June 1, 2006
Brendan Kidwell noted something interesting about the comments on Trackpoint versus Touchpad:
You know, I just realized that there is a lot of LOVE and HATE going on, and it doesn't seem to be very unified.
Software development is basically a religion. It's not surprising that software developers are predisposed to these kinds of religious arguments. Pick your poison:
- Linux vs. Windows
- Mac vs. PC
- C# vs. VB.Net
- Ruby vs. Python
- Static typing vs. Dynamic typing
- Spaces vs. Tabs
.. and so on, and so on, and so on
, ad nauseam.
I am reminded of Robert Mitchum's character in the movie The Night of the Hunter, a preacher with the words "love" and "hate" tattooed on his left and right hands:
Would you like me to tell you the little story of right-hand/left-hand? The story of good and evil? H-A-T-E! It was with this left hand that old brother Cain struck the blow that laid his brother low. L-O-V-E! You see these fingers, dear hearts? These fingers has veins that run straight to the soul of man. The right hand, friends, the hand of love.
Heavily partisan discussion-- lovers on one side, haters on another-- isn't conducive to civilized discourse. Jon Stewart's now-legendary appearance on Crossfire was a textbook indictment of this style of debate. It's not debate at all; it's pure theater. Entertainment. Professional wrestling, if you will.
This is a common problem with online reviews at Amazon and other sites. The people most strongly motivated to post reviews are those who either:
- had such a good experience with the product that they can't stop talking about how great it is.
- had such a bad experience with the product that they dedicate the next few days of their life to warning everyone they can away from it.
Either way, you're getting a distorted opinion based on the select few who had a strong enough negative or positive reaction to spur them into action. It may be fun to read, in a partisan sort of way, but it's difficult to distill the vitriol into something you can base a reasonable opinion on. What about the vast, silent audience who liked it despite its flaws?
That said, a certain amount of controversy is desirable. If nobody cares enough about what you're doing to tell you that they loved it or that they hated it, it's probably not interesting enough to be good, either. As Kathy Sierra points out, you should strive for a little controversy:
If you're out there creating something on the edge, someone's going to hate it. Probably a lot of someones. One thing we noticed from our Amazon reviews was that we get mainly five-stars and one-stars, but not much in the middle. They either love it a lot or they hate it with a passion. Whenever I start to feel bad about a scathing review, I remind myself that Don Norman said, "If someone doesn't really hate your product, it's mediocre." And mediocre is where you SO do not want to go.
Ever since we started this crazy scheme (18 months ago with the release of the the first book in the series), we've been thinking that the extremeness of our reviews was a good thing, and now someone's confirmed it. A NYTimes article looks at a professor who analyzed Amazon book rankings for, among other things, a book's "controversiality index". From the article:
"But the most telling variable is the one star rating. Professor Gronas found that books high on what he called the "controversiality index" are given almost as many one-star as five-star ratings, creating a horseshoe-shaped curve. As it turns out, these books also tend to have high sales."
The last reaction you want to anything you've done is a resounding "Meh." Go out of your way to create something that will inspire people to love it. Unfortunately, that means a few people will probably hate it, too. The love/hate dichotomy has value. But it's also important to maintain perspective. It's all too easy to fall into the trap of framing your discussions in that old timey religious love/hate dynamic.
Posted by Jeff Atwood
great blog! i've enjoyed reading your stuff since discovering this lately.
unrelated to any of the actual entries: you may need to fix your blogging software. the past few entries have been for 11:59 for the day before you actually posted. also, the 'enter the word' (for leaving comments) is always 'orange'
Don't forget Kathy's great post on the "Kool-Aid Point," where people go religious about your product:
For the most part, Amazon and IMDB reviews are meaningless to me; I don't care what Who-The-Heck-Are-You-Anyway thinks, I care what someone thinks whose thinking I have some context for. (Similarly, anyone reading this doesn't care what I think about Amazon reviews ... :-) )
That said, when contemplating a purchase, I will sort the reviews with least stars first and see what about the product drove people to give it such a bad review. You can occasionally get some useful insights that way.
Same, the "this thing is a piece of crap" reviews are a great resource. They're the fastest way to find out if the vendor doesn't support their product, if it has non-obvious flaws or incompatibilities, or if there are trends in failures. Or whether the majority are just problems clueless dorks and people looking for something to bitch about invent.
I pass up stuff that never gets reviewed anywhere, unless it doesn't cost much, because why take the risk? (If only it was as easy to find reviews of enterprise software online, it would save a lot of grief and wasted dollars.)
I'd say that, coupled with flair, panache, and of course elan -- always elan! -- this remains insightful advice when applied to social interaction. It's a fine ENTP way to live.
Coding Horror has rapidly become one of my favorite reads; thanks for the good times!
you may need to fix your blogging software. the past few entries have been for 11:59 for the day before you actually posted. also, the 'enter the word' (for leaving comments) is always 'orange'
Strangely enough, this is all by design..
when contemplating a purchase, I will sort the reviews with least stars first
Same, the "this thing is a piece of crap" reviews are a great resource
Well, I do the same thing-- but I have to do it for the least stars *and* the most stars, and *then* filter out the weirdos on each extreme.
It's the entire problem in a nutshell.
you may need to fix your blogging software. [...] also, the 'enter the word' (for leaving comments) is always 'orange'
Strangely enough, this is all by design..
What's that then?
Security by intimidation?
The 'orange' captcha (I hate this word) is an example of a very simple solution that seems to work (or is Jeff deleting all the spam that gets through? ;-)
I guess this wasn't a very good post topic if all we can talk about is the blog captcha.
This blog is based on Movable Type, which is written in Perl. I don't know Perl. This is the best CAPTCHA solution I could implement with my crappy Perl skills.
And the best part of all? It works perfectly. I have gotten a grand total of two spam comments -- clearly hand-entered -- since I implemented the Crappy Captcha(tm).
Here's a fun scenario:
You are looking at a product that very likely will fill your needs. It's kind of high, and no one you know has one. It's only got 2 raving reviews and 3 bad ones. What do you do?
If you knew that Amazon has sold 3700 of them, those 3 really bad reviews look a lot less bad. I think knowing how many of each item they've sold would make the ratings and reviews more helpful. I don't know that it'll ever happen though, that seems like the kind of thing that retailers may not like getting out.
There's one point to consider in these A vs. B programming religions; that there's a huge investment SOMEONE has made SOMEWHERE, that he doesn't want to lose by switching from A to B. That's why we have Java vs .NET; The windows guys don't want to throw everything away to go Java, and the java guys don't want to be tied to one platform. etc. It goes on and on. Emacs users don't want to learn how to do things the VI way. It's usually NOT the case that A is incomparably better than B; its just that no one wants to make the switch.
There's also the element of having to doubt yourself. By switching from A to B, you're saying "I was wrong for picking A in the first place. I was a fool for spending so much time on A." No one wants to do that; men in particular, we're not wired for it. It takes a lot of confidence to be able to do that, and if we all had a lot of confidence, the world would be a very different sort of place.
Speaking of capatcha's, how about this;
Instead of making us type a word, as us a question, and have the answer be an image. Make it multiple choice. 5 choice, drawn as images, and let us click on the right one. Typing the right one is a pain. Make the answers pictures; "Which fruit is used in tropicana."
If you don't know what tropicana is, tough.
I love this kind of topic because it exposes the psychology and sociology of software development (although, in similiar fashion, many other things as well.)
Too often people fail to dig just a little deeper into the 'why' of something (and religion is definitely a biggie.) For example, why do you preach at the C# pulpit? Why is Linux better/worse than Windows? Why is that HDTV you just bought a piece of junk? What influences have affected your decision? Do you truly strive to see all points of view? Are you attempting to analyze every position with an open mind?
Groovy topic, Jeff.
You know what Jeff? I bet you don't even need the word to be an image. Effectively you've got a Turning Test here - copy the colour into the box. No spam bot is going to do that.
Jeff, you've actually authored a captcha (I love that word) on codeproject, so what gives with the Turing test whose answer is predictable? Why not just use the server control that you made, instead of using the orange thing? It's like protecting your car with the Club, except your 'the Club' is made out of paper mache! It's probably a successful deterrent, in that the thief will move on...