that it is not worth taking them seriously. It's a shame that Hungarian Gallup is tarnishing the name of this oldest American polling company"). So here we also have the words "usually" as a qualifier.
In linguistics, particulary in the field of pragmatics hedging statements are studied in detail. What is worth noting is that these adjectives, adverbs, subclauses act along Gricean principles (search under the term 'Gricean Maxims' to learn the basics) to give additional information about the veracity, certainty, verisimilitude, completeness etc... of an utterance.
But these qualifying statements work at more than one level, particulary as part of a larger and ongoing communicative act (such as an online web log). Whilst Eva is making these statements she is also sure that you and everybody knows who she is and her academic background. As such, most individuals are less inclined to call to question the implicit assumptions she encourages you to make n.b. I do not assert by this that she either knows or understands the nature and manner in which she communicates this information, she is after all a historian, not a linguist, psychologist, logician or orator. For her, after years in academia, she simple presents information and ignores contrary information/statements.
So yes, Eva makes it clear that these are her personal opinions, that is and was not in doubt. Even though she claims that Gallup Hungary is 'tarnishing the good name' of Gallup in general it is a personal opinion. If you are happy with that, then fine, it's hardly worth taking further.
Might you concede in return that one might conceivably revise one's opinions in line with experience and evidence to the contrary? Or are we all doomed to stick to our beliefs no matter what? Worse still not acknowledge or answer for our errors of fact or judgement. As it happens historians belief systems, line of reasoning, use of facts and logic have to obey the same principles as everyone else's, even if their methods differ. You are simply restating panrationalism.
Dumneazu,
You are mistaken. I'm afraid your anger clouded your judgement somewhat. I am not who you assume I am.
I am not your friend Bufo bufo. What is more I have never met any of the established Pestiside or expat crowd. Although I know who you all are etc... It doesn't take much if you spend most of your time hacking (in its positive sense, not destructively) away on computers. So please don't bother to divulge any information for my gain.
A simple look at the content and style of my writing would have told you that I was not who you thought. There are plenty of lexical analysis tools freely available that would statistically confirm this. Our (mine and Bufo bufo's) idiolect are markedly different in structure and vocabularly.
And I presume you know who I am, even though we haven't met. I have never tried to hide it and it is completely obvious to anyone who isn't a complete 21st Century dimwit.
As you remarked on my "affected erudition" and basically accused me of proving by my comments that my level of discourse was of the same content and quality as the worst of Fidesz and Orbán without showing a willingness (or ability) to respond to the actual content of what I wrote, then I thought it amusing. Obviously you didn't.
It had nothing to do with your ethicity, caste, creed, religion or otherwise either - which you again inferred. You couldn't be more wrong on that point. As it happens I already know your music and would also recommend the CD.
I'm interested that you called what I did 'unethical' though. I wish you would say more. I am not familar with the model of ethics you refer to, so presume it is a home grown variety, as such I need to know how linking to your webpage constitutes a breech of ethical contract (assuming we had a such a contract).
The only theorist I can think you must be referring to is that of Emmanuel Levinas' ethical hermeneutics. For the benefit of Adrian: Levinas argues that the pursuit of knowledge is secondary to our 'ethical duty to the other'. Dumneazu, am I on the right track, or was this just a flippant remark meant to arouse sympathy from the audience and direct revulsion against me? Booo, sisss
"- I foresee, Mr Deasy said, that you will not remain here very long at this work. You were not born to be a teacher, I think. Perhaps I am wrong. - A learner rather, Stephen said. And here what will you learn more? Mr Deasy shook his head. - Who knows? he said. To learn one must be humble. But life is a great teacher."
- 'Ulysses' by J.J. p.41
Oh, there I go again. It's a kind of Tourette's. Or may be I'm just a Szent fazék, after all.
n.b. It happens that dispite Eva's claim to the contrary my posts were blocked by her comment filter. That is because I mentioned Bufo bufo by his nom-de-plume. This effectively swallowed my posts. Eva's claims to the contray are simply wrong, though it was a side effect of her campaign to redact comments she found unfavourable.
p.s. Additional words/concepts/phrases to look up (for the benefit of Adrian): hedging statements, Gricean Maxims, pragmatics, hermeneutics, bufo bufo, lexical analysis, idiolect
Posted by: YouKnowWho | May 21, 2008 at 11:40 PM
YouKnowWho. Your last comment is deleted. I don't allow insulting comments to appear on this blog.
Posted by: Eva S. Balogh | May 21, 2008 at 12:44 PM
YouKnowYou
Thanks for dumbing things down for me. I managed to understand nearly everything you wrote.
I'm familiar with Grice having studied him both in Philosophy as an undergraduate, and in Linguistics as a postgraduate. Could I draw your intention to his conversational maxims - especially of quantity - and suggest you try to use them in your future posts.
P.S. I enjoyed the quote form Ulysses.
Posted by: Adrian | May 21, 2008 at 02:30 PM
A comment on comments
Comments on my blog, these are currently turned off. This is simply because I am messing about with a server-database cluster and am having timing and lock problems with complex database queries. I have this almost sussed, but until it is streamlined and working to perfection (with the help of hibernate and memcached) I will be keeping the service threadbare. Once everything is fully tuned comments will be back on.
A Tudás Fájdalom ('knowledge is a pain')
Knowledge is a pain Power
Oh, yes, knowledge is power ('a tudás hatalom'). It's also a pain. It's a pain, because you have to wield power responsibly. And let's face it, all the alternatives are more fun. So for about 20 minutes this morning I was put in a quandary. Do I use my knowledge responsibly, or just have fun? Now, in everyday life I don't have a choice, my conscience and upbringing reign me in, but I've got a suspicion that given enough space and anonymity I'd be a Jack type (from Lord of the Flies), perhaps even like Roger.
Normally, normally I just don't have to worry about what shape my ID takes, but today it just went haywire, because some fool has placed me in two categories of 'The Satin Pajama Awards' over at 'A Fistful of Euros'.
I then had a problem. I knew I could win. It would be so, so easy. I could trounce the opposition with just a few bits of Java code and a list of anonymous IPs.
OK, so how was I going to win? Well, the voting system relies on collecting enough information about each voter to make sure they don't vote more than once in each category. The information that the poll tries to collect is...
Information people collect
1. Your IP (the web address that you are viewing the internet from)2. Cookie information (those hidden things that website use to remember who you are)
3. Computer information (if you don't have a firewall, you may be transmitting lots of information about yourself to the world, try Zone Alarm for a good free firewall).
So, to get round the voting system all I have to do is make sure that they don't get this information, or if they do I quickly get rid of it and vote again. So here is my step by step guide to internet
(If you'd prefer not to know, or you come out in spots at anything vaguely technical you should skip to the end bit "Where is this all going?")
Simple Anonymiser
1. You get access to the internet via an ISP. Some ISPs have multiple addresses. When you sign on to use the internet you are assigned one of these addresses or IPs. You need this IP address to that servers know where to send back the information you request when you click on a link in a web browser.So, just by logging off and back on again, you may be given a new IP address by your ISP. You can check what your IP is by going to What is my IP. So check this before you log off and after and if they are different then you have solved the first problem.
Better still, if you have a laptop and wireless access, then roam around signing up to various services available. Make sure you do step 2 and 3 in between each hook-up and you can vote at will.
2. You can clear your cookie cache by looking in the Tools->Options menu on your browser. If you know how to delete just the cookie set by 'a Fistful of Euros', otherwise you may lose some important information, or have to remember passwords/usernames for sites.
3. Make sure that your computer is not visible to the outside world. If this doesn't work (which it does by the way) then you need to step up a gear.
Medium Strength Anonymiser
1. You have two options, use a web-based service or downloadable program that integrates with your browser. So, for the first option, using a web based anonymiser service. With these the website you visit will be the web address that is seen by your real target. You request a web page through the anonymous proxy, which is why it's called a proxy. There are hundreds of these sites, so try a few (dozen).Otherwise you can use the desktop versions JAP short for Java Anonymous Proxy. This is a program that integrates with your browser and redirects all your web requests via a choice of servers. Both of these work, I've tried them (voting on a few underdogs). I'd also note that in using some of these proxies they were showing me that I had already voted! There is only one explanation for this, other people nominated in the poll are using these services to bolster their votes.
2 and 3. Just make sure you take the same precautions as in the Simple anonymiser.
High Strength Anonymiser
There are hundreds, if not thousands of open and perhaps insecure servers around the world that will let you direct your traffic through their server. Most of these probably don't know that you are doing this, but as long as what you are requesting is legal you are probably safe (though using anyone elses property without their consent is clearly theft and would be seen as such in most countries).When you vote, your choice is sent to the website as either a POST or GET request. Your voting preference is attached to your request. With GET requests the information is visible in the browser address bar. With POST requests the voting information is hidden. This is why most people use POST requests, it stops people seeing sensitive information too easily.
So, to vote, you don't even have to go to the website, you can just simulate voting by mimicking the web form request. First you have to look at the source of the webpage you want to emulate. This tells us the information we need to know to fake a real request. What we need to know is what is the form name, and what parameters do we need to attach to make our request. In this case the form name is given as <form method="post" action="afoeawards.php">So the webpage for voting is "afoeawards.php". Then we need the parameters to attach our vote. For option 10 of the first poll this would be:
<form method="post" action="afoeawards.php">. So we need to convert that into a request. First add the website address (the URL), which is "http://fistfulofeuros.net/". Add the webform "afoeawards.php". The add each of the parameters to this address, which is "?" followed by each parameter, followed by "&". The parameters for option 10 of the first poll are: "option_id=10", "action=vote" and "poll_ident=15".
Putting that all together, a crafted vote request would look something like this:http://fistfulofeuros.net/afoeawards.php?option_id=10&action=vote&poll_ident=15
But that still isn't enough, because if you just submit this, the server sees this as a GET request and it only accepts POST requests. To convert this into a POST request you need to know a bitabout programming and there just isn't space or time here to take you over that hurdle. Search on the internet for "simulating POST requests" and there is tons of information. There's bound to be something in your favorite programing language. Or perhaps some other tools you can use.
If you have Java, then you could just download JMeter to set up the request using this. Using Jmeter, you can also randomise the time of your requests to make them harder to detect. You could even set this up on a number of machines so that you have a whole army of drones voting for you.
Undetectable Anonymiser
There are many, many more ways, but this is my secret. Knowledge is power and I just don't trust you. Also, it's irresponsible to put knowledge into the minds of people that may not use it wisely or legally. It's taken me more than 10 solid years to become a programmer and gain my knowledge and in that time I've realised how fragile the internet democracy is. Only a fool would undermine that democracy.Where is this all going?
So, have I been a fool to tell you all this? No. This stuff is simple stuff, it's obvious for those who know just a little about the internet. And it's clear that these services (only the most obvious I might add) are already being used by people to cheat. Internet polls are not democratic, but they are fun. I don't approve of cheating, so this way we all have a level playing field. If you vote for me, do it once within the rules they publish. I would be honored with any votes.
Also note that Andy over Csikszereda musings needs your support.
I am certainly honored to be nominated, if only for the fact I am so lazy I never get round to dumping my mind much. Also, I love the fact that Catfish and I have been described as 'various' blogging contributors.
