New Season 2010/2011 - Recent changes in functionality

[euler]

Saturday September 4th, 13.00 the new season is starting with the publication of problem 300.

In order to make sure that problems are only published during the weekend (particularly on the Western Hemisphere) the publication scheme has been changed into:
Sat 13:00, Sat 16:00, Sat 19:00, Sat 22:00, Sun 01:00, Sun 04:00, Sun 07:00 and Sun 10:00;
then back to Sat 13:00 again (UK times).

Several changes to the website will take effect at the start of the season:

First of all, both the problems fora and the phpBB forum will support LaTeX.
To get inline formulas simply place your TeX between two single dollar signs.
To get block level formatting place your TeX between two double dollar signs.

Secondly a new Veterans class will be introduced for those that have solved 300 problems or more.
Members achieving the status of "Veteran" will be immortalised in Project Euler's "Hall of Fame". But please note that the number of problems solved will no longer be shown for Veterans; also members in the Veterans class are displayed sorted by member ID, that is "oldest" members first.

The rank lists will cease to contain links to the public profile pages as these pages will not be accessible to other members any more.
For those that want to display a picture with the number of problems solved on another website it is possible to get such a picture directly from Project Euler.
The URI of this picture is http://projecteuler.net/profile/my_username.png (change the obvious).
This URI can also be found on your profile page.
For those in the Veterans class this picture will show 300+

[elendiastarman]

WOOOHOOO!!!

I particularly like being able to use LaTeX...

[hk]

perhaps you should change your profile picture into

[elendiastarman]

perhaps you should change your profile picture into

Ah, yes of course. Thank you for reminding me...

[Lord_Farin]

The rank lists will cease to contain links to the public profile pages as these pages will not be accessible to other members any more.

Why is it called 'public profile' if it isn't public anymore? What is the reason for confining this information? It was really useful and fun to look at those pages...

[ThomasH]

First of all, both the problems fora and the phpBB forum will support LaTeX.

Bravo.
I just must try it:
The definition of the Binomial Coefficient is $\begin{equation} {n \choose m} = \frac{n!}{m! (m-n)!} = {n \choose n-m} \end{equation}$
Very fine!

[Slaunger]

Why is it called 'public profile' if it isn't public anymore? What is the reason for confining this information? It was really useful and fun to look at those pages...

I share the same sentiment as Lord Farin regarding this. The old signature with the "bar of vertical lines showing solved problems" was really rich in information and interesting for me and so were the public profiles. It was fun to make patterns in it. My pattern was boring, just filling in from the left without any holes (but also hard as I never skip a problem until I solve it), others attacking problems from the high id end, were really cool. I have not quite understood the objective for hiding this information in the internally generated signature image.

Regarding the argument that the public profiles impose a problem as it adds constraints to the inner workings of the site, I'd say it should not be a problem that the site team decides to change something in how things are presented as long as the information is still available. then people can surely figure out how to adjust their bots and what not crawling the statistics on the site.

I like the veterans idea of ranking +300s according to the user id (how loyal the user have been to the site). It is a demotivating for fast cheaters.

Last, but not least I really appreciate this: $e^{\mathrm{i}\pi}+1=0$.

Slaunger

[hk]

For those that want the information contained in the profile images for further processing (e.g. in local leaderboards) Colin made this information available in .xml format and .txt format. See viewtopic.php?f=5&t=2003#p21204.
Nice work Colin.
The introduction of the Veterans class (with the obvious reason to demotivate fake 100%-ers, and those trying to hunt down people they suspect of cheating in the problems fora (what happened quite a few times, sometimes quite wrong)) made it necessary to minimise the information in the "public" profiles. The introduction of the Veterans class would have been useless if we had allowed people to crawl the public profile to find out how much problems those in the Veterans class would have solved actually.
Or put in other words we had to remove the list of problems solved.
So the public profile then wouldn't have shown more than now is available in the picture and the text based versions.
An additional advantage of the system with text and image based versions is that they put much less strain on the database than countless bots crawling the profiles.

All that I can add is that the steps taken are a follow up of the steps we took when introducing the fastest 20 lists and the new Eulerians list. If one wishes to compete: there is your place. If not, be happy to solve the problems provided but don't hang to the old system of 100%-ers. That system has been compromised from the outside too much to keep in place.

[Slaunger]

Thank you for the explanation, hk.

It is probably because I am not in that high end of solvers, and have not experienced these kinds of problems, that I was unaware of these sometimes fake cheat hunting in problems threads based on public profiles. Although I do not want to beat a dead horse with a stick could one not just hide the public information for the very highest levels then? Or, even better, make it an option in the user profile if it should be public or private defaulting to private?

E.g., I have this fun PE race with a collegue at work. We are not mathematicians nor computer scientists, just rusty physicists having fun with numbers as digression from "normal work". I am very confident he does not cheat, and I think he is confident that I am not cheating either. So we regulerly "tease" each other with the problems one has solved but the other has not managed to solve (yet), by inspecting each others profile. That part of the fun is gone now to bash cheaters going for 100%. But I am confident that the team members have a much better feeling of the proportions of the current problems you are trying to fight with the new initiative and that it outweighs other minor disadvantages such as this. It is just seen from the perspective of an average mediocre problem solver racing a collegue.

Slaunger

[jui-feng]

Hey, I like most of the new changes!

However, I also liked the possibility to see which problems had been solved already by a member. Sometimes, when trying to find a problem where I at least had a very slight clue how to solve it, I just opened the score board and looked at some member profiles which had about the same number of problems solved as I had. Then I searched their solved problems list to find one that I hadn't solved yet.
Sometimes that was a nice way to discover a (for me) solveable problem.

It's the same with the new fastest solvers table and getting rid of the "eulerians table" with good "recent performance". Although I managed to solve some of the newer problems (and might some day be able to solve half of the most recent problems), I'm afraid I'm too casual to ever solve a problem within the first twenty solvers. But maybe one day I'll do it, a year ago I didn't think that I'd ever solve a problem where the problem thread wasn't closed already.

Oh well, I guess the decision has been made and there were good reasons to change it the way it is now. The XML files are surely very useful for anyone who builds things on top of project euler.

And the most important thing is: new problems, yay! Thanks so much project euler.

[IvayloS]

Perhaps the line that says "(34 more until last level)" should also be changed with the introduction of a new level.
I also liked the option to view the problems solved by a member as for me there are two types of project eulereans - the ones that solve the problems in a sequence(i.e. from 1 to last or the other way round) and the not so serious who only solve the easiest problems.

[Lord_Farin]

hk, thank you for your comprehensive reply. Nonetheless it's sad these actions were necessary because of the 'bad' few. However I now understand the reasoning of PE, and can live better with it.

Heads up for the problems and the facilities. Keep up the good work.
(Edit: Do the RSS feeds come back or are they forfeited?)

[mario62]

Or put in other words we had to remove the list of problems solved.
...
All that I can add is that the steps taken are a follow up of the steps we took when introducing the fastest 20 lists and the new Eulerians list. If one wishes to compete: there is your place. If not, be happy to solve the problems provided but don't hang to the old system of 100%-ers. That system has been compromised from the outside too much to keep in place.

This was a very sad decision. i'm not interested to be one of the fastest or of haveing 100% or whatever. And i am not interested to know who solved 100% of all problems and i am not interested to see if you solved the problem 10min after it was released. (i nearly never have time solve problems from friday till sunday).
i just want to solve, from time to time, one of the problem and it was very interesting to see what other people have solved - e.g. just for selection of a problem i want to solve next.



mario.

[jui-feng]

I wonder if anything can be done about incorrect (attempts of) latex rendering i.e. in old problem threads? http://projecteuler.net/index.php?section=forum&id=129 has an example post by ke9tv: "Extra close brace or missing open brace".

[hk]

This occurs if people have posted code outside code tags AND the code contains dollar signs indicating variables.
For the post mentioned I've put code tags around it.
Should be cured by now.
People can PM me such posts (problem number, page, and member name) and I will go through them in due time.
Or Colin might find a way to do batch processing on that.

[Stache]

In my case, at least, the statement

Once members have solved 300 problems it is clear that quantity ceases to be a challenge. Instead speed and efficiency becomes the new motivation and driving force as members compete for a place in the prestigious Fastest Solvers tables.

does not apply at all. In fact, I would argue that as a motivator, this is more likely to drive people to grungy brute-force solutions rather than thinking through the deeper mathematics for an elegant solution.

Trying to maintain my 100% rating does motivate me, however. Further, it's interesting to see who else has the persistence to do that. The breakdowns by country with the ordered lists of 100%ers are also interesting.

I'm not arguing against the Fastest Solvers paradigm...that's great for those that wish to compete in that way, and presumably that would be difficult to cheat, at least without hacking the website. But I would like to see the "Number of problems solved" maintained. If there are a few cheaters, so be it. You can't forestall every possibility. Provide the data and let those of us that like to look at such things judge for ourselves the credibility of the solvers.

[elr]

if you removed the member public profile because of bots & statistics collector
why not adding a captcha check instead of removing the public profile ?

[Quique12]

How does the knowledge of what problems others have solved help in cheating to get the solutions to all the problems?

[albert]

I regret the loss of the feature of viewing each others profiles.

I used it to get an idea of what kind of problems people I'm interested in,
could not solve, or didn't bother to solve.
Also for finding problems I might get interested in myself.
I can understand the bot problem.

I also regret leaving out the score on the last 25 problems. (red->green).
It was a big motivator for me to keep that at least above 20.

Groetjes Albert

[hk]

Thanks Stache for your comments.
Here follow some remarks.

In my case, at least, the statement

Once members have solved 300 problems it is clear that quantity ceases to be a challenge. Instead speed and efficiency becomes the new motivation and driving force as members compete for a place in the prestigious Fastest Solvers tables.

does not apply at all. In fact, I would argue that as a motivator, this is more likely to drive people to grungy brute-force solutions rather than thinking through the deeper mathematics for an elegant solution.

We've adapted the heading of the Veterans class into:

Once members have solved 300 problems it is clear that sheer quantity ceases to be the challenge. Instead they find themselves in one or more of the following categories:

•Engaging the true spirit of Project Euler by sharing their refined thinking and strategies via the problem forums.
•Getting involved in working with the development team on new problem material.
•Making speed and efficiency a driving force as they compete for a place in the prestigious Fastest Solvers tables.
•Continuing to enjoy the profound sense of self-satisfaction in cracking another "hard nut".
Whatever the motivation it is fitting that tribute is paid to those members that have moved beyond Legendary status by being immortalised in the Veteran's Hall of Fame.

Moreover if members cease to be active their names will be gradually fading. How this works you can observe in the other lists.

Trying to maintain my 100% rating does motivate me, however. Further, it's interesting to see who else has the persistence to do that. The breakdowns by country with the ordered lists of 100%ers are also interesting.

I'm not arguing against the Fastest Solvers paradigm...that's great for those that wish to compete in that way, and presumably that would be difficult to cheat, at least without hacking the website. But I would like to see the "Number of problems solved" maintained. If there are a few cheaters, so be it. You can't forestall every possibility. Provide the data and let those of us that like to look at such things judge for ourselves the credibility of the solvers.

Yes if they were few you would perhaps be right. But they seem to concentrate on particular countries where they aren't few.
For instance out of the top 10 of your own country about 7 are at least of very dubious quality.
So I hope you haven't been comparing yourself with people that haven't been doing anything to place themselves above you several times during the last year.
We were hoping that they would give up in the end but several of them are very pertinacious.
For the team the moment had come that we couldn't go on the way we did the last year.

Regarding following who is consistenly solving new problems you can extract this from following the forum posts.
After all, if members are giving as much as they receive then they will make it a point of sharing their insights when they solve the problems. If they cannot be bothered ever sharing their methods then who cares if they continue solving problems or not; they haven't really engaged the spirit of the Project Euler community.