Re:Ask HN: viability of a startup based on screen scraping?
by _delirium on Thursday August 26, 2010 @05:05PM
(#1637166)
> Databases are protected under copyright law.
This depends on the jurisdiction. I believe the EU has a specific database copyright, while the US explicitly doesn't recognize one. A database may be copyrightable under normal copyright law in the US, but only if it's sufficiently creative (e.g. a database containing nothing but a mechanical listing of stock closing prices wouldn't be copyrightable).
Re:Ask HN: viability of a startup based on screen scraping?
by jacquesm on Thursday August 26, 2010 @05:11PM
(#1637182)
That's quite similar actually. The thing you copyright is the aggregation effort, so a mechanical process does not add anything copyrightable.
But annotation and such do add value, as does the original collection of data.
Also, most databases that you can scrape will contain sentinels to tip off the database owner that you've scraped their content. The sentinels are typically bogus records that are hard to spot, including them in your output will do a good job of proving that you ripped the data.
If you're just going to regurgitate the data you would have to have a better excuse than 'making it more accessible and useable', you could offer your services to the current owner of the data. If they abandon the data that's a different story, but it sounds to me like they are not.
Another option is to license the data, simply contact them to ask if they have licensing options, this is the usual way to go about this.
Scraping is also a heavy drain on the resources of the company whose data you intend to harvest, this means that if they sue you successfully for breach of copyright that they have a fairly clear path to claiming damages.
good luck!
ps: if you have a corporate lawyer that would be a good spot to ask for advice.