Shall creating or cloning services from scratch be considered on-topic?

Question

The current FAQ is pretty generic, and doesn't go into much detail about what's off-topic.

I just reviewed the following: Movie/song database like IMDB.

I would like to flag it for closure as off-topic, but the current FAQ gives me no grounds for doing so. To fix this :-), I will suggest that something like this is added to the FAQ:

The following are considered off-topic at Open Data SE:

• Implementing a service, database or resource duplicating a specific already existing commercial or free service, database or resource.

Edit: Now some other user has changed the question from:

Movie/song database like IMDB

into:

Is there an open movie and/or music database available for commercial use?

I think the latter is a different question than the former, and while the former IMHO should be off-topic here, the latter is clearly on-topic.

(I think it is against the spirit of SE to radically change the meaning of someone's else question through an edit, but that's another topic.)

Answer 1

I agree that the question is off-topic. However, I don't think those additions to the FAQ are the right way to go about this. I could think of plenty of questions about building an open data source from scratch that could be considered on-topic. Here's why the question doesn't fit, IMO:

1. The question isn't specific enough. Poorly-worded questions about how to do some ill-defined thing are generally considered bad form on Stack Exchanges in general.
2. The data they're asking about is generally proprietary/commercial/closed in nature.

The first, I think, is the stronger reason, and could already be covered by the FAQ under "what kind of questions should I not ask here?" The second, though, might not really apply. Although they didn't specify it, perhaps they are looking to create an open database of movies and music. Wouldn't it then be on-topic?

Answer 2

I think you've read the question as if the user is asking how to build a new commercial service that is a clone of IMDB. Whereas I read it as asking how one could create a new Open Dataset that contains similar data, and which can be used commercially.

That seems on-topic to me.

So,I think you're reading too much into that particular question. If there's anything wrong with it then it's that it could be a little more specific.

E.g. are there alternative open sources of movie data? how would one go about crowd-sourcing or collecting new open datasets that are otherwise only available commercially?

I've just reviewed all of the questions on the site and I notice that there are several that have been closed or down-voted just because they cover requests for data in areas that people think are out of scope. There are suggestions that the site if only for Open Government Data, whereas I don't think it is. Personally I don't have any issues with people looking for data on, e.g. fictional characters, surely the role of the site is to support and encourage discussion and discovery of open data, not policing what kinds of data or re-use are considered acceptable.

The FAQ could certainly be improved to help guide people towards asking more useful questions. There are some very broad questions that could do with being more on-topic.

Personally I don't think any of your bullet-points should be added to the FAQ.

The questions of how one goes about building an open dataset from scratch, via crowd-sourcing or any other model is experience that can usefully be shared. So long as we can tease out the critical questions.

Answer 3

Creating something from scratch (or scraping the data) in my view is fine.

Duplicating a commercial resource to, in effect, open the data for unrestricted public use is OK by me.

However, would be really improper to abuse the community's help to create closed datasets. For this, one should hire consultants and cough up real money.