Join the new world

This article is in both English and Romanian. The article is quite lengthy, so I have marked the most important paragraphs with ***. Please vote it so that it gets international attention! Enjoy!

Acest articol este in engleza si romana. Fiind destul de lung, am marcat cele mai importante paragrafe cu ***. Va rog sa il votati pentru a ajunge in mass-media internationale. Lectura placuta!


The wonderful V1 platform allows for much easier redesign of layout, formulas and gameplay from the part of admins. Therefore, in the past few days we were offered a couple of variants for the Congressional elections, but both of them shows that, even though the developers cheered for democracy whenever they had the chance, still now they are serving frozen authoritarianism.

*** Why is the actual system proposed by the last article “Division of congress members” (http://www.erepublik.com/en/article/division-of-congress-members-694342/1) not exactly the best of options? Most remarkably, we have a typical case of “taxation without representation”. In systems of this sort, especially in countries with a high number of regions, it will be like having 40 (or more) presidential elections, in which the greatest party is by far the most likely to get all the spoils. In this case, the Top 5 system is useless – even Top 3 should do – the Congress will no longer hold a fair distribution of seats between the first 5 parties, but a representation of the top party and a handful of lucky left-overs (aka wild-cards) for the runner-up. I do not want to get very much into details here, as the results of the next Congress elections will speak for themselves and everybody will see the flaws of this system. The worst part of it is that the biggest countries will be even more devastatingly affected.

From my point of view, it is obvious that there is a problem, as the purposes and wishes of the admins and those of the citizens are not reached. Without claiming that my research is “scientific”, I have drawn together what I think these wishes are. This is the result:
- a system that is more fair and democratic – a more correct representation given the correlation between percentages per party and the number of seats in Congress (a) OR a more correct representation given the correlation between regional populations and the number of seats every region gets (b) – because any of these two alternatives is normal and users want that;
- greater significance for regions (some sort of substitution for the mayoral system);
- uninominal or pseudo-uninominal elections;
- scalability, now or in V1.1;
- not a revolutionary change (as far as I understand from the aforementioned Erepublik Insider article).
Should the assumptions above stand, I have at least two better systems that could be implemented.
For case (a), Peregrine has already proposed a quite versatile system as I remember.
* the simplest of the variances is two-phase voting, first window for the party and second window for the man in that party; then the distribution of the places in Congress is done partywise by the first vote and then memberwise, according to the positions obtain by each candidate inside the party and observing the limit of seats ascribed to the party in the first window vote (the Simple Proportional Representation system in real life).

* a bit more complicated, but easier to fine-tune would be the “Highest Average” variance, in which the number of seats in congress one party can obtain in relation to the percentage of votes received can be altered (admins could, for example, choose to balance the political play a bit, by disadvantaging the bigger parties by one or two seats in Congress) by a very-well defined algorithm – I could go into further detail if you are interested. The way of voting is identical to the one described above, but the method of distributing seats among parties is slightly changed.

From my point of view, this would be by far the fairest of systems, being readily scalable (only one piece of data is to be changed for each country – or even only one for all countries as a whole) whenever the surge in user count will demand for it – the only problem being that it does not involve regions in any way.

For case (b), there is the initial version of the admins (http://www.erepublik.com/en/article/congressional-elections-improvements-693809/1), which I could not grasp in its entirety.
* As far as I can understand, depending on the population in one region, every region would get from 1 to 10 spots in congress. A citizen could only vote for a candidate in their own region, and with respect to the population in that region, the first x of them would get into Congress. Even though this is not an entirely pluralistic system (as I understood after having read the article twice), in most regions, especially now, when we have only few citizens in most of the regions, the system could bring immense advantage to larger parties and terribly discourage the smaller ones, not to mention the absolute impossibility for the parties that are not in the top 5 of obtaining spots in the congress (that if we assume the admins will ever drop the 5 parties threshold). Therefore, especially now and perhaps in the following year, this system is not exactly the best.

The conclusion would be that while the former system is fair, it does not involve regions (no fun), the latter is not very proportional (no democracy). Therefore, my actual proposal for the electoral system looks like this:

*** (c) Regional lists are created – ideally by all (not only the first 5) parties –, by the principles the admins presented. With the entire preliminary period, with the candidacy intention link etc. Voting will be the same as I have described it at point (a): the voter choses a party and then the individual candidate (note: we could have more than one version here: (c.1) – the voter can chose only among that party's candidates; or (c.2) – the voter can choose among all candidates in the region, regardless of their party. (c.2) would be the best, to my mind, especially if they abandon the small party discrimination). The distribution of seats in Congress is the most interesting part of this system. The Congress would be split in two - say, 60% for regional seats and 40% for “national” seats. The first to be filled are the regional spots, taking into account for this phase only the votes received by individuals, the hierarchy generated by that inside of regions and the number of seats ascribed to that region. The other 40% (or whatever the proportion may be, it may be 30 or 50% just as well) are distributed to the parties, and from here on the possibilities are countless. The most applicable two versions are: (c.i) – depending on the distribution of regional seats in Congress for each party, the global seats are distributed in such way that, in the end, the percentages of seats in congress correspond as accurately as possible to the vote percentages obtained on the overall by each party (e.g.: there are 100 seats in Congress. Overall, party X has received 65% of all the votes, whereas party Y 35%. The votes have been distributed in such way that, of the 60 “regional” seats in Congress, 50 will belong to party X and only 10 to party Y. The rest of 40 seats are to be distributed in such a way that the percentage of seats in Congress is in sync with the percentage of votes, so party X receives additional 15 global spots, whereas party Y gets the remaining 25. The final seat distribution is X=50+15=65 and Y=10+25=35); or (c.ii – the global votes are distributed regardless of the regional-individual results (in our example, after party X got the 50 seats in Congress because of regional voting, it receives additional 40*65%=26 seats, whereas party Y 40*35%. The final seat distribution is X=50+26=76 and Y=10+14=24. It is to be noted that, in our example, should all the seats were to be distributed on the basis of individual votes, the final distribution in Congress would have most likely been X=83 and Y=17.) Distribution of the party (global) seats to their members can be done by the number of votes received (considering that the distribution of places in Congress is made in such a way that entering in the Congress by regional vote in a smaller region is easier) by the candidates that have not made it there in the first round.

*** The final conclusion is that this hybrid system falls well with both the developers and the users, according to the wish list established at the beginning of this article, it is very easy to fine tune (slightly harder to describe, I must admit, but not much harder than the new productivity formula), distinctively modern (I will have to undertake research in order to see the name and the few countries – if any – where this system is employed), scalable – the number of seats can be triggered by the national population, number of regions and regional population – and it both satisfies the admins and the users. I would be more than happy to go into further detail or more complex examples on this last system should you wish. I feel the need to say that my best preferred version is c.2.i.

Thank you for reading!



Minunata platforma V1 permite adminilor sa faca schimbari in design, formule si gameplay mult mai usor decat in beta. De aceea, in ultimele cateva zile ni s-au oferit 2-3 variante pentru alegerile din Congres, insa fiecare din aceste variante ne arata ca, desi de fiecare data cand au avut ocazia, developerii au preaslavit democratia, acum ne servesc dictatura.

*** De ce nu e bun sistemul actual propus in ultimul articol Division of congress members (http://www.erepublik.com/en/article/division-of-congress-members-694342/1)? Avem un caz tipic de, dupa cum ar spune americanul, "taxation without representation". In astfel de sisteme, mai ales in tarile cu relativ multe regiuni, este ca si cum ai avea 40 de alegeri prezidentiale, unde partidul mare ia tot. In acest fel, top 5 este justificat - probabil chiar si un top 3 ar fi la fel de bun - congresul nu va mai fi format pe baza impartirii de locuri intre primele 5 partide, ci doar in partidul dominant si ceva resturi norocoase din al doilea partid. Nu vreau sa intru in detalii aici, daca va ramane acest sistem cel mai probabil va veti da seama la urmatoarele alegeri cat de gresit este conceput. Partea cea mai proasta este ca tarile cele mai mari vor fi si cele mai afectate, dar asta veti vedea abia atunci.

Din punctul meu de vedere e clar ca avem o problema, intrucat obiectivele si dorintele adminilor si userilor nu sunt atinse. Fara pretentia de a fi facut o cercetare "stiintifica", am sintetizat ceea ce cred eu ca reprezinta aceste obiective. Iata ce am obtinut:
- un sistem mai fair si mai democratic - o reprezentare cat mai corecta in corelatia procentaj/partid -> numar de locuri in congres (a) SAU o reprezentare cat mai corecta in corelatia dintre populatia din regiune -> numar de locuri in congres (b) - pentru ca oricare varianta e normala si userii vor asta;
- o importanta crescuta a regiunilor (substitut pentru primari);
- alegeri uninominale sau macar pseudo-uninominale;
- scalabilitate, acum sau in V1.1;
- o schimbare cat mai mica, dupa cum inteleg din articolul anterior din Insider.

In cazul in care premisele acestea sunt adevarate, atunci am cel putin doua variante care ar trebui sa mearga.

Pentru cazul (a), a fost deja propusa o varianta pe care am retinut-o, pe marginea careia s-ar putea construi ceva variatii.
* cea mai simpla varianta este votarea in doua faze, intai partidul si apoi omul din partidul respectiv, distributia locurilor in parlament se face catre partide pe baza primului vot, iar apoi catre candidati, in functie de procentajele obtinute de fiecare in cadrul partidului si in limita locurilor arondate de prima faza a alegerilor (sistemul Peregrine, cunoscut in real life ca Simple Proportional Representation).
* un pic mai complicata, dar mai usor de tunat, ar fi varianta "Highest average", in care se poate umbla la procentul de locuri in congres pe care un partid le poate obtine in relatie cu procentul de voturi obtinute (adminii pot alege sa avantajeze/dezavantajeze cu un loc-doua partidele mari in detrimentul celor mai mici, de exemplu) in urma aplicarii unui algoritm - daca intereseaza pot detalia. Modul de votare este exact la fel, se schimba (foarte putin) metoda de distribuire a locurilor intre partide.

Din punctul meu de vedere acesta ar fi de departe cel mai corect sistem, usor scalabil (este necesara schimbarea unei singure variabile pentru fiecare tara - sau chiar uneia singure, globale - atunci cand cresterea de populatie o va cere) - singura problema este ca nu implica in niciun fel regiunile.

Pentru cazul (b), exista varianta initiala a adminilor (http://www.erepublik.com/en/article/congressional-elections-improvements-693809/1), dar nu am inteles foarte bine cum functioneaza.
* Din cate imi dau seama, in functie de populatia regiunii, fiecare regiune avea sa primeasca intre 1 si 10 locuri in congres. Un cetatean putea vota doar pentru candidatii din regiunea proprie, iar in functie de ponderea populatiei regiunii respective, primii x intrau intr-un congres largit. Chiar daca nu era un sistem totalmente pluralistic (asa cum intelesesem dupa primele doua lecturi), in cele mai multe regiuni, mai ales acum, cand sunt destul de putini cetateni, ar fi fost first past the post in majoritatea regiunilor, sistem care ar fi putut avantaja covarsitor partidele mari si descuraja teribil pe cele mici, ca sa nu pomenesc de imposibilitatea absoluta pentru partidele care nu sunt in primele 5 de a obtine vreun loc in congres (asta presupunand ca adminii vor renunta vreodata la limita celor 5 partide). De aceea, mai ales acum si poate cel putin in urmatorul an, sistemul acesta nu e cel mai bun.

Concluzia pana aici ar fi ca primul sistem este corect, dar nu implica regiunile (no fun), iar cel de-al doilea nu e foarte democratic (no proportionality). De aceea, propunerea mea pentru sistemul electoral pentru Congres arata in felul urmator:

*** (c) Liste regionale sunt create - eventual de catre toate (nu doar primele 5) partidele - , dupa principiile expuse de admini. Cu tot cu perioada aceea preliminara, cu linkul catre prezentarea candidaturii etc. Votarea se face ca in modelul descris de mine la punctul (a):  votantul alege un partid, apoi candidatul individual (cu mentiunea ca aici exista o discutie, sunt doua variante: (c.1) - poate alege doar dintre candidatii partidului; sau (c.2) - poate alege dintre toti candidatii din regiune, indiferent de partid. Varianta (c.2) este recomandabila, dupa parerea mea, mai ales daca se renunta la "discriminarea" partidelor mici). Distributia locurilor in congres este partea cea mai interesanta. Congresul se imparte in doua (sa zicem 60% pentru locurile regionale si 40% pentru locurile "nationale"). Primele se ocupa locurile regionale, luand in considerare numai voturile primite de indivizi, ierarhia obtinuta in regiuni si numarul de locuri alocate regiunii respective. Celelalte 40% (sau care va fi proportia, poate fi 30% sau la fel de bine 50😵 se distribuie partidelor, iar de aici posibilitatile sunt nenumarate. Cele mai aplicabile doua variante sunt: (c.i) - in functie de distributia locurilor regionale din congres pe partide, locurile globale se distribuie astfel incat, la final, procentele de locuri in congres sa corespunda cat mai bine procentelor de voturi primite in total de fiecare partid (e.g.: sunt 100 de locuri in congres. Per total, partidul X a obtinut 65% din voturi, iar partidul Y 35%. Voturile au fost distribuite astfel incat, din cele 60 de locuri regionale din congres, 50 au ajuns la partidul X si doar 10 la partidul Y. Restul de 40 de locuri se vor distribui astfel incat sa se ajunga la echilibrarea numarului de locuri in acord cu procentul de voturi, astfel ca partidul X mai primeste doar 15 locuri globale, in timp ce partidul Y restul de 25. Distributia finala de locuri: X=50+15=65, Y=10+25=35); sau (c.ii) - voturile globale se distribuie indiferent de rezultatele regional-individuale (in exemplul nostru, dupa ce partidul X a obtinut 50 de locuri in congres datorita voturilor individuale, partidul X mai primeste inca 40*0.65=26 de locuri, in timp ce partidul Y 40*0.35=14 locuri. Distributia finala de locuri: X=50+26=76, Y=10+14=24. De remarcat ca, in exemplul dat, daca toate locurile se acordau pe baza votului individual, distributia finala a locurilor in congres ar fi fost, cel mai probabil, X=83 si Y=17.) Distributia locurilor partidelor ("nationale") catre membrii acestora se poate face pe baza numarului de voturi obtinute de acestia (cu conditia ca distributia locurilor in congres sa se faca in asa fel incat intrarea in primele 60 de procente dintr-o regiune mai slab populata sa fie mai usoara - asta se face prin ajustarea distributiei locurilor pe regiuni astfel incat in regiunile mai slab populate proportia dintre numarul de locuri in congres si populatie sa fie usor mai ridicata), intre candidatii care nu au intrat deja in congres.

*** Adevarata concluzie este ca acest sistem hibrid impaca foarte bine si capra si varza, este foarte usor de reglat (ceva mai greu de pus pe hartie, ce-i drept, dar nici cu noua formula a productivitatii nu mi-e rusine), este deosebit de interesant, destul de modern (trebuie sa fac ceva cercetare in ceea ce priveste denumirea si cele cateva state in care e folosit, daca e folosit), scalabil - se regleaza numarul de locuri in functie de populatie, numar de regiuni si populatia locala - si este o combinatie intre ce vor adminii si ce vor jucatorii. As fi mai mult decat bucuros sa prezint orice aspecte ale acestui ultim sistem mai clar, mai in detaliu sau cu mai multe exemple daca e nevoie. Simt nevoia, de asemenea, sa mentionez ca varianta mea preferata este c.2.i.

Multumesc pentru lectura!