Discussion on question by caird coinheringaahing: One OEIS after another

Feeds

02:49

70

This is an answer chaining question that uses sequences from OEIS, and the length of the previous submission. This answer chaining question will work in the following way: I will post the first answer. All other solutions must stem from that. The next user (let's call them userA) will find the...

Brian J

Man this is hectic at first. I've made code for three sequences and then by the time I've figured them out someone beat me to it :)

Stephen

For future reference: if at all possible, please try to leave your bytecount the same after posting. If you have to edit, please do so quickly.

BlackCap

Can we freely reuse languages an unlimited number of times after 150 answers?

Lynn

can we intentionally pad the byte count to give the next person in line a hard time?

"You can assume that neither the input nor the required output will be outside your languages numerical range, but please don't abuse this by choosing a language that can only use the number 1, for example." ← Is assuming, say, 32-bit ints and hard-coding for that "abusing" this rule? (I think it's a banned common loophole, but this sounds a little like you're excepting it!)

Nathan Merrill

Why is the scoring based off of the second-to-last vs the last? Don't you want people to pick hard sequences that are difficult to write?

Stephen

02:49

@NathanMerrill you want them to pick ones that are hard, but not impossible, to write. If it was last, it would be over soon, and that would be no fun.

Nathan Merrill

@StepHen I'm not convinced that there are impossible ones.

caird coinheringaahing

@NathanMerrill A000017 is close to impossible, if you take a look at it.

@Lynn you can pad the source code if you want, for whatever reason you want. Check the 20th answer as an example. And as for the integer limit, it's just poorly worded. It is essentially the standard loophole: "don't use Boolf*ck or languages like it."

Stephen

@cairdcoinheringaahing no, it's not impossible - the sequence only goes up to a certain amount, so you can hard code it up to that, and since the sequence doesn't go any higher, anything after that is undefined behavior (at least in my opinion)

ETHproductions

@StepHen The ranges are rather nice, thank you for that.

Stephen

@ETHproductions no problem, it was getting a bit lengthy, although this might cause some people to miss some if they Ctrl-F it or something

R. Kap

02:49

I hope my edit to the stack snippet was acceptable and not too drastic. Again, feel free to rollback if desired. Also feel free to improve it in any way deemed fit. :)

@BlackCap Nice! I was going to add the functionality of sorting by clicking on the labels, but it seems you have beat me to it. :)

BlackCap

@R.Kap I think that is a good compromise, because then you still have the ability to skim trough ranges for unused bytes. Thanks for adding the search, I think we needed that badly :)

Do you realize that trumscript cannot handle numbers smaller than one million?

Scott Milner

@BlackCap Maybe we could allow 6 trailing zeros?

caird coinheringaahing

@BlackCap yes, I've coded in it. However, you can enter 1000001 - 1000000 to get 1.

ppperry

If somebody posts a solution that's longer than the number of OEIS sequences, and the next person picks the lowest unused one, can somebody else later use that number of bytes?

Brian J

Building on @Lynn's comment about input/outside within your numerical range: can we assume that if the formula fails only because intermediate steps exceed our numbers integer size, but would work if we had a bigger integer size, then the code is valid?

NieDzejkob

02:49

@cairdcoinheringaahing How do you do input in TrumpScript?

ppperry

@NieDzejkob VARIABLE hear

NieDzejkob

@ppperry Oh, nice. Where did you find it? There is no mention of it in the readme.

ppperry

@NieDzejkob I looked at the example for prime testing.

Scott Milner

Does Python w/ SymPy count as a separate language? codegolf.meta.stackexchange.com/a/7246/67881

Peter Taylor

@ScottMilner, that would make the entire question pointless, because it would create a loophole the size of an aircraft carrier through the non-repetition of languages restriction. Any language with 150 libraries could be used for every single answer, and there are way more than 150 libraries in CPAN or NPM.

NieDzejkob

02:49

I think I broke the snippet. Could it be caused by the emoji? Can you investigate?

caird coinheringaahing

@NieDzejkob I've edited your answer and it works now.

NieDzejkob

For some reason the snippet does not show any answer with 17 bytes but says it's taken...

caird coinheringaahing

@NieDzejkob yes, that is weird... I'll alert ETH, see if he can find the problem

NieDzejkob

@cairdcoinheringaahing just a guess, but maybe some deleted answers shenanigans?

@cairdcoinheringaahing I found it. It's not a bug, it's a feature. Beautify the javascript using jsbeautifier.org, line 89. A000017 is uncomputable so it was added as an exception.

mbomb007

@NieDzejkob There are a lot more (> 1400) erroneous sequences than that one. dead OEIS sequences

NieDzejkob

02:49

@mbomb007 don't ask me. I didn't add that, but in these cases the sequence is defined as hardcoded numbers, so I don't see why it should be blacklisted

caird coinheringaahing

@NieDzejkob It was blacklisted before people realised that hardcoding was allowed, as it was thought that the sequence was incomputable and therefore wouldn't be fair to make that the sequence.

mbomb007

@cairdcoinheringaahing I think you meant to say that hardcoding isn't allowed.

caird coinheringaahing

@mbomb007 no, hardcoding is allowed, but discouraged quite a lot

Jonathan Frech

Are C (gcc) and C (clang) considered to be different languages or are they too similar?

totallyhuman

If gcc and clang are considered the same, then are CPython and PyPy? 'Cause they have been used before...

caird coinheringaahing

02:49

@icrieverytim Good point. In light of that, I'm going to change my previous decision and say that clang and gcc are different languages.

NieDzejkob

@cairdcoinheringaahing I think that a better solution might be to ban different compilers/interpreters, but to grandfather in the answers made before the rule change. This allows 6 Python answers, 4 C answers, 2 C++ answers and 3 C# answers every 150 (10 percent), which is, in my opinion, a little too much, in a challenge that is supposed to be about learning various languages.

caird coinheringaahing

@NieDzejkob Defining different languages is a tough decision. I've already changed my mind once, and changing the rules would invalid around 10 current answers, which isn't a good idea. While that does allow 15 "different" languages which are very similar, that still means that there are 135 different answers that must use a new language. While it isn't a perfect answer to this problem, I think it is best to not invalidate 13 answers. I'm keeping the current rule.

NieDzejkob

@cairdcoinheringaahing I think you missed a large part of this proposition. This should only affect new answers - invalidating answers isn't cool. One thing you might use in the definition of different languages is that the different interpreters/compilers aim to be identical.

I think Seed would be a good candidate for a bounty language.

caird coinheringaahing

@NieDzejkob Agreed, but why do I get the feeling that you're working on an answer in Seed for the next sequence?

NieDzejkob

@cairdcoinheringaahing look at what the next sequence is, then look at what is the juice of Seed after you get rid of all the layers of complicated RNG. However, if I manage to crack the Mersenne Twister and a sequence feasible for Befunge shows up, you can be sure there will be a juicy explanation.

So ,,, broke the snippet, somewhat.

caird coinheringaahing

02:49

@NieDzejkob ,,, always breaks snippets. I'll edit the answer to fix it

Engineer Toast

I feel like this has become the ultimate "code-golf this OEIS sequence" challenge. We'll never cover them all, of course, but 222 in a row (as of this posting) is a sizable chunk for a single challenge.

caird coinheringaahing

@EngineerToast who says we can't get them all? :P

Engineer Toast

@cairdcoinheringaahing I'm not saying it's impossible but it's highly unlikely. As of right now, OEIS has 293,235 sequences. Per the FAQ, they get about 30 new ones per day. This question has averaged about 2.7 answers per day since it was posted. We'll have to increase our efforts by more than tenfold just to keep up with new submissions, much less cover the old ones. At the current rate, it'll take about 294 years to recreate all the sequences already in the database.

Post frequency per day as of 2017-10-16T11:40Z. Less helpful graph of the same data but sorted by posts per day instead of by date.

caird coinheringaahing

@EngineerToast Yeah, that looks about right :P

Transcript for

Discussion on question by caird coinh…