Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Crowdsourced Coders Take On Immunology Big Data

Soulskill posted about a year and a half ago | from the still-no-cure-for-cancer-oh-wait dept.

Medicine 31

ewenc writes "Mercenary computer coders are helping scientists cope with the deluge of data pouring out of research labs. A contest to write software to analyze immune-system genes garnered more than 100 entries, including many that vastly outperformed existing programs. The US$6,000 contest was launched by researchers at Harvard Medical School and Harvard Business School, both in Boston, Massachusetts. TopCoder.com, a community of more than 400,000 coders who compete in programming competitions, hosted the contest. The results are described in a letter published this week in Nature Biotechnology."

Sorry! There are no comments related to the filter you selected.

Genes as strings (2)

K. S. Kyosuke (729550) | about a year and a half ago | (#42838153)

With the help of employees at TopCoder.com, Guinan’s team created a contest that expressed the problem in generic, non-biological terms, such as strings and sub-strings instead of gene sequences and gene segments.

That's funny, because I can't remember that any other approach than a stringological one has ever been used.

Re: Genes as strings (1)

Nic (2818063) | about a year and a half ago | (#42838447)

I agree!

Crowdsource Coding? (2, Insightful)

Anonymous Coward | about a year and a half ago | (#42838695)

Next up on Stack Overflow:

Can some1 pls post the codes for in generic, non-biological terms, such as strings and sub-strings instead of gene sequences and gene segments. full codes pls email me at topcodersplspostfullcodes@hotmail.com

What a fuckin' scam...FTFA

The US$6,000 contest was launched by researchers at Harvard Medical School and Harvard Business School, both in Boston, ....

Their contest ran for two weeks and awarded weekly $500 prizes to top performers.

In total, 122 people submitted computer programs to characterize the genes involved in immune responses. Half of the entrants were professional computer programmers, yet none worked as computational biologists. Contestants spent an average of 22 hours on the problem, accumulating a total of nearly 2,700 hours of development time.

Who do you want to bet owns the copyright?

How about open source instead of $6000 tossed at a starving pack of desperate coders to solve "Big Data" problems for them.

Re:Crowdsource Coding? (2)

thoughtlover (83833) | about a year and a half ago | (#42839787)

Exactly what I was thinking! In the world of the graphic artist, you see this shit a lot... It's called 'spec work' and it's the reason that there's fewer and fewer good-paying graphics positions out there. Hungry artists make logos, interfaces, and designs with the hope that their submission will be chosen. Who wins? The person on the receiving end. What a fucking joke. Seriously people, DO NOT do work for free! This is reminding me of the recent post where the submitter asks, "How much should I reveal in an interview" --Really?? I had no idea that people were becoming that scummy (read: greedy) that they give people the false hope of a job while trying to fish answers out of them to help them with their problem.

spot on, its a freelancer site dressed up (1)

decora (1710862) | about a year and a half ago | (#42840863)

as some kind of leet hacker haven. its full of the same horse shit you find all over those places

"need powerpoint conversion ASAP!"

"java lx2e zorbog buzzheavy lightyear lcick layer"

paywalled? (1)

v1 (525388) | about a year and a half ago | (#42838433)

The results are described in a letter published this week in Nature Biotechnology [nature.com]

looks to be paywalled, @ $32 for a single article?

not much return (4, Insightful)

v1 (525388) | about a year and a half ago | (#42838713)

It's too bad the winning entry, at 970x the speed of the algo it replaced, only received $6k. Surely this was worth a lot more to the eggheads than that? You'd have difficulty contracting even simple, low grade code for that amount?

Re:not much return? think again. (3, Interesting)

Kozz (7764) | about a year and a half ago | (#42839339)

It's too bad the winning entry, at 970x the speed of the algo it replaced, only received $6k. Surely this was worth a lot more to the eggheads than that? You'd have difficulty contracting even simple, low grade code for that amount?

I think you're overlooking the fact that a coder who wins the contest gets something far more valuable: a demonstrable proof of one's mettle and a fairly admirable accomplishment that can only pay dividends for years to come when they're hired by a company who pays them what they're really worth.

Re:not much return? think again. (2)

v1 (525388) | about a year and a half ago | (#42839409)

I think you're overlooking the fact that a coder who wins the contest gets something far more valuable: a demonstrable proof of one's mettle and a fairly admirable accomplishment that can only pay dividends for years to come when they're hired by a company who pays them what they're really worth.

true, but in the short-term, bragging rights and resume' bullet points don't pay the bills.

Re:not much return? think again. (1)

Kozz (7764) | about a year and a half ago | (#42839485)

If you're conquering the challenge for the short-term, you're doing it for all the wrong reasons.

Re:not much return? think again. (1)

v1 (525388) | about a year and a half ago | (#42839613)

It's a matter of priorities. If you're lost in the wilderness, you work on obtaining (1) shelter, (2) water, and (3) food. in that order. It doesn't do you any good to have a good food and water supply if you freeze to death.

If you want a resume bullet or experience, that's what workstudy and unpaid apprenticeships are for. This was supposed to be offering financial reward. And I'm sure it also factored in the bullets, but I think they assigned the bullets too much value.

I think if I was on a big project and found someone that gave me a way to save at least hundreds of thousands of dollars in my budget, they'd get a lot more than 6k and a handshake. More like 20k and a job offer.

Re:not much return? think again. (1)

easyTree (1042254) | about a year and a half ago | (#42839667)

The whole premise of contests is a scam. Everyone works their ass off for some 'prize'. Only one wins yet the contest hosts get to benefit from all entries.

Re:not much return? think again. (1)

Morpf (2683099) | about a year and a half ago | (#42842079)

That is why I only take part in contests, where my work is not usable in any production environment. Contests for the contest sake. Google CodeJam, Project Euler, ACM ICPC, you name it.

Re:not much return? think again. (1)

Jmc23 (2353706) | about a year and a half ago | (#42851887)

Whine, whine, whine. Why such a whiner if you couldn't even win the prize? $6k for 2 weeks of work is nothing to sneeze at. Some strange people aren't even greedy self-important pricks and actually feel that the advancement of science is payment enough.

Re:not much return? think again. (0)

Anonymous Coward | about a year and a half ago | (#42840035)

true, but in the short-term, bragging rights and resume' bullet points don't pay the bills.

But $6,000 for two weeks of work does.

Re:not much return? think again. (1)

doesnothingwell (945891) | about a year and a half ago | (#42841621)

when they're hired by a company who pays them what they're really worth.

Same kid of shit my old boss would say like "when our ship comes in". The bigger boss pulled him aside and told him not to say things about when the important people got rewarded, and that we didn't belong to that group. My job and yours is to weigh anchor and row vigorously while they water ski.

I got the last laugh, the company barely exists now, that ship was really a trap, our captain gave away the company jewels without verifiable data metrics.

Re:not much return? think again. (1)

Morpf (2683099) | about a year and a half ago | (#42842055)

But what if the company supposed to hire this winner just starts a TopCoder Contest, too? This example shows quite clearly that there are people out there selling their souls for bragging rights. Why do you think, this will not be exploited? Remember: That contest yielded 2,684 hours of development time with an overwhelming result for just 6k USD, there is no way to get cheaper. You can brag, that's it. I think those competitions ruin the income of software developers in the long run. Not to mention that someone has to maintain the code.

Re:not much return? think again. (0)

Anonymous Coward | about a year and a half ago | (#42845341)

> That contest yielded 2,684 hours of development time with an overwhelming result for just 6k USD

Go to Africa, pick a kid you can't even read. I'm sure that he will sell you 10 000 hours of development time for just $1000. There is no way to get cheaper. Oh, does it matter that he can't code? So suddenly we are talking about quality here. Well that will change things. I'm a software developer and I am not afraid of these contests, because it would be pretty difficult to run a software company using just contests.

Re:not much return (0)

Anonymous Coward | about a year and a half ago | (#42839891)

There are plenty of mapping algorithms which can do this (BWA, TopHat2, BFAST, Lastz) - map a sequence back to reference (with gaps and mismatches) and they're extremely fast. Not clear why the authors of the paper didn't try any of these out before using topcoder.

I wonder who maintains the code once the contest is finished. If there're any edge cases that are not present in the test sets that are used to judge the competition.

Re:not much return (1)

Daniel Dvorkin (106857) | about a year and a half ago | (#42841563)

You're trying to bring facts and logic into the narrative of "1337 h4x0rz are sooo much smarter than those hidebound academic 'scientists' with their fancy 'degrees' and expensive Big Science labs!" that we're all supposed to love so much. Stop that.

Re:not much return (0)

Anonymous Coward | about a year and a half ago | (#42865227)

eh? more for a 2 week contest? with a 970 fold improvement, you would think that the code he was replacing was itself low grade. Assuming he worked 40 hour weeks for 2 weeks he was making $75/hr, not bad. More than a lot of CS grads make out the door.

Open government (1)

robot5x (1035276) | about a year and a half ago | (#42839159)

Very timely piece, following on from the Slate bio of Aron Swartz and O'Reilly's decision to make 'Open government' free for download.

I've just finished reading the first few chapters of that book and i have to say it really blew my mind. I work in a quasi-IT role in a public sector health organisation; it's really amazing to me how difficult everyone makes it all out to be. Our IT systems are cobbled together from a range of hugely expensive 3rd party solutions - none of which inter-operate and all of which require a change request, 6 months of paperwork and a cheque for at least $50k to make very minor modifications. As it says in 'Open government', there is now an expectation that any solution involving computers or data has to be hugely expensive and time-consuming.

So, I'm really inspired by this story. It says to me that a bit of openness and thinking outside the box is a Good Thing. I'm submitting a paper soon recommending that we develop a strategy moving towards more open platforms and, yes, even merge our IS and HR thinking to do something like competitions and code-outs to get the community and CS enthusiasts working on real world problems.

This begs the question - why is there so little of this thinking currently in the public sector?? Maybe that's a debate for another day!

how lond have you worked there? (2)

decora (1710862) | about a year and a half ago | (#42840899)

in the public sector, your dick is measured by how big your budget is. why the fuck would you want to spend LESS money than you did last year? then your budget gets cut - not only do you have to fire some people, you destroy any chance of advancing up the hierarchy. not only that, you dont have anyone to sue if the 'solution' is behind schedule or doesnt work or, in healthcare, kills someone (Therac 25 anyone?).

Also, you typically get kickbacks from vendors for 'implementing' their 'solution' after a 'careful decision process involving all the stakeholders'.

The only people who believe in this open stuff are the nutcases like, you know, doctors and scientists. What the fuck do they know?

Re:how lond have you worked there? (1)

Jmc23 (2353706) | about a year and a half ago | (#42851919)

Some rare people have this thing called a conscience. You might want to research it.

Re:Open government (1)

sorisos (2702365) | about a year and a half ago | (#42841965)

Perhaps the lack of out-of-the-box thinking in public sector is because it's not a wide spread practice and you have to fight a lot of bureaucracy for these kind of projects. Also it might be hard to fund a project without any guarantees of success, as compared to buying proprietary sofware x with a huge spec list.

Shitty prize (1)

Alejux (2800513) | about a year and a half ago | (#42842557)

Crowd sourcing is a great concept, but this sounds like an abuse of it. It's like someone thought: "Hey, let's throw some bread crumbs and watch these suckers do the work for us."

Re:Shitty prize (0)

Anonymous Coward | about a year and a half ago | (#42842937)

Crowdsourcing, for all but the sexiest projects, will not work for pennies forever.
Let's do the math:(fake math, bear with me)
Number of entries: 20
Losing entry time investment: 40 hours (Avg)
Winning entry time investment: 60 hours

Total dev time: 40 * 19 + 60 = 820
If prize is $6000, expected pay is 6000/820 = $7.3
US Minimal wage: $7.25

Note: I did NOT choose the numbers to make the payoff match min wage, it just came up this way.

Where the code is (0)

Anonymous Coward | about a year and a half ago | (#42847843)

Everybody yaps about it, but don't tell where the code is. Here's the link: http://apps.topcoder.com/wiki/display/tc/Marathon+Match+Overview

Working for free (1)

MotorMachineMercenar (124135) | about a year and a half ago | (#42849199)

Coders work for free? Looks like they've taken a tip from gaming companies, which do Q&A and product testing by outsourcing it to gamers who do it for free - or even pay for the privilege, as has been seen in various betas requiring payment.

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?