Jon Skeet earns 1mm rep on Stack Overflow | Hacker News Reader

Jon Skeet earns 1mm rep on Stack Overflow

stackoverflow.blog

135 pointsjuliasilge30 comments

Open articleView on HN

> Jon Skeet's SO reputation is only as modest as it is because of integer overflow

From

https://meta.stackexchange.com/questions/9134/jon-skeet-fact...

I like his answer about Chinese time zones

https://stackoverflow.com/questions/6841333/why-is-subtracti...

Note how the answer has 30+ comments and yet no admin has dared delete them...

There are 123 additional deleted comments there too...

Once you have a large body of answers your score will just keep going up.

Mine has gone up about 40% in the last year and a half, despite not having time to answer any more questions.

Yep. Where that gets strange is being granted moderator-style privileges despite no longer being an active user.

One can be big on rep from ancient Q/As and yet have no idea about the "culture" of the site. That's me.

It gets a bit ridiculous over longer time scales. Almost all of my SO activity was in early 2009, when I got up to about 12k reputation. I had another small burst in early 2010 that took me to 17k or so.

I've had basically zero activity since then, and am now at 56,725. So I've gained about 40k while doing absolutely nothing for 8 years, and am still ranked #1331 (top 0.34%) on the site overall.

Here's a screenshot of my reputation graph: https://i.imgur.com/qgkdTz1.png

I'd be trying to drop six ranks in that case.

I've only been a contributor on SO for the last 1.5 years or so (though a long-time lurker since almost the beginning) and can tell you that it's not that easy or outright impossible to reach that kind of figures nowadays. Not that I care that much about my SO rep but still. Folks would often not even respond to long answers, use throw-away accounts, and the quality of questions is depressingly low, like in "How would I go about (description of a multi-year web project)?" without even the slightest effort of basic research or checking already answered questions. And of course the frequent case where students want to crowd-source their homework (sometimes answered by others who want to quickly grow their rep).

Was it always like that, or is it a recent phenomemon with more Web freshman coming to the scene since around 2012?

All the easy questions have already been answered so most new questions are obscure, poorly formed or duplicates.

And obscure, poorly formed duplicates don't yield much long term rewards. As it should be, but it still means that these numbers are out of reach. Except maybe if you manage to work in the hipster shop that only uses the new technologies before they become cool. Then you can go back to answering the basic questions.

But isn't that the way it's supposed to be?

I have always thought of reputation on SO as a measure of the total usefulness of your answers. When people find your reply useful, they upvote you. It's thus only natural that long-standing useful answers will keep gathering points, because people will continue to find those answers useful. When your answers are no longer useful (say due to technology change), your points too will stop growing.

> When your answers are no longer useful (say due to technology change), your points too will stop growing.

Ideally this is how it would go, but in reality it's more likely your outdated question will just continue to get points while misinforming new users by recommending older methods.

It does say quite near the start that Jon joined SO in 2008 and further on "In the near decade since then, Jon has answered over 34,000 questions, which works out to about 10 questions per day!" So I don't think he's just doing nothing and getting rep. I'm sure there is a bias, often people have a few good questions or answers that stand out and pull in the majority of their rep, like you mentioned about being idle for circa 1.5 years. However, if you read that post it seems Jon is still actively contributing.

I've long felt it would be worth trying some "diminishing returns" mechanics in SO. It might result in more attention given to the "long tail" of problems rather than people seeking "retirement investments."

I didn't know reputation was measured in millimeters.

Could be the amount of ink required to print out the actual number.

mm is a less common abbreviation for "million," but it does appear often in the US in finance and accounting

I don't get this either, it's not as if there are two "m"s in the word million. I must be spelling it wrong.

I believe it come from mille - Latin for 1000. So 1mm is one mille mille or one thousand thousand. Why we use the Latin? I have no idea.

MM is widely, though decreasingly, used in legacy finance:

https://english.stackexchange.com/questions/181917/mixing-us...

It’s also used by finance-outsiders who want to look like they are finance-insiders, a bit like how people will casually write GOOG and AMZN instead of Google and Amazon.

Anyone know who that other user is in the first graph, who just surpassed Jon? They seem to be answering at a crazy consistent constant clip.

https://stackoverflow.com/users/1144035/gordon-linoff

I came here to ask the same thing!

Looks to be #2: Gordon Linoff (https://stackoverflow.com/users/1144035/gordon-linoff) as he is at 41k answers.

If only stack overflow had a crypto currency, he'd be a rich.

I'd assume he is rich for a software engineer anyway. As he has written multiple books and is currently working as a senior engine at google.

The answers given by the model mentioned at bottom don't even make any sense. I wonder if ML will even be useful at harnessing and redistributing knowledge.

It's just a simple Markov chain model, obviously included for laughs.

That said, a relatively simple character-by-character RNN would probably have been more interesting.

It does give a relevant response w/r/t airspeed velocities.

I wish I had his job: spending my time answering and playing (it's a gamified Q&A) on SO all day and even being paid regularly.

Reply on news.ycombinator.com