IOCCC28 winning entries

[lcn2]

Let's discuss the winners of IOCCC28.

What are some of your favorite entries under the 2024 directory?

Besides the initial release bug where we had 2024/anders and 2024/straadt swapped ...

SORRY (tm Canada 🇨🇦) about that

Are there any problems, typos, and/or mistakes in the IOCCC28 release?

[lcn2]

The 2024/anders and 2024/straadt swap has been fixed.

If you download tarballs before 2025 Aug 03 03:00:00 UTC, please re-fetch the tarballs.

We SINCERELY APOLOGIZE to both Anders Kaare Straadt and Sebastian Macke for the mix up!

[lcn2]

Per request from Sandro Maffiodo, we have edited the 2024/maffiodo/README.md file, updated the related 2024/maffiodo/index.html and updated the tarballs as needed.

[xexyl]

Let's discuss the winners of IOCCC28.

What are some of your favorite entries under the 2024 directory?

Hmm I wonder! :-)

I haven't had a chance to look at the others actually but I should guess Endoh's are brilliant (as always) and same with @SirWumpus's. Those are the ones I am hoping to look at when I have the time and energy. It is not clear to me when that'll be but those I am excited to look at if not others (which I am but these stand out as ones I want to look at).

[xexyl]

Are there any problems, typos, and/or mistakes in the IOCCC28 release?

Problems? Yes. I'm not the only winner :-)

Typos and other mistakes? I've made some fixes (as have some others) already but I barely looked. I guess there are more.

I noted in another comment that 2024 is still referenced in copyright messages - but I am not going to fix that as I don't want to cause any git conflicts.

[xexyl]

BTW shouldn't this in the index file of 2024:

Using a lot of goto statements to make your code harder to understand might not help it pass the final rounds of judging.

... instead be changed to the new guidelines after what I did? :-) Since obviously that's not the case this year - I smashed the record at 134 (up from 24 in 1987) and I guess nobody else is going to do that for the very reason that it is probably now 'old' and boring (in which case I am happy to have done the contest a great service). What do you think @lcn2 ?

[lcn2]

BTW shouldn't this in the index file of 2024:

Using a lot of goto statements to make your code harder to understand might not help it pass the final rounds of judging.

... instead be changed to the new guidelines after what I did? :-) Since obviously that's not the case this year - I smashed the record at 134 (up from 24 in 1987) and I guess nobody else is going to do that for the very reason that it is probably now 'old' and boring (in which case I am happy to have done the contest a great service). What do you think @lcn2 ?

The above remark was something Leo wanted to add to this year. Think of that as a reminder that when we rework the guidelines, we need to mention something like this.

It is not that we will ban using lots of goto statements. What it means is that simply obfuscating code flow by using lots of goto's the jump around is in itself, not a sufficient obfuscation technique in the future.

[xexyl]

BTW shouldn't this in the index file of 2024:

Using a lot of goto statements to make your code harder to understand might not help it pass the final rounds of judging.

... instead be changed to the new guidelines after what I did? :-) Since obviously that's not the case this year - I smashed the record at 134 (up from 24 in 1987) and I guess nobody else is going to do that for the very reason that it is probably now 'old' and boring (in which case I am happy to have done the contest a great service). What do you think @lcn2 ?

The above remark was something Leo wanted to add to this year. Think of that as a reminder that when we rework the guidelines, we need to mention something like this.

I see. As below - that goes for everything but also as below it might be confusing to some. That's why I asked. Most would see my code (I guess) and just think it won because of all the gotos. I mean obviously that had some help (how could it not with how many I did?) but I did a lot of other things too.

It is not that we will ban using lots of goto statements. What it means is that simply obfuscating code flow by using lots of goto's the jump around is in itself, not a sufficient obfuscation technique in the future.

That goes for every technique of course! I just found it bit odd as although I did plenty of other things it can't be denied that the goto amount was ludicrous.

Even if you don't 'ban' them I guess after what I did they're going to have far less significance. I certainly hope so. That was one of my goals. That and showing just how much is lost. But as above the reason I brought it up is it might confuse some people if it's here when one of the winning entries has 134 gotos!

[xexyl]

As for typos I do see more. In one of Endoh's - in the judges' remarks. I'll have to leave soon though .... I know I already did a huge amount of typo fixes but I can help more - I hope - if you wish. Tomorrow is a hectic day and a number of things (as you know) have to be resolved but I can try and get some of those in if I have the energy (and remember).

[lcn2]

As for typos I do see more. In one of Endoh's - in the judges' remarks. I'll have to leave soon though .... I know I already did a huge amount of typo fixes but I can help more - I hope - if you wish.

Yes please 🙏

We will hold off on re-building the 2024 tarballs until you have performed the above.

[xexyl]

... also I wonder if I should do it today: since some people might still be making corrections in their entries.
Let me know what you think about that @lcn2.

It would be a good idea to get the corrections in early. That would allow us to rebuild the tar balls as well. Thanks 🙏 @xexyl

Fair enough. I'm working (kind of - today is crazy and much of this week will be too) on some updates to my entries and then I'll see if I can do anything. I should have finished mine a while ago. Have been working on other issues. I hope to resume soon. I must say I doubt I'll have the energy to do this request of yours today but we'll see.

[SirWumpus]

@xexyl What is involved with rebuilds? I think if you share the process, we not reliant on just two people to assemble things in future.

[xexyl]

@xexyl What is involved with rebuilds? I think if you share the process, we not reliant on just two people to assemble things in future.

Are you referring to rebuilding the html files and the manifest etc.?

[SirWumpus]

Yes.

[xexyl]

It actually depends on what. There are a lot of scripts and Makefile rules.

There are also dependencies. The scripts will tell you about that if they're missing.

To rebuild ALL html files:

make www

but there are many others that are more specific.

As for the manifest that's more complicated. You can do:

make entry2csv

and then open the CSV file in a spreadsheet application. Then update the spreadsheet appropriately (the description of each field is beyond the scope here) and then run:

make csv2entry

.. and then build the necessary html files.

All the rules can be seen by:

make help

Does that help some at least? See also bin/index.html (you can read the markdown file - bin/README.md - instead).

[xexyl]

How bizarre .. there was a manifest error. I can't understand it because you, @lcn2, would have run the proper make rules. An entry in 2024 had a space in one of the fields. So either something went wrong in the script or it sneaked in somehow. But how could that be with the commits passing the test suite? Or did we not test that? I could have sworn we did but maybe not the csv2entry rule? I guess that might make sense - how could it deal with the spreadsheet? Even so strange it failed ... doing updates now. Soonish I'll have to stop so I guess I won't get to typo checks today. We'll see.

[xexyl]

How bizarre .. there was a manifest error. I can't understand it because you, @lcn2, would have run the proper make rules. An entry in 2024 had a space in one of the fields. So either something went wrong in the script or it sneaked in somehow. But how could that be with the commits passing the test suite? Or did we not test that? I could have sworn we did but maybe not the csv2entry rule? I guess that might make sense - how could it deal with the spreadsheet? Even so strange it failed ... doing updates now. Soonish I'll have to stop so I guess I won't get to typo checks today. We'll see.

Even more bizarre: 2020/yang had the wrong version in .entry.json ... not sure how that could have happened.

diff --git i/2020/yang/.entry.json w/2020/yang/.entry.json
index db8f535516b0819950fdcd1be6fd109f2c98f440..179b64accf086e49a5af5e7b1c42e434f01271a3 100644
--- i/2020/yang/.entry.json
+++ w/2020/yang/.entry.json
@@ -1,8 +1,8 @@
 {
     "no_comment" : "mandatory comment: because comments were removed from the original JSON spec",
-    "entry_JSON_format_version" : "1.1 2024-02-11",
+    "entry_JSON_format_version" : "1.2 2024-09-25",
     "award" : "Best abuse of CPP",
     "year" : 2020,
     "dir" : "yang",
     "entry_id" : "2020_yang",
     "title" : "2020.yang",

I pulled changes earlier today.

[SirWumpus]

Comment from https://news.ycombinator.com/item?id=44774104 :

Unfortunate that the redesign broke all links to the old entries (which, since they've been around for decades are on quite a few sites - wikipedia for example)

Doubt there is much that can be done about the old links.

[SirWumpus]

"Cardigan" tends to be a button-up sweater as I recall.

[xexyl]

"Cardigan" tends to be a button-up sweater as I recall.

Something like that anyway ... with interesting etymology:

---

cardigan | ˈkɑːdɪɡ(ə)n | noun a knitted jumper fastening down the front: I wore a light cardigan over a sundress. origin mid 19th century (Crimean War): named after James Thomas Brudenel, 7th Earl of Cardigan (1797–1868), leader of the Charge of the Light Brigade, whose troops first wore such garments.

---

But even knowing that I can't keep track of it. It's worse when you throw in terms that mean different things. Don't get me started on the word 'literally'. I very rarely use it now because it's basically meaningless. Yet the lexicographers had no choice. It was abused so much that it had to be added as a definition. Still I wish they would have drawn a line - somehow. Quite different from clothing of course but ....

[SirWumpus]

Well having lived in Australia, I literally learned a lot of names for things.

[xexyl]

Well having lived in Australia, I literally learned a lot of names for things.

Mostly mate I am sure! I will not acknowledge the other word beyond saying I will not acknowledge a word is not being acknowledged.

[xexyl]

Well having lived in Australia, I literally learned a lot of names for things.

And you should try German. They have so many ways to say 'the', 'you' and 'bitte' (for example) can mean (or be used for) so many things too ...

Plus they have compound words so they can make up a word for anything by putting more than one word together.

[SirWumpus]

Ok. This is broke. HTML links from the web site that POINT BACK TO GITHUB do not render, while the .md files do.

For example from this page https://www.ioccc.org/2024/howe/index.html click and compare the visual output for the links prog.html vs prog.md. The prog.html shows you the source HTML in the GitHub repo (not the pretty printed page), while the prog.md is formatted by GitHub and easy to read.

I've always wonder about this, why bother generating things like prog.html, when the prog.md is sufficient to view (especially if with Github formatting). You only want the prog.html if it is hosted (off GitHub) by a web server unless you can get Github to present the HTML without the Github repo related stuff.

I do not know if GitHub has a API URL that provides the proper presentation, but this is not it.

This happens in OTHER winners directories too. Consider https://www.ioccc.org/2024/ferguson1/index.html. Click the link for deobfuscation.html vs deobfuscation.md.

It appears that any winner that presented supplemental *.md files has this problem. The list is...

2024/burton/deobfuscated.md
2024/burton/extra.md
2024/ferguson1/deobfuscation.md
2024/ferguson1/obfuscation.md
2024/howe/prog.md

[xexyl]

Voice would never work anyway

I could still screen share and type wildly.

Ah ... well all I can say is I've not forgotten about it I just am not sure I could deal with it - yet. And right now I'm too overwhelmed with other things. Anyway what matters is I see what you mean now. Definitely bug.

[SirWumpus]

BTW I just check (only one) OLD winner with supplemental *.md files.

https://www.ioccc.org/1992/buzzard.2/index.html hover over buzzard.2.design.html and notice the URL points to the IOCCC web site link https://www.ioccc.org/1992/buzzard.2/buzzard.2.design.html - correctly formatted.

So it appears to be an issue with link generation of NEWER winners.

[xexyl]

(Its times like this I wish you used Discord and then I could explain in voice and screen share what I'm seeing.)

Voice would never work anyway .. I have CAPD (central auditory processing disorder) (hearing aids help but are not perfect) and it would also be too much for me. I'm not good with voice conversation except with a few people. In fact I rather dread it. But I see what you mean now anyway. It's a bug.

It might be that the manifest has them as the view on GitHub field. I'm not sure.

Thanks for reporting anyway.

My suspicion was right. The display_via_github was set to true for these files; they should be false. I can fix it but first I have to recreate the CSV (due to recent updates). I can then fix it and then rebuild the 2024 html pages.

[xexyl]

So it appears to be an issue with link generation of NEWER winners.

Yes you're right. In particular 2024. 2020 should be fine. Fixing ... but it takes a bit of time.

[xexyl]

So it appears to be an issue with link generation of NEWER winners.

Yes you're right. In particular 2024. 2020 should be fine. Fixing ... but it takes a bit of time.

Had to take care of some other things ... commit 951a42b should take care of this. Hopefully I got them all.

Thanks again!

[xexyl]

Wrt the tarballs in commit acbfd35: I know I haven't been able to get to typo checks yet. I am sorry. It's been kind of crazy. Server still not reachable .. that's another story entirely but that and the other things I've worked on (here) and other things unrelated - it's all conspired against me helping with typo checks. I will try doing a bit more today if I can but I don't know how much that will be. We'll see. I just wanted to let you know I have not forgotten.

[xexyl]

Wrt the tarballs in commit acbfd35: I know I haven't been able to get to typo checks yet. I am sorry. It's been kind of crazy. Server still not reachable .. that's another story entirely but that and the other things I've worked on (here) and other things unrelated - it's all conspired against me helping with typo checks. I will try doing a bit more today if I can but I don't know how much that will be. We'll see. I just wanted to let you know I have not forgotten.

No worries. The tarballs were mainly to catch some of the changes to source code and make files and transcripts. People who download the tables are probably less interested in the webpages. When you complete your typo changes, we can rebuild the turbo again.

That is good to know. Unfortunately I have run into more snags so I am kind of stuck with what I can do. I do hope it is all resolved soon.

You wouldn't think that preparing for the new service by buying a router early would cause so many problems but for once you would be very wrong!

It's a long story and it's completely bonkers.

I am actually thinking I might turn on the old network temporarily so I have at least some connectivity.

Time to get some food as well. I HOPE tomorrow I can continue but I suspect if it is fine I will have to do some final configuration changes.

I already did most of it but I can't finish it until it's ready as otherwise there will be connectivity issues.

I might then just collapse. Take care and hopefully I can get back to the typo checks soon!

But that does bring up a question that I will ask in another comment.

[xexyl]

As far as my question/comment.

My understanding is that I should just check the judges' remarks.

Or maybe that's what I had in mind. It was one thing when working on the old entries when redesigning everything but it seems like going forwards it shouldn't be too necessary other than some typos that slipped through when you were preparing the write-ups.

This would simplify the process and also speed it up.

But of course I have to first have the other issues resolved. It MIGHT be that I might work on them even offline but that's certainly not happening today. I hope that the connection is restored by today too.

Of course it would not really hurt that much (I would think) if these were not resolved until after you get back from holiday. But hopefully I can get to it sooner than that: or perhaps it's not preferable? Maybe you want to get going soon after all!

Did you decide yet where you're going?

Best wishes!

[lcn2]

As far as my question/comment.

My understanding is that I should just check the judges' remarks.

As some of the winners do not have English as their primary language, checking for spelling and grammar issues for the rest to the index.html page would be a service to them too.

[xexyl]

As far as my question/comment.

My understanding is that I should just check the judges' remarks.

As some of the winners do not have English as their primary language, checking for spelling and grammar issues for the rest to the index.html page would be a service to them too.

Fair enough.

I will keep that in mind when looking at them.

Thanks and hope you have a great rest of your day!

[xexyl]

As far as my question/comment.
My understanding is that I should just check the judges' remarks.

As some of the winners do not have English as their primary language, checking for spelling and grammar issues for the rest to the index.html page would be a service to them too.

I haven't entirely forgotten this - but it's been hectic and stressful. Things are calming down now but I'll need time to break. I did discover a few typos and have fixed them but I haven't committed them and I don't see myself doing so for a while.

Just so you know though.

[SirWumpus]

Wikipedia

@lcn2 I'm going to look at it tomorrow. I showed you some links I think will hopefully address their (comments at tope of page) like 3rd party links that are independent -books, mags, review etc.

If you know of any other books (I have Don Libes) that discuss the IOCCC or its winners; course materials (have Harvard CS50) ; etc. drop them here please.

[xexyl]

REQUEST

We aer briefly in range and want to follow up, @xexyl on your discussioncomment-140590480.

David, the admin, is trying to contact you via discord direct message about this issue as it is important that regain access to the discord server to see resources people have posted about IOCCC for the purposes of adding references as part of improving the Wikipedia page for IOCCC.

I have replied below on this subject.

As for wiki: like I said somewhere (think I deleted it as I must have responded not in the thread directly) I have a disliking for their politics but I have mentioned one place an entry is shown - one I had to fix for modern systems (it not only did not work with clang but it did not work with gcc and iirc it didn't work in linux or macOS: the 12 days of Christmas one). I know some articles out there have talked about the contest some including citing you but I'd have to find those and I'm not sure if they're worth having.

Being on a channel - the thought of it - is kind of overwhelming but if I limit it I might be able to do some. @SirWumpus suggested it's not different from mastodon but it's unfortunately (for me) not nearly that simple. Even mastodon I often pause before I say anything at all and I might go many days - weeks - before I even open it.

Still the wiki page for the IOCCC obviously needs help otherwise we wouldn't be having this discussion!

There is also another project that I believe you will be interested in where the discussion will also occur on discord. So PLEASE work with David to restore your access.

Something else, eh? (I hope you appreciate that wording!) This is curious to me ... although I hope that I will have a bit more time this next contest versus the few short months I did IOCCC28. It took me two months to the day to finish 'A Pact With the Devil On the Oregon Trail' - I had mere hours left! I was totally knackered by the end of it. And as you know I had to help with mkiocccentry even after the contest opened. The fact everyone else had near 5 years and I had about 2.5 to 3 months and I still won two entries again - that's a pretty significant accomplishment (I'm also very proud of the wins and I enjoy them very much).

Even so if I can help I'd love to!

Anyway we'll see what David has to say about what I provided earlier today.

Best wishes!

[xexyl]

@lcn2 Wikipedi - I have update already the contest dates on the page and the changes accepted/visible. I'll be looking further to see how I can work in the newly found references. It would be useful to know parts of the page to focus on.

We are not allowed to modify the IOCCC page.

That's one of the things I hate about wiki's politics. That last word can be derogatory but it's the right word for them. They even have a page talking about how they're not a reliable source! There is the infamous story where an author tried to correct information about a book HE wrote (hence 'author' :-) ) and they undid the change. If anyone would know the book I think the author would. I have seen medical articles with serious - in fact deadly - false statements.

Unfortunately some people rely on it first and I could just imagine someone doing what they read there only to lose someone (or whatever the case is).

Still in this case it might be nice to update indeed.

We would offer this general advice: read the IOCCC Wikipedia page and ask yourself:

• Are there factual errors that need to be corrected?

There is one on your page last I knew - a funny one, one I shared with you before. They claim you're an amateur astronomer! Astronomer yes but not amateur!

• Are there things that, due to omission, that present the IOCCC in a misleading way?
• Are there ways to add more content to the page such as going into the history, or changes to the IOCCC that could be documented?

One of the craziest things I have seen there, and I also shared this with you (you were as stunned as I was): there is a screenshot or else copy of someone running an IOCCC entry as root (!!). The amount of people who stay root in case they'll need it, it's astonishing.

For example, the introduction to the iocccsize tool? .A table that talks about how Rule 2 size limits have changed over time. Etc.

Perhaps that could link to the FAQ item?

Do we want anything about the mkiocccentry toolkit? That seems maybe useful - but I don't know wikipedia organisation for pages and I don't know what they would want - or would reject. I could see them rejecting some of this.

UPDATE 0

• Maybe Wikipedia could have sub-pages for each of the IOCCC contests instead of displaying a few examples of winning code?

That's an interesting idea. I wouldn't know how they would want it structured though. Thankfully for me I'm not the one doing the editing :-)

• A table of locations where winning authors are located, by country?

Perhaps this would be a link back to the website? Since what we two did, the JSON parser and the website scripts, it's all automated now.

• Top winning authors, etc.

What do you mean by top? Obviously Yusuke has to be there but besides that I could see it being arbitrary, unless it's simply by count. But then that number would be arbitrary in a sense.

I'm not sure if having winning authors is necessary but it's an interesting idea anyway and I'm not trying to say yes or no there. I just wonder if it should stay more about the contest itself? Just a thought.

[xexyl]

@xexyl No worries, take it easy, relax, rest, you deserve it. It will be waiting when you get back. Ping me when you are ready until the enjoy some much earned R&R.

I have solved the problem! It was as I guessed - the need to rejoin. Thanks @Alchemist-Production!

[xexyl]

@xexyl No worries, take it easy, relax, rest, you deserve it. It will be waiting when you get back. Ping me when you are ready until the enjoy some much earned R&R.

I have solved the problem! It was as I guessed - the need to rejoin. Thanks @Alchemist-Production!

You'll be happy to hear, @SirWumpus, that I posted a couple comments, one in reply to you about Slashdot, one about sources of the IOCCC (or places referencing it).

Probably won't do much more beyond that but ... since Landon said there will be something I might be interested in I'll have to check once I know what that is and where it is. Which brings me to my next comment when I can find the right one to reply to.

[xexyl]

There is also another project that I believe you will be interested in where the discussion will also occur on discord. So PLEASE work with David to restore your access.

I got that sorted. It was as I guessed. But please, @lcn2, let me know when you posted something about the project and a link to the post. I'm not likely to see it otherwise. Too much chatter is way too overwhelming to me. Plus with all the threads it's easy to miss.

As I think I probably told you before even at family gatherings (when we used to have them) I would be in a room all by myself (or maybe just with my now late grandparents) most of the time. Never liked crowds and it's fair to say I have ochlophobia as well. I also have claustrophobia and agoraphobia. Those might not seem relevant online but they are in ways I am too tired to explain. Point is I tend to stay away from communities even if it's one with a shared interest of mine.

Thanks and enjoy your holiday!