Seeding a startup without using fake accounts: anyone seed a startup with wikipedia content? How did that work out? | Hacker News Reader

Seeding a startup without using fake accounts: anyone seed a startup with wikipedia content? How did that work out?

2 pointsamichail18 comments

View on HN

As long as the content you're seeding your site with is valuable to your users go ahead and get whatever you can. We got a ton of great content for a gardening application from usda.gov and are able to manipulate it in ways that makes it useful to our users.

I'm not sure how useful this is.

But it would give users some idea as to what the site would look like with more contributions from users.

Someone recommended that I seed http://studystickies.com with wikipedia content.

So I've been working on that and plan to make an update next week with some wikipedia content.

At least in this way, search queries would yield something interesting.

But at the same time, it would be clear that this data was imported from wikipedia. So it's not obvious what sort of impression this would have on visitors to the site.

Also, it's not clear whether wikipedians would be attracted to the site since this is not really a tool for wikipedia .

That site needs an explanation of "what the heck is this?!" either on the front page or linked from it.

See the flash demo. Maybe I should change the text on the flash demo button to "what's this? (flash demo)"

Why would you do something as silly as force people to watch a flash video to know what your site does?

Demo not working for me. Even if it did though, I agree that you should make more apparent the immediate use that the site has for the new user.

The site is up now and includes wikipedia content.

For example, try this query:

http://studystickies.com/#number%20theory

Ok, that makes it much clearer.

Another thing that I found annoying is the 'please wait' screen. It plays around with my mouse cursor and gives me a 'forbidden' (circle with a slash) rather than the proper hourglass/waiting cursor that is appropriate for this situation. You might consider canning the 'please wait' bit entirely, or make it less prominent.

Too complex. I saw that, and I thought - too much information at once.

Your site appears to be down right now but I stumbled upon it once before and not only did I grok it straight away, I also thought it was a great idea. That said, if some people are saying they don't get it, you should do something about that.

The site is up now and includes wikipedia content.

For example, try this query:

http://studystickies.com/#number%20theory

Is the GNU Documentation License that Wikipedia is published under viral in the the same manner as the GPL?

I'm curious what technology you used to grab content from wikipedia?

it's pretty easy to write a scraper in most languages and customize it as needed. I've done in PHP.

Easy, but really if a site is nice enough to provide a legitimate way for you to get data you should respect that and download it.

You can download it here:

http://download.wikimedia.org

Wikipedia provides a query api for accessing pages and links within pages: http://meta.wikimedia.org/wiki/Query Its very useful and I think people should use it if they want to scrape Wikipedia. At least it gives the Wikipedia guys a chance to throttle bandwidth if they need to.

Reply on news.ycombinator.com