Month: February 2007

…I am! Ok, so I wrote a post yesterday called “Digg Comment System is Retarded” in which I was complaining about the *new* way story comments are displayed. A short time later after the post was posted to Digg, one user pointed out the following:

you know you do have the option of viewing the comments in differing formats:

sort by most diggs
sort by date (show all)
sort by date (-10 diggs or higher)
sort by date (-4 diggs or higher)
sort by date (+0 diggs or higher)
sort by date (+5 diggs or higher)
sort by date (+10 diggs or higher)

it’s a drop down box right above the first comments number of diggs
sounds like this would solve most of your problems with digg’s comment system

I’m feeling a little sheepish about the whole thing now. In my defense, Digg adds features regularly and after a fairly recent feature addition, the comment view defaulted to the “sort by most diggs” selection which is what I was seeing every time I viewed story comments. I believed that this was just a feature change that I had to live with and it was really getting under my skin how illogical it was.

Frankly I can’t imagine why anyone would want to view the comments this way except to verify that comments from more liberal/irreligious posts get dugg up and conservative/religous posts get dugg down, but the Digg programmers are pretty sharp and I’m sure they’ve got some reason for it. So really, I just need to say, hey folks at Digg, I apologize for the “Retarded” comment. I see you’re really just trying to make it accommodate as many features as possible. You all do great work and I’m sorry for doubting you.

Now, with all that being said, I still have to take issue with one of the items I named in my original post:

Posts are not rated based on whether they are good or well thought out, but rather on whether or not the person rating the comment agrees with what was said.

I acknowledge that this doesn’t take issue with the comment system, but rather with the commentators. And I do take issue with the fact that people do this. I doubt my mentioning it will cause people to change, but I do remember a time when we used terms like “good netizen”–meaning someone who is responsible on the net (tubes), and there were whole documents on ‘netiquette”–that is, net etiquette. In case you’ve forgotten etiquette means (and yes it’s a dictionary definition) “conventional requirements as to social behavior; proprieties of conduct as established in any class or community or for any occasion.”

Having proper netiquette in the context of the Digg comments section I think would mean digging comments up or down based upon whether they are demonstrating well reasoned points and whether they are written with temperate language rather than whether or not you agree. If I disagree with someone but they’ve made their arguments well and with respect, I digg them up. There is so much juvenile name calling (yes I realize I used the word Retarded to describe the comment system–nobody’s perfect) that it makes it a place not of enlightenment and discovery, but rather a place of vitriol and anger. It’s completely counter productive.

In hindsight, it’s not the whole system, just the “sort by most diggs” feature that is umm, not so great (retarded is probably too strong), but at least I have choices. I am now viewing the comments using “sort by date (show all)”.

Since its inception every aspect of Digg has gone through refinements, the vast majority of which have been good. This latest version of the comment system, however, is the most retarded thing I have ever seen. Usually comment systems are either what you call “flat”–meaning the comments are just posted in ascending order according to the date when the comment was posted, or they are “threaded”–meaning when someone responds to a comment, their comment is placed indented right below the comment to which the user was responding.

The folks at Digg have tried to keep some control over the way people post by limiting replying only to the top level comments (e.g. avoid infinite thread cascading)–which was good, but now it appears that posts are listed on the page only according to the number of diggs–positive or negative–a post may get. Here’s what this has amounted to:

• Replies to posts are scattered about and you never know who the person was responding to without doing a page search to find similar keywords.
• Posts are not rated based on whether they are good or well thought out, but rather on whether or not the person rating the comment agrees with what was said.
• Often the top rated comments are just jokes that barely relate to the topic at hand.
• When the posts are of a controversial nature (e.g. religious, political, etc.) clearly the majority of “diggers” are non-religious and left wing. If you don’t believe me, just look through any of the political stories at digg and you’ll see that if you’re a conservative, you just need to file to the bottom of the page and if you’re liberal, you will get promoted to the top of the page.

I suppose that you can never create a system that will make everyone happy, but this current version is an illogical atrocious mess that really needs to be re-thought. I’m not sure what the underlying issues were with the previous iteration of the system, but this is definitely a step backwards.

In his follow-up to hist post on Coding Horror about the FizzBuzz problem (Why Can’t Programmers… Program? (February 26, 2007), Jeff Atwood says:

It certainly wasn’t my intention, but a large portion of the audience interpreted FizzBuzz as a challenge. I suppose it’s like walking into Guitar Center and yelling ‘most guitarists can’t play Stairway to Heaven!’* You might be shooting for a rational discussion of Stairway to Heaven as a way to measure minimum levels of guitar competence.

And then a paragraph or so later he says:

The whole point of the original article was to think about why we have to ask people to write FizzBuzz. The mechanical part of writing and solving FizzBuzz, however cleverly, is irrelevant. Any programmer who cares enough to read programming blogs is already far beyond such a simple problem. FizzBuzz isn’t meant for us. It’s the ones we can’t reach– the programmers who don’t read anything– that we’re forced to give the FizzBuzz test to.

What I believe Jeff has discovered is that programmers, all programmers (especially the ones who read programming blogs) are insecure about their ability to write code.

Programmers, when we first come into contact with a rudimentary problem that might call into question our competency, need to make sure we agree that the problem is in fact rudimentary because if we are not able to solve it then there can only be two possible conclusions, either the problem is not rudimentary after all, or we are not competent.

As I see it, upon encountering this post, most programmers went through four phases of competency validation as follows:

1. Am I able to solve the problem at all?
2. As a ‘senior’ programmer, am I able to solve it in under 10-15 minutes (remembering that a quote from Imran in the original post stated, “I’ve also seen self-proclaimed senior programmers take more than 10-15 minutes to write a solution.”)
3. Now that I have solved it under that time, how long did it actually take me? In other words, where do I rank among true programmers?
4. Now that I know I can do it in under 3 minutes, I must create the most cleverly written one-liner ever, ever!

In my mind, when Jeff says,

And instead of writing FizzBuzz code, they should be thinking about ways to prevent us from needing FizzBuzz code in the first place.

he’s missing what actually happened. When programmers saw the post, they said, “I’ve never had to solve that problem before and if some senior level (self-proclaimed or otherwise) programmers took 10-15 minutes to solve it, I had better check to see that I can do it in less time (or at all for that matter)”.

Dont’ you see what you’ve done Jeff?. Every programmer out there who considers himself competent just went and took a test to make sure he really is. Imagine that there was someone who should know how to solve the problem who currently holds a programming job who just failed the test miserably. Maybe he will come to his senses and stop wasting his company’s time and his own life. If this is truly a problem, that people get into positions because nobody asked them the FizzBuzz question when they were being interviewed, then the FizzBuzz post has just shown them the truth and maybe they will now ‘see the light’ and finally stop making their fellow programmers look bad and go away.

Then again, they probably won’t, but at least now they know. They can’t deceive themselves any longer. They are not competent. The jury is no longer out for them.