Pattern of the Year 2015 (Votes)

[Kazyan]

7, 9, 14, 13, 15, 17, 16, 4

I know, that's a surprising order, but to me, new tools being used to build new tools says "pattern of the year" louder than the individual new parts themselves.

[muzik]

15, 6

[triller]

15

While the bulk of accomplishments this year are interesting to me as curiosities, the Simkin Gun has direct impact in my particular bent, that of p30 circuitry. Its side shooting action makes it a great fit in some optimizations.

triller

[Alexey_Nigin]

15, 16, 14, 13, 6, 3, 8, 2, 4

My site now has the "official" countdown to the deadline. I will be asleep when it reaches zero, so I would ask somebody to go ahead and publish "The competition is over" using huge font.

You cannot edit your votes after the deadline.

The results will be posted in a few days.

I'm sorry for not answering all your questions, but be sure I will do it in the near future.

[drc]

15, 16, 14, 13, 6, 3, 8, 2, 4

My site now has the "official" countdown to the deadline. I will be asleep when it reaches zero, so I would ask somebody to go ahead and publish "The competition is over" using huge font.

You cannot edit your votes after the deadline.

The results will be posted in a few days.

I'm sorry for not answering all your questions, but be sure I will do it in the near future.

I'll do it. Also, what exact time does it end?

[Kiran]

Ending in 1 hour!

[Kiran]

It ended!

I'll do it.

You failed!

[drc]

It ended!

I'll do it.

You failed!

I have a life you know

[Kiran]

I have a life you know

http://rationalwiki.org/wiki/Argumentum_ad_cellarium

Should an exception be made to the deadline to allow Extrementhusiast to revise that vote (it does not have the proper form)?

[calcyman]

I have a life you know

http://rationalwiki.org/wiki/Argumentum_ad_cellarium

Or, even more accurately, argumentum ad cellularium.

[drc]

Oh great, everybody's turning on me.

[Extrementhusiast]

I have a life you know

http://rationalwiki.org/wiki/Argumentum_ad_cellarium

Should an exception be made to the deadline to allow Extrementhusiast to revise that vote (it does not have the proper form)?

I thought I revised it already.

[Alexey_Nigin]

Should an exception be made to the deadline to allow Extrementhusiast to revise that vote (it does not have the proper form)?

Don't be so fastidious! That missing line break you are referring to will not cause any problems.

[Kiran]

(13, 14, 15), (16, 5, 11), (9, 6, 8, 17), (1, 7, 10), (3, 12), (4, 2)

So this will work?
I assume you will turn it into this:
13,14,15,16,5,11,9,6,8,17,1,7,10,3,12,4,2

[Alexey_Nigin]

1. Syringe (107 points)
2. Simkin Glider Gun (102 points)
3. Demonoids (91 points)
4. Syntheses of Spaceships (55 points)
5. Recursive Filter (48 points)
6. G-to-Weekender (45 points)
7. Bob Shemyakin's Syntheses (34 points)
8. HBK Gun (33 points)
9. H-to-MWSS (31 points)
10. Sawtooth 177 (30 points)
11. Single-Lane Construction Arm Toolkit (29 points)
12. Elementary Conduits Collection (28 points)
13. [L|M]WSS-to-G (26 points)
14. Collection of Guns (23 points)
15. Dragon Lightsaber (19 points)
16. Quadratic Sawtooth (14 points)
17. t*log(log(t)) Growth (13 points)

[Alexey_Nigin]

In order to generate that ugly list, I organised the entries and the votes into two separate text files.

Entries.in

Code: Select all

Elementary Conduits Collection
Quadratic Sawtooth
Recursive Filter
Sawtooth 177
H-to-MWSS
HBK Gun
Collection of Guns
G-to-Weekender
Bob Shemyakin's Syntheses
Dragon Lightsaber
[L|M]WSS-to-G
t*log(log(t)) Growth
Syringe
Demonoids
Simkin Glider Gun
Syntheses of Spaceships
Single-Lane Construction Arm Toolkit

Votes.in

Code: Select all

drc               11, 15, 8, 16
dvgrn             13, 8, 5, 3, 2.
biggiemac         13, 15, 14, 3, 16, 8, 1, 2
Scorbie           14, 13, 15, 17, 3, 9, 10, 4, 1, 16, 7, 5, 2
A for awesome     13, 7, 1, 15, 5, 11, 16
calcyman          15, 13, 4, 17, 16, 3
Saka              8, 14, 6, 3, 2
Kiran             15,6,8,14,4,10,13,3,5,2,12,11
gameoflifeboy     3, 17, 14, 12, 15, 2
isaacg            13, 15, 14
codeholic         15,13,14
Extrementhusiast  (13, 14, 15), (16, 5, 11), (9, 6, 8, 17), (1, 7, 10), (3, 12), (4, 2).
AwesomeFace       15, 13, 12, 4, 8
mniemiec          16, 14, 9
Sokwe             16, 13, 8, 14, 5, 4, 15, 6, 9, 11
wildmyron         14, 13, 15, 3, 16.
chris_c           13, 15, 16
simeks            13, 14, 1, 3, 15, 11, 9, 16, 2, 12, 4
Kazyan            7, 9, 14, 13, 15, 17, 16, 4
muzik             15, 6
triller           15
Alexey_Nigin      15, 16, 14, 13, 6, 3, 8, 2, 4

These files were fed into the following program:

Judge.dpr

Code: Select all

//  No parameters are hardcoded, so this program can be used for
//  any competition with the same voting system. The inputs are
//  contained within "Entries.in" and "Votes.in". These two
//  files and the program should reside in the same folder.

program Judge;

{$APPTYPE CONSOLE}

uses
  SysUtils, Classes;

var
  Entries,Votes,Outcome : TStringList;     //  Some array indices start from
  i,T1,T2 : Integer;                       //  zero, some from one, hence
  A : array of array [1..2] of Integer;    //  the confusion. For more
  RepThis : Boolean;                       //  info on the topic, see
                                           //  https://xkcd.com/163/

procedure Canonize(var S : String);
var
  i : Integer;

begin
  for i:=1 to Length(S) do
    if not CharInSet(S[i],['0','1','2','3','4','5','6','7','8','9']) then
      S[i]:='.';
  for i:=Length(S)-1 downto 1 do
    if (S[i] + S[i+1]) = '..' then
      Delete(S,i,1);
  if S[1] <> '.' then
    S:='.' + S;
  if S[Length(S)] = '.' then
    Delete(S,Length(S),1);
end;



procedure AddVote(S : String);
var
  i,Sum : Integer;

begin
  Canonize(S);
  Sum:=0;
  for i:=Length(S) downto 1 do
    if S[i] = '.' then
      begin
        Inc(Sum);
        Inc(A[StrToInt(Copy(S,i+1,Length(S)-i))-1,2],Sum);
        Delete(S,i,Length(S)-i+1);
      end;
end;



begin
  try
    { TODO -oUser -cConsole Main : Insert code here }
    Entries:=TStringList.Create;
    Entries.LoadFromFile('Entries.in');
    Votes:=TStringList.Create;
    Votes.LoadFromFile('Votes.in');

    SetLength(A,Entries.Count);
    for i:=0 to High(A) do
      begin
        A[i,1]:=i+1;
        A[i,2]:=0;
      end;

    for i:=0 to Votes.Count-1 do
      AddVote(Votes[i]);

    repeat    //  Bubbles are cool!
      RepThis:=False;
      for i:=0 to High(A)-1 do
        if A[i,2] < A[i+1,2] then
          begin
            T1:=A[i,1];
            T2:=A[i,2];
            A[i,1]:=A[i+1,1];
            A[i,2]:=A[i+1,2];
            A[i+1,1]:=T1;
            A[i+1,2]:=T2;
            RepThis:=True;
          end;
    until not RepThis;

    Outcome:=TStringList.Create;
    for i:=0 to High(A) do
      Outcome.Add('[size=' + IntToStr(((High(A)-i)*100) div High(A) + 100) +
       ']' + IntToStr(i+1) + '. ' + Trim(Entries[A[i,1]-1]) + ' [i](' +
       IntToStr(A[i,2]) + ' points)[/i][/size]');

    Outcome.SaveToFile('Outcome.out');
    WriteLn('Everything is OK.');
    ReadLn;
  except
    on E:Exception do
      Writeln(E.Classname, ': ', E.Message);
  end;
end.

This is all I have for now. I will hopefully answer your questions tomorrow.

[Alexey_Nigin]

Yesterday, I think I saw a "vote" tab next to the "upload assignment" tab. I don't see it now. Did anyone spot that? Did I see just a testing beta or what was that??

That's not my work; actually, I know nothing about it.

These don't really have a place in the "pattern of the year" entry but Adam's work definitely fits my "this pattern/accomplishment helped make 2015 a groundbreaking year" criterion so I'll say it anyway.

Agreed.

If the voting system hasn't been solidified yet, I think that the first pattern in each vote should be given 1 point, the second should be given 2/3, the third should be given 2/3, etc. The nth should be given S^-1, where S is the partial sum of the first n terms of the harmonic series. I think this method is close enough to Zipf's law to be a semi-accurate representation of how much people favor patterns relative to each other, while decreasing slow enough to account for the fact that some people might have more than one favorite pattern and be forced to tie.

I honestly don't know which system is better, but I generally oppose runtime changes to the rules. For example, they are occasionally used to cheat at some math olympiads: "look, one guy solved problems 1,2,3, the other guy solved 1,2,4, and the first guy (is from respected school/has good history/participated in my paid courses) so let's value 3 more than 4." It was of course not the case with this competition, but the idea still has some bad flavour. I do not oppose using your system for the next competition.

Could you make the context here more explicit? Did you (plural) publish something?

Adam asked Ivan to send prize money to charity, which Ivan did.

#1: I think Chris Cain made the most significant recent contribution to the conduit collection by writing a script to compile it.

You linked to Herschel thread, while #1 is about the results from Elementary thread, right?

#7: "scote" is Scot Ellison.

Amended.

----------------

And now that everything is over, it is probably time to start the Pattern of the Year 2014 competition. But... maybe this workload can be put on someone else's shoulders?

[Kiran]

And now that everything is over, it is probably time to start the Pattern of the Year 2014 competition. But... maybe this workload can be put on someone else's shoulders?

I think I can do it.

[dvgrn]

And now that everything is over, it is probably time to start the Pattern of the Year 2014 competition. But... maybe this workload can be put on someone else's shoulders?

I think I can do it.

A new 2014 Entries thread could start with this starter list of entries for 2014.

However, all this nominating and vote-counting is hard work, and seems a little less urgent somehow when the title has to be awarded retrospectively. Seems like we might not get much participation.

Here's a much simpler voting system:

The Provisional Pattern of the Year 2013 is the Snark.

The Provisional Pattern of the Year 2014 is the waterbear.

The Provisional Pattern Collection of the Year 2014 is the complete 16- and 17-cell still-life synthesis collection. (Collections seemed to need a separate category, since it's really hard to compare them with single patterns, except maybe by some kind of I*{total-inspiration}+p*{total-perspiration} metric.)

If nobody takes issue with any of these enough to start a new thread and collect official entries and votes for 2013 and/or 2014, then that must mean that everybody is in agreement... or not reading this thread, but that's not any different from the 2015 vote really.

In either case, after some arbitrary deadline passes -- let's say Star Wars Day 2016 -- we could remove the "Provisional" prefix and consider the matter closed.

-- Obviously if people are really interested in mulling over 2013/2014 B3/S23 Things Of The Year, then my suggestion is not a serious one! It just doesn't seem like it will be the same kind of competition as the 2015 one.

[simsim314]

I know I'm late, but I want to share some thought of my own on this topic. Because I'm not limited to the rules of the voting I want to widen the scope a bit to include forum threads and search utilities.

#1. The Hunting of the New Herschel Conduits thread. This thread has encouraged more novelties than any other thread in years. New utilities were written during it (CatForce), and old utilities were fixed and improved (ptbsearch, bellman). Bellman was used extensively, and realized much more of its potential since 2013 when snark was found, but never extensively used in conduit search.

In this thread alone we get: 1,5,11,13,15.

And I would place 13 in front of 15, because fast G->H was very desired and missing component for a long long time. But I think they all part of the same collaborative effort.

#2. Demonoids. As I consider Gemini one of the most beautiful design patterns of all times, Demonoid 0hd brings the best of them all into single neat and spectacular self replicating spaceship.

Why I like Demonoid 0hd:

1. It's the smallest designed ship of all times. Compare it to the original Gemini: its replication cycle is 75 times faster, and the cell count is 30 times smaller. This is reduction of almost two orders of magnitude (!). I can't think of any design pattern that was improved by so much. It looks more like Geminoid from GeminoidParticles rule, than Gemini.

2. It's the first self constructing ship that uses latest novelties. Eater2 synth kit, synthesizable syringe, correctly utilizes self destruct, and uses 0hd recipes kit. It's basically well executed, the edge of the current technology ship. This ship just couldn't be done in the past year.

3.On top of this all, the 0-hd uses doubled tape trick, it's amazing and very simple idea, which deserves special exclamation mark.

#3 Synthesis effort. There is no single thread for the synth effort, but it's undoubtedly one of the most developing areas, and is very challenging. After the dart synth breakthrough in 2014, many other spaceships synths were found, and Bob's collection is also very impressive.

This area of research is amazing, considering everything is still done by hand, using knowledge experience and intuition, and no real search utility exists to simplify significantly the effort.

-----

All in all I basically agree with the collective mind assessment, except I would place the [L|M]WSS-to-G and H-to-MWSS somewhat higher.