Dealing With Minority Opinions

There have been some complaints that fans of "minority technologies" (not "ethnic" minority people) insert their point of view in many topics. For example, "As a foobar fan, my opinion on static typing is......". Some have complained, saying this is a pro-OO and pro-Pattern wiki, and that such minority opinions should be sidelined or perhaps deleted. Is there a way to at least have a link convention to minority opinions so that they are given a fair voice without upsetting the majority?


There are a few WikiProcesses? you can do to deal with minority opinions.

You can state that the site is NPOV, which means NeutralPointOfView. Basically, pages are worked to the point that no side disagrees with the phrasing of the page. It works surprisingly well, but people aren't able to give complete expression to their ideas, on that particular page. (The trade-off is worth it, in some cases.)

If the site is not NPOV, it may have a perspective policy. In that case, the policy determines how the page is framed. I would, personally, think that the wiki would have a page saying what that policy is, and then linking to "neighboring" wiki on the same subject, but from a different perspective. Wiki may become possessive, however, and argue that "we want to give people more time to consider our view," and thus not link to a neighboring ("competing") wiki.

I frequently argue for WikiNodes, which includes the notion of linking to neighboring sites.

This wiki, I think has neither a perspective policy, nor an agreement to be NPOV. I'm not sure, but it just seems to be that whoever is here at a given point of time, basically decides how it will go. That said, the PageDatabase is a vast and powerful thing, and has a magnetic power of it's own, pulling similar people to itself, and repelling very distant people. So, it's probably not as territorial as it would be otherwise.

-- LionKimbro

You're seriously misrepresenting NPOV. The intent of NPOV is to suppress value judgments, but this is blatantly impossible. Why? Because the mere voicing of an opinion implies it is a legitimate position to hold, the mere discussion of a subject implies it can be discussed, the holding of different views implies there is legitimate room for disagreement. As a result, only strong value judgments are forbidden on an NPOV site. This skews discussion and perverts values (it is a blatant lie that NPOV is value-less) so that, for example:

--anon.

To quote the MeatballWiki page, NeutralPointOfView:

"In reality, it is meaningless to accept all perspectives as valid or sound. No one could reasonably believe that the United Nations was an evil plot by space aliens to conquer earth, even if it were true. To have to accept all perspectives is to deny that any one opinion could be right or better, and worse, is to ForceConsensus?, which is the worst thing you can do. Typically, the perspectives represented are of those involved in the discussion, and perhaps a few others thrown in for good measure. If no one is willing to defend the position, or not enough people, then that position may be excluded from the NeutralPointOfView." --anon., retrieved November 23, 2003

Yup, and if you follow this to the letter then you've still got problems. Because there are always quacks and wackos who believe the weirdest / stupidest false things. Like for example, that "anarcho"-capitalism isn't a blatant contradiction in terms given that anarchism is a branch of communism. The opposite problem also occurs with NPOV.

With NPOV, you get total suppression of "controversial" facts which end up dismissed as mere "opinions held by a tiny minority". For example, anyone can see that corporations are psychopathic institutions, that they don't care a whit about human beings or human society, that they have no emotions, et cetera. And if you surveyed psychologists then more than 99% of them would agree. But try to put that on a wikipedia page and it gets dismissed because you can't point to a "study" proving that corporations are psychopathic. Geez, do you need a study to prove that the sky is blue?

With NPOV, your judgment that someone else's position is beneath contempt gets edited out. And your judgment that any position other than yours, which is blatantly obvious, is beneath contempt certainly gets edited out!

-- anon.

Who cares what Meatball says? I can quote my own wiki if I like, quoting as an anonymous poster. Woo-whoopty-doo.

At any rate, the issue isn't whether NPOV works or not. The issue is how we deal with minority opinions.

"Putting people in their place" sure doesn't sound cool to me.

Personally, I think every perspective needs it's own place to grow, and that we should have some sort of "MentalRightOfPassage?" idea that links perspectives. Sure, some perspective is going to want to exclude all of the others. But if it's even remotely connected to the network, it'll be pretty obvious, pretty quick. We don't have to LIKE other perspectives, but I do believe we should link to them. MentalRightOfPassage?, and all.

-- LionKimbro


I think the originator of this page (TopMind) is exaggerating the opposition to minority opinions on this wiki.

Somebody did tell me that this is a pro-OO wiki and that I am interfering. I can find links to it if you want.

Please do.

Minority opinions shouldn't be discouraged here. We all have them. But it's annoying to see the same minority opinion expressed every time the word "object" shows up on a page, especially when the person holding that opinion doesn't seem willing to accept that we are convinced we hold our opinions for good reasons.

No, it could simply be PersonalChoiceElevatedToMoralImperative.

But our personal choices are made for good reasons. There's no moral imperative involved.

We obviously don't agree, but telling us we shouldn't be using OO languages and techniques isn't going to stop us. Provide a compelling reason to stop and we will. Coming to a place where OO programmers hang out and telling them they shouldn't be OO programmers whenever they talk about OO programming isn't productive. -- EricHodges

I just want real, open evidence that OO is better before you belittle me for criticizing it. Is that asking too much?

I haven't belittled you for criticizing it. I enjoy arguing with you or else I wouldn't do it so often. But you are asking too much if you expect people to remain silent about the appropriateness of your contributions here. This isn't the place to get evidence that OO is better than procedural (or that functional is better than OO, etc.). You get that on your own. If you can't convince yourself to use OO, don't use it. But there's nothing to be gained by contesting its use by others every time they discuss it. -- EricHodges

I also agree with the above, this is a pro-OO Wiki, your not going to convince us we're all wrong. We all more or less arrived at the same conclusion that OO was better than procedural, and many think functional is better than OO (myself included). So yes, asking for evidence is too much, you must obtain your own evidence through use of the techniques. Until you've actually tried OO for a while, your criticizing it falls on deaf ears. You can't criticize something you don't understand well enough to put into practice without sounding like a beginner who doesn't understand. The fact that you always want evidence is proof that you don't have an open mind about the subject. Open minded people don't ask for evidence, they try things out and prove it to themselves. Open minded people take other people at their word when they say "hey it is better, you should try it". You don't take us at our word, you always insist on trying to force us to prove it, and that my friend, is why no one listens. I've not once seen you post asking a question about actually implementing an OO solution that you were trying, because I doubt you ever try, you just bash it. Look at comp.object, newbies are always posting stuff trying this or that, and they receive guidance until they understand it. You (TopMind) make no attempt to understand, you want some mythical proof laid before your before you'll even consider using it. Personally, I like the dissenting point of view, it seeds discussions, but your contempt for OO biases you, you really should try and see that.

I'd say that trying to find truth(which is what I assume Wiki is all about), inevitably leads to having minority opinions about something. The procedural versus OO folk can fight all they want, but I'd rather see it in a broader context. OO is not the penultimate. Aspect oriented, mathematically provable(functional), agent-based, and genetic programming are arguably more important than procedural or OO precisely because the decision to use them is currently a minority opinion. Want to prove why OO is better? Explain how it relates to the above. Of course arguing procedural versus OO is like arguing whether a wrench or a crane is better. Try building a crane without using a wrench. . . Now try lifting that car with the same wrench. As always, ZealotsPresentThemselvesAsTheMajority?. Also, I never considered this wiki as a "pro-OO" wiki, I thought it was about design patterns, and OO is just one of those patterns. Claiming pro-OO sounds like PrematureCalcification?.

Well, I read a lot of wiki, and I'd say I definitely come away with the opinion that OO is in the majority here. As far as agent and aspect based go, I see those as particular flavors of OO, but OO none the less. Functional is quite cool and superior to OO in many ways IMHO, and I think a lot of us see that. They are all tools to use, and arguing about them is fun, so don't call me a zealot, it's called passion.

Aspect oriented is not a flavor of OO. That's like saying OO/CeePlusPlus is a flavor of procedural/Cee. OO is a low-level part of aspect, not vice-versa. Give me a better noun than zealot meaning "passionate person" and I'll use that(who is this response to anyhow?). My point was that people like to present themselves as holding the majority opinion - when in truth most people sit on the fence. Perhaps Object Oriented is ill-defined, or worse, it falls victim to LaynesLaw. The ultimate "worst" scenario is if the wiki were so pro-procedural or pro-OO that it ceases to be agile - it's slow death. After all, LudditesAlwaysLose.

How about just passionate, zealot has other connotations, mostly negative. I'm certainly not trying to represent the majority, I just think that OO is the majority, here anyway, it's certainly not in the real world. In the real world I think the vast majority of programmers are procedural, one of the reasons I'm so passionate about OO/Function/Anything else! I'd love to be able to do Aspect Oriented programming, unfortunately my mandated tools don't do it very well yet, but that's slowly changing. I'm all for agile, if agile means easier faster better, I'll throw out any practice or belief the second I see something better, and often do, so zealot seems a bit harsh, to me at least.


OO is probably the majority in the "real world" as well. See http://www.tiobe.com/tpci.htm.

Well, I'll disagree with that, though I wish it were true. That chart is about language usage, and using an OO-enabled language doesn't mean you're doing OO. I've found in my own personal experience, that many people think they are doing OO just because they are using Java or C++ or something, but OO is a state of mind, not a language. I don't think OO has at all captured the mainstream, though I could be wrong. IMHO, OO is still fighting for mainstream acceptance.

Could it be that what we consider OO(design patterns, abstract interfaces, etc) is not what the mainstream considers OO? I still have my college textbook on C++, and there were no examples of interfaces, just inheritance. The design patterns course had the GangOfFour patterns, but still no interfaces. . . Maybe our definition of OO is really OO++ ?

Interesting idea, there does seem to be plenty of confusion on what actually constitutes OO. Hell, I can barely describe it myself, but I know when I see it, and I know to me, it's about polymorphism and structural patterns, not inheritance. Actually I have a great one word definition of OO, SmallTalk. But that usually just makes people look at me funny!

NobodyAgreesOnWhatOoIs


Some suggestions that at least work for me.

One should write with the goal of increasing his own understanding. Use written text to clarify one's thoughts, to ask questions of another, or to respond to questions.

Avoid use of "You." You is often used merely as a generic reference, but it is quite easy for a reader to interpret it as a personal reference. I tend to use "one," which although it sounds quaint, seems to come across as less accusatory.

Do not assume the motivations of another, do not even worry what the motivations may be. Assume the other person is also writing with the goal of increasing his understanding.

If someone makes a statement one disagrees with, ask questions to lead to personal understanding of the statement. Do not ask questions merely to try to "trap" another.

Do not try to have the last word in a discussion. Once one has reached his goal of increased understanding, walk away. Respond to serious questions by others who are trying to understand, but also allow others to disagree without comment.

One should not repeat one's self. Say something once. If there is disagreement, try to understand the view of the one disagreeing. This understanding of others will only lead to greater understanding of one's own beliefs.

Accept disagreement. No harm results from disagreement, and one should be able to walk away with a clearer understanding of his own beliefs. One is unlikely to change the beliefs of another, especially in the course of a few days. Changes in beliefs may occur, but will likely take months or years.

-- Anonymous

I like the last contribution quite a bit. I disagree with parts of it ("One" seems a bit stiff, compared to "You"), but I like the general spirit.

My only problem with it is that it's on the page DealingWithMinorityOpinions, rather than GoodStyle. I'd like to see it moved over there.

-- LionKimbro


CategoryWikiHelp, CategoryWikiDiscussion?


EditText of this page (last edited December 3, 2004) or FindPage with title or text search