New Brilliant Solutions Layout
Discussion thread for the awesome feature! Any questions, suggestions, etc. It'd be interesting to see what everyone thinks. Personally, I love it. One suggestion, though -- it'd be. Just like the new feature, it'd be kind of nice to see, on people's profiles, how many problems they have contributed to Brilliant (as in own-problem submissions).
Just a thought; thank you so much, Brilliant!
Note by
Ahaan Rungta
7 years, 11 months ago
No vote yet
16 votes
Easy Math Editor
This discussion board is a place to discuss our Daily Challenges and the math and science related to those challenges. Explanations are more than just a solution — they should explain the steps and thinking strategies that you used to obtain the solution. Comments should further the discussion of math and science.
When posting on Brilliant:
*italics*or_italics_**bold**or__bold__paragraph 1
paragraph 2
[example link](https://brilliant.org)> This is a quote# I indented these lines # 4 spaces, and now they show # up as a code block. print "hello world"\(...\)or\[...\]to ensure proper formatting.2 \times 32^{34}a_{i-1}\frac{2}{3}\sqrt{2}\sum_{i=1}^3\sin \theta\boxed{123}Comments
I also really like the idea behind this change. However, I'm not sure I like how it was implemented. As Muhammad said (even though I think he went a bit far when he called it racist), it really is unfair to and discouraging for people who like submitting solutions when they know their solution will probably only be glanced over by most people since the one that was submitted by someone in a more favorable time-zone already has 20 up-votes. I don't think that the featured solution should be completely chosen by this upvote system, as it will usually find the earliest solution which solves the problem rather than the "best" solution to the problem.
Also, the current system is not one that will often lead to solution "discussion," but rather to people simply posting solutions and others just reading them. It is, of course, way too early to judge how the system will turn out towards the end of this week and in the future, but I have seen many more solutions posted than comments posted on those solutions. I don't think that was what was intended with this system.
In addition, I personally got very excited when the "You've been chosen to submit a solution" message popped up on my screen. That encouraged me to try and cover every possible detail in my solution, since I knew that some people who hadn't solved the problem really could learn something from reading my solution.
Finally, I'm sure that there are many people here on Brilliant who are not very well-acquainted with writing full solutions to problems, and that's ok! Brilliant is a place to learn just as much as it is a place to test yourself. Unfortunately, some people may feel that they shouldn't write a solution using the new system for this exact reason, as they don't want one of their first solutions being judged by their peers or simply ignored. In my view, this theoretical "side-effect" of the new system goes against the spirit of the entire Brilliant website.
I urge you guys to keep the old system in place while you make a few tweaks to this new system, and when some of the main issues have been fixed, to re-implement this new system. This system has potential to become a great change for Brilliant and, in fact, encourage more people to write solutions, but I don't think that its current implementation does that effectively.
Log in to reply
A hybrid approach could address the issues raised so far. All those who have answered a given question may be invited to write a solution, but none will be made available for others to see until a few days into the week, and no voting will be allowed until the problem is closed.
Although the concern about fairness with respect to early responses is valid, the reason why it may not be a good idea to wait to post everyone's solutions until the entire week is over is because doing so would likely lead to many identical (or near-identical) solutions. Having to read through hundreds of submissions for each question would be taxing. So a balance could be struck by delaying posting for a few days, after which anyone else who has a novel solution can feel free to add to the discussion. It also gives students a chance to try their own hand at writing up a clear, step-by-step solution, rather than seeing someone else's response immediately.
Because posts are sorted in order of decreasing number of votes, there is also the concern of popularity bias--that is, voters will tend to more carefully review, and vote for, those responses that have already received numerous votes. This gives early responders a visibility advantage that could be compensated for by sorting posts in random or newest-first order and not permitting voting until the end.
Log in to reply
Something like this is very similar to what I had in mind. Make it happen, Calvin.
Delaying posting for a few days does not really solve the problem, does it? If in someone's timezone, posting is opened at 3AM, then their disadvantage is huge, as many others had their solutions ready and posted those immediately.
Delaying the voting is a good idea, but an entire week might be a bit long. If it would be allowed after one or two days, then you have had quite some time to work on the problem and write your solution, before the voting starts.
I think that the solutions should be visible from the very start, because if I solve a problem and I'm wondering how others have solved it, then it would be a bit strange that some others have posted their solutions but those are just not visible to me. It is true that you could just copy someone else's solution (before the voting starts) in order to get the same chance of your solution being featured, but this issue could be solved by asking the user right after solving the problem whether he wants to submit a solution, or view the solutions by others. In this way, original ways of solving the problem cannot be stolen by others.
Log in to reply
I really like your idea involving only submitting a solution before viewing everyone else's solutions. That could help stop a lot of cheating.
Also, I think what Hero was getting at was having a 5-ish day period where everyone can submit their solution and having them hidden until then. In the meantime, there could be a true discussion-based period where people who have solved the problem (and perhaps even some people who haven't, if they're willing to give up a fair amount of points to join in on the discussion) can talk about the problem and its subtleties. Some great insights about the problem and its generalizations could sprout from these discussions. Then, after the 5 days pass, the solutions could be revealed (anonymously perhaps) in random order and users could browse them and vote for a certain number of their favorite ones. Hiding the upvote count might help, too. After the week ends, the top solution(s) would then be featured and have the writer's name attached to it(them). I think that might end up being a good system that encourages both discussion and submission of solutions regardless of how long it takes a person to write up their solution and without fear of being "judged."
I sorta rambled on and on xD I hope everyone gets the idea.
Log in to reply
I totally get what you are saying. And I agree with it, for the most part.
You are saying that submitted solutions can be browsed after a period of 5 days, but I think that is way too long. My interest in most problems fades away over time, and I'd love to see other solutions right after I solved the problem, but I would not read solutions to all problems extensively half a week after I solved a problem. So, I'd prefer solutions to be visible instantly, but in random order, and voting only being possible after a certain time. I think 5 days is a bit too long though, 1 or 2 would already get rid of the disadvantage some people have due to different time zones.
I have thought about the anonymous part as well, but I don't think it would work out that well, as people might comment on it and the author of the solution would give himself away. Also, you would get a lot of "This is my solution" comments.
Log in to reply
I agree that 5 days is quite long and 1-2 does make much more sense. I think the underlying ideas are more important to discuss than the actual numbers, and I think we're seeing eye to eye on most things :)
Yeah, you're probably right about the abuse of an anonymity-based system, but I think that without it, a lot of people might just upvote their friends. I guess we'll have to wait and see how everything plays out.
Log in to reply
They might still upvote their friends by just asking what their solution was. Those 'unfair' votes will probably not prevent the best solutions from coming out on top.
I think that most things are already going quite well, as the best solutions seem to be somewhere near the top, usually.
Thanks for your encouragement!
We are working through the implementation of this feature (which is why it's not rolled out to all the other topics as yet). We decided that it was best to see how the students would use this feature, instead of sitting in our comfy chairs theorizing about possible uses (and abuses). Over the next few weeks, we will work out the best way for us to host these discussions in a fair manner.
As always, your feedback is greatly appreciated!
Log in to reply
For the record (and in case it wasn't absolutely clear from my previous post), I think it's great that you guys implemented something new and are actively looking to improve the website :)
Also, sometimes people's judgement on what makes a "good" solution and brilliant's judgement of what makes a good solution are not the same. Most people will look at a solution, say "hey, that makes sense," and upvote it. However, brilliant (using the traditional method) requires much more in a solution. It needs to be intuitive and not assume things without either proving them or mentioning other proofs of why they can assume things, and this is checked by the staff. However, most members will not catch that, so it can lead to problems.
I think that the system should be such that all the solutions posted are side-by-side, rather than listed by chronological order. This way, people can see easily whether their type of solution has already been posted.
agree... like the old system...
I really like this new layout. It is definitely not perfect yet, as Muhammad and others have pointed out. Here's a summary of what I've posted before, and some other thoughts:
This feature (when enrolled throughout the whole website, especially) together with the Practice sections that has been introduced a while back, is taking away a lot of weekly work from the Staff, which is great. I hope to see more of these awesome advancements on this website in the future. :)
Log in to reply
These are some great ideas! I think these could work. They could also get rid of of some of the problems pointed out by others..
I think your fifth point should be the main purpose of this new layout. We should be able to discuss about solutions, comment on others, talk about different techniques and approaches rather than read ten solutions that look almost the same.
However I still think that this change is great. It is not perfect but I've seen some great ideas on this thread. These ideas will surely get rid of some of the problems that others have pointed out.
Log in to reply
You're summarizing pretty well what I meant to say with that fifth point. :) That's really something that could be improved in this layout. Something else that would be nice is discussing the problem with others and then write the perfect solution together, but that would encourage the stealing of ideas - I'm not sure if that's even a bad thing or not. Maybe we then could remark which users came up with which parts of the solution. Anyways, that last thing would be quite a bit harder to implement.
With reference to the 4th point: Atleast I am not able to vote for myself. Maybe they excluded self vote recently.
For the fifth point I would like to suggest that the solutions discussion and problem discussion be separate. Otherwise it may look like a big mess.
I strongly agree with all the other points. :D
Log in to reply
This is a great list Tim. As always, we are very appreciative of your feedback and constructive criticism. We will take this into account as we modify the system.
We are really excited that many students are willing to put themselves out there and display their solutions to others. We also understand that some students might not be so bold and might hold back. As such, we will be modifying the submission slightly, to allow students to write up their solutions but not require them to post it. In this way, they can refer to their own writeup and compare it to similar solutions, to learn from the style of presentation.
Under the old system, we had built up to having 75% of those who were asked to submit solutions actually submit their solution. This is a pretty amazing stat, especially when compared to most math circles who have trouble getting their students write up a proper solution after attempting the problem. We want to build off the excitement and energy of the students, to encourage them to learn from one another.
Give us time to modify the flow of solution submissions. For things that can be done quickly, you should see these changes coming soon. As Smabit mentioned, we did exclude self vote yesterday, as that was a quick thing to do which we agree with. I believe that within a month, solution discussion will be a favorite of many Brilliant students, as it offers you great opportunities for distilling your ideas around the problem that all of you worked on.
You're right, the ability to vote for yourself has been removed. And yeah, of course those discussions should be separated, a real discussion part, and a formal solution part.
Perhaps so much focus on enforcing fairness of voting is not the best way to approach the issue. After all, if anything can be inferred from the way voting has progressed so far in this week's test run, it is that maybe the most popular solution should not automatically become the featured solution.
A voting process could be used as a guide for selecting featured solutions, but I don't think it should be the sole deciding factor.
Log in to reply
I don't completely agree with you. I do also feel like the most up-voted solution should not automatically be the featured one, but this voting system does save a lot of time for the Brilliant staff.
If they are going to keep the voting system, it should be done right. Having a voting system that is just a guideline is okay, but that should not be a reason to not make it perfect. I'm writing my solutions not necessarily to have it featured next week, but to share my solution with the rest of the solvers of that problem. For that reason, no matter if the highest voted solution will be the featured one or not, I would still want my solution to be on top.
If voting would be very unimportant, they could just as well abandon it, which is -I think you can agree- not the way to go.
Log in to reply
Where did I say that "voting would be very unimportant?" Nowhere did I suggest such a thing. I also did not imply that not implementing a properly fair voting system was acceptable. What I implied is that it may be unrealistic to try to devise rules for a voting system that always selects the best solution, given that what is considered "best" is subjective, and voters do not necessarily actually know what is best. See the post by Michael T. here, who mentions this exact concern.
Therefore, I feel that some level of staff involvement remains necessary. Expediency is not necessarily the only concern; selection of good solutions, especially those that are elegant in their approach, and may reveal further insights, is something that should also be valued, and that is something Calvin has demonstrably provided. Voting, executed properly but without obsessive concern for things like anonymity, cheating, and so forth, can be a good way to winnow the number of submissions that may need to be reviewed.
The fundamental goal of Brilliant and the reason why any of us are here in the first place, is for the love of learning through problem solving, and fostering and discovering problem solving talents. Those who would seek to cheat have nothing substantive to gain. Cheat for a t-shirt and some other minor prizes? Cheat for site recognition? That's pitifully small-minded in the face of the real reward. My point is that while fairness is laudable--one shouldn't make it convenient to cheat--fairness shouldn't be imposed to its own end, and especially not to the detriment of learning, for those who are earnest.
Log in to reply
I wasn't trying to accuse you of saying that voting isn't important, although that was how it sounded like, I'm sorry for that. What I meant to say, was that the voting system (if done right) will save the Brilliant staff a lot of time, but not all the time they put in checking and featuring solutions: solutions with many upvotes might have its flaws, and some very elegant solutions might not have very many upvotes because it was submitted very late, or something similar. The voting system just helps the good solutions towards the top and the bad solutions towards the bottom of the list. I think our views on this part are very similar.
It's true that when you cheat on Brilliant, it won't gain you very much, except for a t-shirt and and ebook. However, cheaters might make the Brilliant.org experience a little less enjoyable for the non-cheaters. If I came up with a very cool way to solve a problem, and someone just copies that idea (or the whole solution), then I wouldn't like that, no matter whether he will be 'caught' or not. As Muhammad A. pointed out, some users have a geographical disadvantage, and although that might not stop the best solution from getting featured, it would still be better if everyone would have an equal chance.
Although it is not the primary goal of Brilliant, competition between the users might stimulate great things. Trying to get their solutions featured will encourage users to do their very best solving a problem and writing a rigorous proof. It would be sad if they would be discouraged by cheaters, or unfairness in the system.
Thank you all for the excellent feedback so far. Many of the suggestions made in this discussion will be implemented in future weeks of solutions discussions. Week by week it will get incrementally closer to the ideal environment to have a good conversation among a large group of people who have a shared experience of a math problem. It is a lot easier to design websites when they are used by people who care enough about it to put as much thought into their experience of it as all of you. Thanks!
Quick question for anyone who is looking. We have received little to no feedback yet on what the experience is like for people who struggled with a problem, couldn't solve it, and were able to enter the solution discussion instead of waiting a week to figure out what they did not understand.
Has anybody in this thread entered one of the solution discussions before getting the problem right yet? If so, it would be helpful for us to know how you felt that went.
Log in to reply
Do you mean buying access to the solutions, with points? I did that for one problem. At first it said I got all three tries wrong, while I didn't even try once, but it seems to be fixed now. Also, I couldn't submit a solution, which makes sense. Overall, it felt pretty stable.
I was wondering one thing. If I would have submitted 3 wrong answers, would that give me free access to the solutions, or would I still have to spend points?
I love it too! I hope they extend it to other topics as well instead of only Number Theory & Algebra. I'll be looking forward to that.
And that is a nice suggestion, Ahaan! I think everyone would love that.
My suggestion what if the other submitted solutions do not open until our solution.is submitted, for others that do not provide a solution then all solutions will be open after the end of the week.
Perhaps it would be good to randomize the order the solutions are shown, and hide the vote counts during the week.
Why this system is only applied for Number theory and Algebra? Why not for Geometry and Combinatorics ?
Log in to reply
This system is currently experimental. If this gets popular, they will extend it to other categories too.
I think that Brilliant should provide answers to each user after the 3 attempts are over.This way,no cheating,no injustice.
THis change is Awesome because I had solved one problems in very hectic way but when I saw the solution, it was simply awesome!!! Thumbs UP! Brilliant! :D
Dear calvin, i disagree to have brilliant solution layout system. Why ?
We had different timezone around the world, like me... The problem release on 7 am in my country, which is time to go to school (Which mean, i only can access brilliant after school). So, what about if our solution same with the previous user ? It's not about cheat, just the time. And, our solution will be bury into deep post, and they will not vote up. So, i think it's look like racist.
And, u also provide user to submit their solution. And, may be they can cheat from the live solution. And, it's unfair. Maybe, we will change the sentence, but the idea came from cheat. So, it's unfair for me. Thx...
Log in to reply
Thanks for your feedback. We are looking at improving this system based on how students use it. We will definitely modify the way that solutions get submitted, as we learn how students use the feature.
I would strongly disagree with your claim that it is "racist". By using such a politically charged term, it detracts from the validity of your other comments.
Just because you feel that you have been disadvantaged by your timezone (which is a geographical condition), doesn't necessarily mean that the "grass is greener on the other side". You do not know how others are also reacting to the release of the problem sets. Those who get the problem sets released at 12am their time are not necessarily in a better position than you are. I would argue that since you are already up 7 hours before them, you have more time to work on the problems and present your arguments. Of course, they could wake up at 12am, and deal with their parents screaming at them to go back to sleep as it is a school night. WIth this interpretation, are we 'racist' against everyone?
We have taken time zone into consideration, and are looking at grouping students according to when they view the problems. We may choose to split out the solution posts, so that those who only access the problems on Friday night, do not get disadvantaged in terms of presenting their arguments. This decision is dependent on how students behave.
Log in to reply
I would suggest that voting on worked solutions is not possible within the first 24 hours after the problems have been published, so everyone gets a day to work on the problems and write solutions for problems they think they solved in a nice way. If the solutions are then presented in random order (which is different for everyone), the best (rather than the first) solutions will get the most votes.
Why do you reveal the solution immediately after we answer?
A better alternative would be to let us submit our own solutions and then see the others. If we choose not to submit a solution we can directly view others solutions and vote accordingly, maybe without knowing the vote ups that have already been casted.
I am 50-50 agreed with you. This will lead to cheating.
But if you want to enjoy the fun hidden in these questions , you should attempt questions first and if you feel yourself frustrated, then you should go with the 'show solutions'.
This layout will help to know about others way in solving of particular questions. and also let us away from frustration.
i too support Muhammad about this. we get a fresh set every week at about 6:00 in the morning by 7 we have to leave for our school so by the time we answer there will be lot of answers and our methods my be similar to that of others we are not able to participate.
I liked the old Brilliant way. Selecting at random... And also solutions shouldn't be available instantly, it was better(though frustrating but kinda good :D) when we had to wait for a time to get solutions...
totally agreed!
I have an idea... Brilliant cando as following - --> If the user attempts the question and get it correct then he may be asked to the submit his own solution, without showing the other solutions. --> But if he fails in all three attempts, then he will be allowed to know the correct solution (that is submitted by others) for the problem, and so his doubts are clear.
Please tell me about my idea.
Totally agree with you, Kahfi. The old one is better.