Hi guys,

I'm a last year student soon graduating (hopefully). I find it really fascinating how the web is changing so many aspects of our life and how it is able to bring people together from so many different parts of the world. That's why I chose to write my master thesis about this subject (I even got to personally interview our dear Faburisu for this )

Anyway, as a case study for this thesis, I have developed a proof of concept for a website that is primarily aimed at language learners. The idea is that by translating and correcting for each other, we can learn languages from each other (not specifically Japanese).

The reason I'm writing about this here is because I now have to gather feedback to predict the chance of success or failure for such a website. For this we have setup an online survey that includes a quick tour around the website.

So by writing here, I was hoping to find volunteers to fill in this survey. It can be filled in anonymously and it would only take 10-15 minutes.

The survey can be found at :

http://www.translapolis.org/tour


Thank you :-)

[ファブリス]

I told Edwin he could post his survey here but asked him to not bump the topic. So here's a little bumpity bump in case people didn't notice it.

Sounds like Lang-8.com

Umikuma Wrote:Sounds like Lang-8.com

This, sounds exactly like Lang-8.com.

Would there be anything to keep this from being abused, for instance by fansubbers who want someone to translate an entire episode for them? :p

*Just to clarify this, I saw that you are limited to something like 140 characters, but I was wondering if you can prevent people from spamming LOTS of separate posts to get around the limit

Google allows for users to augment translations, doesn't it? I always see that 'suggest a better translation' link.

[ファブリス]

Zarxrax Wrote:Just to clarify this, I saw that you are limited to something like 140 characters, but I was wondering if you can prevent people from spamming LOTS of separate posts to get around the limit

Edwin told me something about a token scheme, you earn points and then you spend them, I think that takes care of abuse.

@Edwin: keep in mind this will only be vaguely indicative. What users think and what users do is often very different. You have to read between the lines. More relevant would be to do a lot of "market research", try to find all possible competitors, how do they sell themselves, what is their primary audience, and so on.

I have received great feedback already, very informative.

The reason why I'm passionate about this is because of the potential. But to actually crack the code and go from 'gimmicky' to 'useful' is much harder than I initially thought.

For example, yes, credits / tokens are meant to counter possible abuse. But it's clear that it could lead to alienating users as well. Someone mentioned that a website that forces the use of credits shows an inherent distrust in its users, which is also something that has to be taken into account.

@toberroth, umikuma : it's a cross between lang-8.com and cucumis.org as I discovered later. As Fabrice said, market research will be key to advance.. if we can't do it better than it's not worth it.
@nest0r : yes, they use it to improve their machine translation system, but it's still machine translation and there's no social component
@IceCream : I agree that it's worth to try and see whether it sticks. And great comments about slangy versions

The more time passes the more my respect grows for people like Fabrice who actually manage to create something useful from scratch!

At first read, this did sound like Lang-8, but having visited and taken the survey, it looks quite different. I think it shows promise.