Yesterday's ABC at AtCoder was the second successive contest which suffered due to slow servers. Which is fine, I can understand, shit happens. But what actually pissed me off is that the admins didn't even care to announce if the contest is rated or not and ignored messages asking for it. Great going AtCoder!
→ Обратите внимание
До соревнования
CodeTON Round 9 (Div. 1 + Div. 2, Rated, Prizes!)
18:38:54
Зарегистрироваться »
CodeTON Round 9 (Div. 1 + Div. 2, Rated, Prizes!)
18:38:54
Зарегистрироваться »
*есть доп. регистрация
→ Лидеры (рейтинг)
| № | Пользователь | Рейтинг |
|---|---|---|
| 1 | tourist | 4009 |
| 2 | jiangly | 3823 |
| 3 | Benq | 3738 |
| 4 | Radewoosh | 3633 |
| 5 | jqdai0815 | 3620 |
| 6 | orzdevinwang | 3529 |
| 7 | ecnerwala | 3446 |
| 8 | Um_nik | 3396 |
| 9 | ksun48 | 3390 |
| 10 | gamegame | 3386 |
| Страны | Города | Организации | Всё → |
→ Лидеры (вклад)
| № | Пользователь | Вклад |
|---|---|---|
| 1 | cry | 167 |
| 2 | Um_nik | 163 |
| 3 | maomao90 | 162 |
| 3 | atcoder_official | 162 |
| 5 | adamant | 159 |
| 6 | -is-this-fft- | 158 |
| 7 | awoo | 157 |
| 8 | TheScrasse | 154 |
| 9 | Dominater069 | 153 |
| 9 | nor | 153 |
→ Найти пользователя
→ Прямой эфир
Codeforces (c) Copyright 2010-2024 Михаил Мирзаянов
Соревнования по программированию 2.0
Время на сервере: 22.11.2024 22:56:06 (h1).
Десктопная версия, переключиться на мобильную.
При поддержке
The reason is obvious. That's because of really really really really really really really really vigorous inflation in number of participants.
12 months ago, it was normal that there were less than 1500 participants in AtCoder contests. But now it is over 5500. Such inflation is not comparable to recent slight rise in participants of Codeforces or TopCoder SRM. More interestingly, ~4700 of them is from Japan.
That's because, very recently AtCoder became very famous in Japanese IT community. There are two reasons.
- On February 17th in 2019, there was a contest which was collaborated with NIKKEI, which is one of the most readed newspaper in Japan. In some days of newspaper, there was thing about AtCoder and NIKKEI contest, in all one page. This publicity of AtCoder was succeeded — lots of Japanese knew the existence of AtCoder.
- On April 26th in 2019, a very big IT company Dentsu Inc., funded 300 million yen (equivalent to about 2.7 million US dollars) to AtCoder. This also increased the trust and popularity of AtCoder.
Remark: I guess that AtCoder will use some part of 300 million yen to the enhancement of judge system and server.
Thanks for the information. Happy to hear that AtCoder has some more funding now. I like AtCoder and never miss its contests.
But the issue here was not slow servers. Like I already said, I understand that shit happens. But the least they should have done after this is to inform us that the contest is unrated, which they didn't. I asked them twice as clarification but even that was ignored.
I agree.
I think the only bad part in AtCoder is the clarifications(especially in the beginner contest) , when I ask for a thing , I get the reply after 20 minutes which is too late.
I have to say something as a writer of some contests.
I was one of the writer in AtCoder Beginner Contest 100. In this contest, the number of participants were 2500. And 2 or 3 people were working on answering the clarifications. However, since many people not noticing on corner-case thought that the constraints are invalid, and was asked lots of clarifications. Since we kindly answered each of clarifications, we typed very fast just to answer, and, finally exhausted.
Now the number of participants are 5500, which is x2.2 leap from ABC 100. It is sometimes, somewhat near-to-impossible number for contest organizers' side.
I hope that the inflation in number of participants and its popularity will not become like burst of soap bubble...
Hey you, Bakry, I know your handle name well because I often write or test AtCoder rounds. I am very impressed to see you here!
OK, here is a quiz for you. What name is the competitive programmer, who sends a lot of clarifications “explain plz” after we answered “Read the statement carefully”, and finally sends many nasty clarifications in Arabic?
HEY! You DO NOT speak to kid that way
Apologise immediately!!!
Please clarify the request.
what request
please apologise to kid immediately. What you said was rude and disrespectful. Such behaviour should NOT be tolerated
Are you his daddy or what?
hey , I'm sorry for what happened at this round , after the contest I realized my fault , it was because I was nervous as I didn't get answer for my clarification until ending of contest (I asked before ending with one hour) , I know that's not let me to write that so I'm very sorry for that.
for my statement above : in many rounds when I ask for clarifications I had some late replies (I asked some people and found they have the same problem) and I think it's the only bad thing in AtCoder and hope it gets better :)
Most of the clarifications are answered within 5 min. We do not answer to the questions, which exceed the purpose of the function such as the question which requires hints. We do not explain about the test cases you made. We do not answer whether your solution is correct. We do not debug your solution. Before posting clarification again, think whether it is suitable.
my question was about testcase in the statement (it wasn't testcase I made it) , I didn't say solution and asked if it's correct , and I didn't ask for debugging or anything and I didn't ask for hint.
so my question didn't violate the rules.
These are just examples of unsuitable clarifications. I can confirm that, if the clarifications are written clearly (Bad example: just to say "plz explain!", "The sample output is wrong, it should be $$$X$$$") ,do not include the request of hints (Bad example: "Why it holds?"), and it is about that we can answer now (Bad example: "Is it rated?" before we decide whether), answer is given within few minutes, during the contest.
my clarification was asking for something in testcase in the problem statement. it wasn't a hint or any other thing.
Even if your question is about the explanation of given sample test case, something can be a hint. The answer for "Why it is maximum?", "Why the answer is $$$X$$$?" can be a hint for general cases. Sometimes we do not write why we should output it: we sometimes cannot write other than intended output, without giving further information. In such case, we intentionally weaken the explanation and cannot answer such questions. If you write problem statement, you can understand me.
DEGwer got it. I know sometimes the output maybe hint for the problem.
but atleast say it's invalid question (if you see it invalid) or anything instead of making me waiting for the reply for like hour.
OK. I try to do it in further contest. Thank you for advise.
I mean, your clarification is not answered because it is hard to answer. Appropriate clarifications are answered in short time.
as I stated above , my question was about something in statement and it didn't violate rules (so I think it should be answered) :) .
Any why ,he still just a kid but he make a great performance and we should encourage him note : متشكرين يرجولة means thanks in Arabic :D
What the word means does not matter. Do you think we, ordinary Japanese people, can read your language?
Ordinary Japanese people can't read English either.
Why are you singling out one person? Seems kind of unprofessional to me...
I agree!
I wrote and tested a lot of AtCoder rounds, and he was the worst questioner I have ever seen. Usually I do not remember the handle names of questioners, but I remember him. And he complained about the response for his clarifications. That's all.
and what gives you the right to call him(note, a 13 year old kid) out publicly? Honestly, it just makes you seem like an asshole
How does age matter?
Younger kids tend to get more emotional about these sorts of things. It might be wise to think twice before posting comments that may negatively affect a user
I think it's just unfair to change attitude to someone base on their age. We are competing on the same ground. It's not a place to raise a kids, it's a place to compete each other, and his action was unnecessarily hampering contest admins.
EnumerativeCombinatorics I agree with your opinion , there isn't any difference between contestant 8 years old and contestant 40 years old , and my age doesn't intercedes what I did wrong.
you're right, regardless of his age, DEGwer had no right to call kid out publicly
I was just angry because he said like it is AtCoder's failure. But now, I finished to write what I wanted to tell the community. I forgive him and apologize. I wish he enjoy future contests. If he send appropriate questions, we will answer soon.
There is no indication of age in your profile, so can I ask you to fuck off?
Read this
Um_nik, your trolling's sort of ineffective if you reply to the wrong comment.
In defense of kid, it's important to understand that shitty problem statements exist
I agree. I don't like Arabic music. You should try listening to twice instead. Korean's a lot more soothing
I didn't intercede with age or anything because it's contset and there isn't any difference between any age.
"How about trying to solve problems"
It was the remaining problem and I didn't understand something in it so I asked in clarification (If they saw my question is wrong) then write it's invalid question instead of making me waiting for an hour for reply (I tried to read statement many times and I didn't understand it)
3 — I agree what I did was wrong in writing with arabic and I apologized for that.
Don't try to blame me for racism or that I hate Arabic language. The site has two official languages: Japanese and English. It is ok to not know Japanese. It is also ok to not know English. But to write clarification requests in language that is not supported by platform and most certainly is not known by contest moderators — it is not ok.
Stop saying that he is a kid. In this situation he is a participant just like everyone else. And he is deliberately trying to disturb contest admins work.
Um_nik by the way I sent my question in clarification in English.
and I agree that they should stop saying kid (because I'm participant as any other participant).
Lol can you read? Where did I accuse you of racism. All I said was that kpop is better than Arabic music.
You and I both know that the context of this argument wasn't that he sent Arabic messages. It's that he had perfectly valid clarifications which weren't answered. And you're right, as a participant, he should be treated with respect. That includes not calling him out on a public forum for something that he'd allegedly done a while back. Well, I mean I know that calling out participants is your kind of thing since you're a wannabe numb, but you should try and limit your edginess sometimes
Okay, let's look at your clarification if you insist on it being valid and not trying to get a hint
The context of this argument is sending a lot of unnecessary questions after getting an answer which wasn't good enough for him. I do believe that AtCoder contest moderators can decide if the question should be answered properly (not by "No comment" or "Read the problem statement"). There is no need in sending more questions if you didn't get proper answer the first time. Come on, if you get "Read the problem statement", be nice kid (if you insist on this term) and read the problem statement carefully.
The lack of an answer regarding a testcase was the problem
I agree, but do you think that it's right for atcoder admins to publically call out users who supposedly asked a lot of questions?
Just to clarify, I'm not kid.
Also you sort of ignored the most important part. Do you listen to twice?
"The context of this argument is sending a lot of unnecessary questions after getting an answer which wasn't good enough for him. "
and after that I apologized for my fault here (and I didn't get answer for my clarification which was in English and even they didn't say invalid).
"Come on, if you get "Read the problem statement", be nice kid (if you insist on this term) and read the problem statement carefully."
by the way I don't like this term and I don't insist on it.
There was an answer. "Read the problem statement". This is a perfectly valid answer.
Yes, I think AtCoder admins should mention this in response to accusation of not answering the questions. In contrast with my case, when I was just enraged (though also answering some accusations of taking a lot time to answer clarifications), DEGwer wrote this only in response to this given user. This is also an answer why did he singled out him. Because he singled him out himself by writing initial comment.
No, I don't listen to Twice. Can we now close this totally unrelated part of this conversation?
"There was an answer. "Read the problem statement". This is a perfectly valid answer."
I'm talking on my second question which I didn't have reply until ending of contest.
"No, I don't listen to Twice. Can we now close this totally unrelated part of this conversation?"
yes , I agree with you.
I'm sorry, the only your comment I answered was this. You actually behave quite nice now, and you already apologized, so the incident is (probably) over, at least DEGwer accepted your apology.
But I'm outraged by wtmoo comments in your defense.
You can check what comment I answering by clicking on ^ button in the comment header.
Also CF sends you a notification bell if someone answered your comment. So if you don't see a bell, probably it wasn't the answer to your comment.
wtmooooooooooooooooooooooo
my question was : "I read it for a million time until now I can't understand why BPB isn't a level ?"
BPB is something in statement and it wasn't a level in test case that was mentioned in statement , so I asked about that. (which I didn't get any reply)
Can you please also provide link to the problem and your first question which was answered?
yes , sure
link
my first question : "As I understood level 2 burger is Bread Bread Meat Meat Meat Bread Meat Bread Meat Meat Meat Bread Bread I need clarification...aren't it level 3 burger "
by the way this case was from problem statement and I didn't made it and they told me to read it again , I read it and asked them in second question why BPB isn't considered a level (which you can find it isn't level in problem statement) and they didn't reply (at least of question is invalid then they should say that).
I will rephrase your dialogue:
Basically, first question asks "Are you sure that problem statement is correct?". And the answer is basically "Yes". Then you send a question which doesn't really make sense because just "level" is not a term in this problem. Probably they could answer something like "Question is unclear", but
1. They already answered "Read the problem statement" to the first question, second question just confirms that you haven't read it. If a participant asks more than one question for one problem it is already some kind of dialogue which shouldn't happen. Answering this question raises chances of you asking more questions, more and more trying to make authors retell the statement just for you. This shouldn't happen, you already have good statement, read it.
2. You started question with "Why" which is not ok. All the questions should assume answer Yes/No. If authors would like to explain some reasoning, they would write it in the statement (and they did write how level-L burger is constructed, nothing to add here).
So, I think that authors decided that any answer to your question would lead to more questions and didn't want to escalate the situation with you asking more and more questions because you didn't bother to read and understand the statement. Probably it wouldn't happen on CF because of different policy, but it was a reasonable assumption that you would reread the statement after getting "Read the problem statement" for the first question.
Yeah, I know that you are saying that you read the statement million times. No you didn't. You didn't take your time to understand the statement and tried to shift this work to admins.
You are right
The number of upvotes is disturbing. How many angry problemsetters are in the crowd? :"D
At least Atcoder Blog should be updated about rated or unrated things.
One more thing why don't they provide english editorial for ABC . It always hurts me .
Please PM rng58 so he may look up ur concern. i did
What was the reply ?
Yes, I also think they should write about it. In Japanese version, the information is provided in the contest post. Official Twitter also says about it. Writing it in English is not so hard task for them, I think.
ABCs are managed in not the same way as ARCs or AGCs. Although rng_58 decide whether the problems people sent are suitable for ABCs, he does not associate with preparation (because he is busy to choose the future AGC problems). Writing English editorial is tough work. If they impose to write English editorials to writers, they cannot hold contests every week. That is why English editorial is not provided.
Honestly, I don't think it's difficult to say just "We've decided to make this contest unrated for the technical issue." I suppose they just forgot to do that.
Writing English editorials is different; realizing it could be an expensive business operation for them.
Writer here.
I am going to write about how I feel about the general stances of AtCoder towards international participants (it may be different from what the admins actually think).
\
We are always short of those who can communicate in English. Hire a translator? Well, previous rounds (years ago) showed that those who just know English but nothing about competitive programming made it much worse. Of course, the translator has to know Japanese as well, so even misof couldn't fill the position.
\
Yeah, AtCoder's official announcement is terrible. They usually announce only on Twitter in Japanese. (If you are interested, follow @atcoder (and @chokudai) then you can see the announcement in Japanese.) Speaking of the contest's case, the final decision has been made hours after the contest ended. At the time, the part-time English writing staff was already offline. (Another bad point of AtCoder is that there are no full-time employees who dare to write in English. Why don't you do this, rng_58?)
\
Generally speaking, editorials are really cost-inefficient. Even if the officials don't prepare good editorials, some people write better ones. Also, those who are preparing the problem sets are good coders (orange, red, ...), and they don't know what is the difficult point of easy problems (or do you still want to read "This is just a straightforward problem." for all the questions?). So, AtCoder tends to rely on the community for good editorials instead. (Again, it's impossible to keep translating editorials every week).
\
Honestly, it seems that AtCoder only cares about these two groups.
If you are not in either group, ... well, at least neither am I, pals.
\
Ask the community. Use Google translate. Learn Japanese.
\
If you can communicate in Japanese, can write English and can do programming contests (at least orange), I think you might have a chance to work as a part-time. Their payment is good especially if you work a lot.
tl;dr
I don't like this kind of cynical analysis especially about "ratism", a little bit harmful and subjective rumor. However, some are somewhat constructive...
Well, a certain source you can also see confirmed this is not wrong, saying that they don’t write English editorials for ABC because they are not the main target currently.
hello , EnumerativeCombinatorics . what about ksun48 and petr mitrichev.
Hello! Nice to see you again ;)
Here is the link you'll probably think helpful: Beware an annoying person sending messages using multiple accounts
U r right, atcoder editorials even for the most toughest problems are like "u can do it in o(nlogn) using starighforward dfs and dp using min heap."
now what is this , please rng_58 how can anyone understand it who is not so pro like u. simply saying it can be easily solved using ... very bad .
I’ll write an editorial in English if you pay me $300 for each contest, or if many people pay for reading my editorials on my Patreon.
Wow, what the kind wolf!
IMHO, if you can't understand editorial, it simply means lack of the skill. That's all.
I'm actually sceptical to this... That's because I can see some reactions from editorial readers that, "there is obviously easy-to-read editorial and hard-to-read editorial".
When I am writing editorials of ABCs in Japanese, I am really trying to be understandable for many coders including beginners. That's because I personally think that, if we teach as many things that is concerned to the problem, the beginners can obtain many opportunities to learn new algorithms and techniques.
So, at least I don't want to throw away thinking understandedness of people by reasoning that "it's just lack of skill", and I sometimes writing editorial with strong thinking of it.
P.S. My English skills are not very good compared to others like university student coders, so it could not be understandable for others because of language barriers. Also, it is commonly said that even my Japanese is not accurate.
Seems it would be best if you write all the editorials.
That's not true. A translator that satisfies all other requirements — understands competitive programming and is good at English writing — is able to take a description of a problem in broken Engrish and rewrite it into a good statement. If Atcoder can't communicate anything in English, anything at all, though... then that's a much bigger problem.
+1 I feel that's what TopCoder has done so far.
I don't think it works without having much trouble. You know, sometimes you have to explain complex rules or constraints in your problem statement. (e.g. replace the integer sequence A with B.) Unlike most European languages, verbatim replacing from Japanese to English doesn't work even to make broken Engrish. Sometimes it (or Google Translate too) generates a sentence totally incomprehensible to anyone, sometimes it generates a sentence of the opposite meaning. Still, for many writers here, it takes many times longer to write a statement in English as well as Japanese.
As you can see in Codeforces, rng_58 technically could write an English statement. But don't you think it is better that he can concentrate on improving the quality of problems, data sets, time limits, etc, rather than spending hours for writing English?
Outside the problem statements, Google Translate works well, and they could hire students for onsite events too. So I guess it is not a big issue.
Maybe you don't understand the process I'm talking about.
There's no verbatim replacing J->E involved. There most definitely isn't anything like Google Translate involved, it's based on human-to-human bidirectional communication where necessary to make sure both people understand the same problem in the same way. It doesn't require anything except a writer that understands English on a level above "will suicide when lost in a foreign country".
Go ahead, tell me it's not doable.
Well, currently evima translates all problem statements with high quality and I think everything works fine.
From my point of view, translations also worked well enough — well, editorials are delayed, but I try to avoid reading them anyway unless I'm stuck for days (and who cares about ABC oneliners lol). I was just responding to that comment about not enough English speakers, that 2-level translation can work just as well as 1-level.
It's quite different from survival. You can get rescued by repeating "I'm lost I'm from Japan", but that doesn't make a programming contest. I don't believe more than half of the problem setters can and willing to do.
Well I believe even with just the Japanese statements and nothing else (or maybe 1 or 2 samples to help), it is possible to translate entire statement to English (maybe not so clear about the entire story of the problem, but at least an equivalent English formulation) using just Google Translate. Maybe this is just speaking from my experience since I have done quite a number of contests on Atcoder (and some past JOI) in Japanese without really knowing Japanese and relying on broken google translate, and was able to understand statements fast enough to compete (and this sometimes apply to editorials too, though if I can't copy paste the text then it might be harder and I have to do some guesswork). I am not sure if this is completely relevant to your point, and maybe this is just me, just I think that maybe translating statements to English might not be that hard, but then I guess this is irrelevant since for now the problems are already translated well.
https://atcoder.jp/contests/abc126/clarifications
As we know, human often make mistake. Contest admin is no exception. He forgot to announce to all.
Sorry for my mistake...
To be more precise, he forgot push "publish button" for the announcement.
but what about english editorials
we really need English Editorials. :(
Sorry, it was not announced because we accidentally made the annoucement unpublic. We'll try to avoid this in the future.
Engineers in AtCoder analyzed the trouble, and it seems they found the reason (and the reason was the same for the two recent troubles) and now try to fix it. So, probably the next contest will work fine. We'll hold two ABCs this weekend, and if they work fine as expected, we'll hold an AGC in June 2nd (also it's possible that we hold an ARC in June 1st).
My main interest in AtCoder is to create a prestigious, high-level international contest using high-qualitied, thinking-oriented problems. (The ideal form of such a contest site is TopCoder in 10 years ago.)
Thus, I devote most of my work for AGCs and ARCs — I check proposals, choose good tasks, and carefully test all of them myself (with other testers). Also, I watch most of AGC/ARCs during the contest and sometimes translate editorials myself. (All <2800-rated contests are equivalent to ARCs, regardless of their names. For example diverta2019.)
ABCs are prepared in a bit different way — there is a pool of prepared easy problems, and before each contest one problem from each difficulty slot is chosen at random, and we put them together as a contest. Currently we don't have sufficient resource for writing English editorials for all of them, sorry.
chokudai is also very interested in popularizing competitive programming in Japan — and I guess the union of our main interests resulted in EnumerativeCombinatorics's guess.
By the way, there is an idea to add a "discussion" tab to AtCoder's contest pages with the link to corresponding blog on CF. What do you think about it? Do you want discussion blogs for ABCs too?
I think discussion blogs are necessary
I agree with GaryZ005. Please, atleast put a cf like comment section on atcoder too so that we can discuss problems.
rng_58 why ARC is so irregular nowadays