The 21st IOCCC Winners | Hacker News Reader

The 21st IOCCC Winners

ioccc.org

61 pointsdlowe17 comments

The way the site is run really lets the contest down.

Why? How? They have a single page with all the winners ever, and for each C file they also have a hint.text/hint.markdown/hint.html file. They also have a Twitter feed.

What's so bad about that?

The problem is they haven't published the winning entries.

How can I appreciate the achievements of these coders if I can't see/run their code? There isn't even a description of what they've submitted (except for the clue given by the award title).

That's because they're still packaging them up and writing hint files and such. They just announce the names first and post the files later.

I understand, but that's the point: for the audience it's not so interesting to see the results without the winning entries, so to maximize interest, the jury should wait to present the results until they're ready to publish the winning entries too. It's not as if the IOCCC is working on a tight schedule anyway.

This is like hearing who won in the Olympics and then watching the games later at night; sure, you can do that, but it's much more exciting the other way around.

At least, it shouldn't be posted to HN when the code is not published yet. As it is now, this doesn't satisfy my intellectual curiosity.

Weeeelll:

1) As others have pointed out, the separate release of code/writeup. I wager that most people really just want to see the code. Why bother doing both? If it's just to attract attention, that's a reasonably valid reason -- but it probably has the opposite of the desired effect, people get annoyed that they can't view the code now, and therefore start disliking the contest in general.

2) Even when the code is released, the format they use makes it difficult to actually look at the code. Let's take the 2011 winners for example [1]. One would assume that clicking each entry (e.g., the top entry[2]) would perhaps show the code, perhaps annotated in a useful way... nope! Surely it must link to the code then... nope! Perhaps it links to a page that links to the code... nope!

You have to go to the 'Winning Entries' page from the main menu (from [1]), scroll down to the appropriate year, then play a fun game of matching up entrant's surnames to their entries (yes, I know the surname is in the URL of [2]).

[ 3) While not a 'real' complaint: the I-just-learned-how-to-use-a-3D-modelling-program logo really is awful. It makes the contest look juvenile, when in fact the quality of the winning entries is very high. The site looks much better without the logo -- try it yourself, delete the img node from the homepage using Inspector/Firebug/whatever. ]

In conclusion: show us the code, delete the logo :).

[1] http://www.ioccc.org/2011/whowon.html

[2] http://www.ioccc.org/2011/blakely/hint.html

For a contest about obfuscation, the site could be a lot worse. ;)

1) The primary reason to do separate releases is because the winning entries do not appear in the website as is. As far as I know the judges have to write separate remarks, the authors have to check any remaining problems and judges and authors have to agree on the finished write-ups. While it is a bit tedious, given the number of winning entries it seems reasonable.

2) Yup, agreed. For example the current website does not allow inspecting the code without downloading it first. (That's why I love www2.us.ioccc.org...)

3) They are not ordinary logos. They are made from the ray tracers from previous IOCCC winners (2004/garave and 2011/zucker, respectively). I do think that those logos should really link to the relevant entries, however.

(2) and (3) are now fixed.

(2) The mime types for the .c and .h files are now set to text/plain on http://www.ioccc.org.

(3) I (Simon) have added a link to the winning entry that generates the current logo.

Oh, thank you a lot! Should have sent emails before, but I always forget...

Some interesting points. If you have some better ideas about how to organize the data, please submit a proposal to the judges (info here http://www.ioccc.org/judges.html). Even better if you can volunteer the XX or probably YYY hours to actually implement it. The site is statically generated and we want it to remain that way.

As to point (3), the current logo is generated by Matt Zucker's 2011 winning entry. I (Simon) have added a link to it from the logo and a little annotation :-)

So - where is the code?

I wondered that too. But in the news section[1] of the front page:

The winning source will be released later this year after the winners have reviewed the writeup of their entries.

[1]: http://www.ioccc.org/index.html#news

The whole fun of the ioccc is to go through the code and figure out how it is done. Without the code this is a non-event.

Reading the write-up in combination with the code is usually really entertaining. Having the code before that would make it a lot less entertaining. But you are of course right, as long as there is no code, there isn't much to actually see here.

"writeup"? Ouch! Ah, it's the The International Obfuscated C Code Contest.

Reply on news.ycombinator.com