Groom your programming skills and make some money | Hacker News Reader

Groom your programming skills and make some money

amebopost.com

56 pointstundebabzy29 comments

So... you can get up to 50 points for the fastest code, and 20 points for the smallest code. And you only lose up to 20 points for code that doesn't work.

I think I just detected a wonderful loophole in this scoring method.

I think you only get the fast code and small code bonuses on working solutions.

Right, submitting fast but incorrect+buggy code nets you -20 points.

It needs some polishing, there are quite annoying aspects:

1) Mandatory gender and country? Seriously, why? Even e-mail shouldn't be mandatory IMO. Hitting 'Enter' on register/login forms doesn't work as submit, I need to grab the mouse for that (especially annoying on a programming-related site)

2) The "How are programs run" section is completely useless. It's never mentioned how is my code actually invoked, just how it is compiled (when applicable). Let's take C for example, it says 'gcc contestx.c -o contestx.c'. Assuming the argument to '-o' is a typo, this way of compilation includes linking... so it seems I'll need `int main()` to avoid linker errors, but what's to be in `main`, I'm never told. I'm just assuming .c code is just `-c` and `dlopen`ed, but I'm not sure, since 15 minutes after submitting a solution, there's still no feedback.

Other nitpickery includes email validation code that's not clickable (okay, this is arguably a good thing) and other stupid memes implemented on the site like asking for the email twice or showing me my local time in the upper-right corner even though all other times on the site are displayed in the server's time zone (e.g. upper-right corner shows 16:45, but the solution I just sent is shown as submitted at 7:45)

Another interesting thing on the "How are programs run" page is the fact that the programs are apparently run on a Linux system, yet C# files are compiled with Microsoft Visual C# somehow!

Good day

I created www.amebopost.com. I am extremely sorry if I annoyed anyone. If I had known that I'd be getting this many users, I would have endeavored to make it more professional and thoroughly fix the bugs before making the site online.

My original goal was to create an avenue where Nigerian programmers could convene and share programming ideas and code.

The About Page was just me being Silly. It will be taken down.

I mean, is this a joke or a viral site or something? The About page explains that it was done by someone called "Sir Pastor Alhaji Babalawo omotodun, Ph.D, MSc, etc...", which sort of brings to mind the Nigerian emails. Couple this with the outregous "also get paid tons of money($1 for winning a special challenge per week)" and P(legit site) goes really low.

On a different note, in the longest palindrome problem, the C++ function header doesn't pass (or return) references, which seems to be amateurish.

He/she clearly states in the About page that he is from Nigeria.

Yeah, that's what I'm saying. I don't want to disparage this guy too much but if I'm from Nigeria, I would bend over back backwards to appear legit. I wouldn't advertise gimmicks like "cheap money" or "Phd., MSc, etc."

As someone who works for a Nigerian boss, I get very tired of this kind of racism. I see it frequently. I see people refuse to do business with us because of his origin. I also see people try aggressively to scam us because of his origin, assuming we are naive.

But you cannot judge an entire country based on some highly visible scammers from there. Not all Nigerians are trying to scam you. And many Americans ARE trying to scam you. Many Nigerians have left their country, gotten educations, and become highly successful businessmen throughout the world.

This specific guy who runs this site? I have no idea. But I will judge him on his own merits, not his country of origin.

You are absolutely right. Two points, though: (i) As you see from my comments, my judgments were based on the site, the Nigerian thing is an afterthought and (ii) unfortunately, people do have Bayesian prior probabilities, commonly referred to as prejudices. You cannot avoid them and they need to be addressed sometime. What I said was that if I were from Nigeria, I would be work more on my site to erase any spammish connotations. This is probably true of your boss, too.

I thought everyone just assumed that the "Nigerian Prince" thing was a front and it was likely some Eastern European or American behind it all.

I thought the "Nigerian" bit was just to add to the allure or something.

I am surprised that scammers, Nigerian or otherwise, don't know better than to still reference Nigeria in their emails.

I also found it surprising that scammers devote so little effort to proper spelling/punctuation/grammar.

But then it occured to me: maybe errors help filter out attentive respondents, who have a low chance of completing the whole scam funnel. Those people who don't notice the errors, or don't consider them a red flag, are better leads.

In the context of a cliche'd Nigerian origin, sure, mentioning Nigeria puts anyone familiar with the 419 scam formula on guard. But anyone who does reply is likely a fresh, naive prospect. Ka-ching!

This is perceptive. 419 is like any other advertising: if you think it is stupid, you are not in the target market, but someone else is.

Who communicates in all caps? The elderly and the mentally ill. Target market.

The "make some money" part is only $1, unless I'm missing something, and currently there's only 1 contest paying. So that's not really going to be incentive to participate.

It sounds like Project Euler except code size, code speed, and code submission time matter. Neat idea but how's this significantly different? Project Euler already has a huge number of challenges, a leader board, and a vibrant community. What differentiates amebopost?

IMHO code size and code speed is already a great and welcome difference in making this more geared towards programmers than mathematicians.

Implicitly Project Euler already takes speed into account, but more as a step function not as a fine differentiator between players: Either you get the answer in your lifetime, or you should work towards a faster algorithm.

Are you normalizing the execution time across languages? For example, when they introduced java, the USACO assumed that java code would run 5 times as slow as equivalent C code (so they would give 1 sec for C and 5 secs for java)

Mh, i don't think giving points based on code size is a good idea, developers really really shouldn't try hard to minimize the code size, this just leads to unreadable garbage. Time to Solve and Speed to Run is ok, imo..

Why not have "time to submit" points work as follows:

The problem details aren't shown, only a brief description. User clicks a button "accept challenge", is given problem details, and the server starts recording time until a solution is submitted. Base points on that time difference.

Sure, there are loopholes (such as someone else posting the question on another website so you can create a solution before accepting the challenge). I think it's better than checking for new problems and get lucky by noticing a new challenge came out very recently and there are no submissions.

Hopefully the command line for the C compiler invocation is a typo, because "gcc contestx.c -o contestx.c" will overwrite your source with the executable--which (while catastrophic if done at home with no backup) is probably not a big deal, so long as the relevant code length metrics are extracted beforehand...but still amateurish.

It would be nice if Ruby was one of the permitted languages.

Right? They're alienating a very large developer community.

I tried to send them an email about ruby support and the typo on their compilers page but their email form doesn't even work. I take this to be way too amateurish and scam-like.

"very large" and "very loud" are two different things. I'm sure you'll take exception with any metric that shows Ruby to be a less common language but

http://programmers.stackexchange.com/questions/tagged/ruby 127 questions.

http://programmers.stackexchange.com/questions/tagged/python 272 questions.

http://programmers.stackexchange.com/questions/tagged/php 466

In a website for programming challenges, the market status of a language should be the least regarded metric to decide whether to include it or not.

Just check this out: http://projecteuler.net/languages

What about the infamous TIOBE Index? http://www.tiobe.com/content/paperinfo/tpci/index.html There should be atleast 3 other languages before ruby then ;)

https://github.com/languages

http://www.tiobe.com/content/paperinfo/tpci/index.html

I wonder what made Lua take the leap.

The rails ecosystem moved to github bringing along the community. Ruby eggs, one basket.

Reply on news.ycombinator.com