Comment by sparkzilla | Hacker News Reader

Comment by sparkzilla | Hacker News Reader

I run a crowdsourced content site that is a kind of Wikipedia for news (see my profile for link). When we first started we wanted to attract Wikipedia writers, who write for free. Those writers are mainly young, white Americans, like the Mechanical Turk writers mentioned in the article. But they wouldn't come over, because they like fighting on Wikipedia for free.

So I offered $1 per post to write 50-100 words summaries of the news. Depending on the content a writer could add four to eight posts an hour, getting faster at it as they improved their skills. We had about 50 writers on the program, but could easily have had many more. We were overwhelmed with people wanting to join the program and had over 1000 people on our waiting list.

At first, But the $1-per-post program attracted a whole different community: mainly women, many of them minorities, who like to work-at-home. We actually saw this as a big plus because they were no-nonsense about doing the work, and gave diversity to the content, something Wikipedia struggles with. We paid out in Paypal and some people would literally do some work so they could get some food for that day.

I think too many companies think of crowdsourcing as a way to get cheap work, but it's really good to get diverse workers.

After paying out $20,000 for 20,000 posts we changed to a revenue-share system, where writers earn more as the site grows. That presents a different challenge: finding writers who will put in work now, with the hope of a bigger payout later.

Would be interesting to see some side by side comparisons of the content written by the minorities versus white male teen suburbanite (ideally evaluated objectively by someone who doesn't know who wrote it) to see what the tangible results are of a diverse writing staff.

The amount of "creative" writing on our site is very limited. Each post in our system has to follow a particular format, and we have a feedback loop so that the writer can improve their posts, based on those formats. For example a movie post is formatted in a very specific way [1].

I found no difference at all between the output of minorities and white writers. A bigger and more rigorous sample would have given more insight. It's interesting because Wikipedia is known for its gender gap: 90% of its writers are young, white, highly educated males. So therefore most people think that Wikipedia-like content can only created by those kind of people. Instead, I found I could create almost exactly the same content by restricting the format and the users actions, irrespective of the class of person adding the information. In fact, as the site stands it was created maybe 70% by women.

[1] http://help.newslines.org/knowledge-base/add-a-movie-post/

Oh ok. I remain always curious about the benefit of diversity in these projects then. It is constantly touted as a good and it makes intuitive sense to me that it would be, but I was wondering if that manifested in the content they produced in particular.

I see what you mean now. In our site, a larger percentage of women contributors doesn't affect the quality of the posts, but does add more diversity to their content. For example, women are more likely to add news about magazine covers, music performances, and celebrities, while men seem to prefer to add posts about politics and issues. If we only had men contributors then the latter would be overrepresented.

We can see the effect of the gender gap on Wikipedia in the content too. Topics like porn stars, games and politics are overrepresented, while other topics that women would be more interested in, are neglected.

> "finding writers who will put in work now, with the hope of a bigger payout later."

It is weird how corporate success demeans those at the bottom. I mean ... you've just exposed "the great exposure" fallacy.

"Work hard enough and we will start to include the carrot."

I'm not really sure how giving writers an opportunity to use their time as an investment demeans them, but perhaps I've read your post the wrong way. There are three ways to get people involved: 1. Pay them directly; make a system where they work for free for a noble cause; or let them invest their time now with the hope of larger revenue-share payments later. This latter way actually empowers contributors by treating them like investors.

I see no difference between an investor putting in $1000 in cash, and a contributor investing $1000 of their time, because all we would do with the investors $1000 would be to pay the contributors. In the case where the investor adds cash, as the company grows the investor gets the long-term rewards, while the writer just gets paid $1000 and has no long-term position. However, when the contributor invests their time, we don't need the investor. Instead, the contributor gets the long-term reward for their work as the value of their content rises. So rather than consigning contributors to a fixed wage and giving the benefit of their work to investors, we allow writers to invest in their own future. Of course, that all depends on whether they think the business will grow so that a decent amount of revenue can be shared.

As I pointed out above, many of the work-at-home workers need to get instant cash so this system doesn't work for them, and the Wikipedia people will work for free (the fools!). Our challenge is to recruit people who are willing to treat their participation as an investment. But if it works, they will make a lot more money than they would in any other system.

How did you recruit those people?

There's a large-work-at-home community on the web, with sites that list sites, as well as forums and review sites. Search under "work at home". We also got good word-of-mouth because the pay was reasonable, and the payouts were fast, although you continuously have to reassure people that it's not a scam :-(

>"although you continuously have to reassure people that it's not a scam :-("

It's unfortunate, but that is the nature of the broad term "work from home", especially on the internet. I think tech-savvy individuals have, in parallel, been constantly telling individuals that such keywords are usually scams.

Most of these schemes put front and center the idea that you're working at home and getting paid to do simple tasks. They aren't putting up the actual requirements of the work or why you're doing it, which is what a serious employment offer would be concerned with.

Reply on news.ycombinator.com