« Obscene burgers, stop the madness | My twine.com saga continues » |
Thanks to the invite I received I now know the answers to the questions I previously had posed. Twine does not retrieve web page info when a user creates a bookmark, instead they offer an edit box for the user to enter a description. One Twine user told me that twine will spider a site when a user doesn't add a description but my quick perusal of the twine website I couldn't verify this. I'd like to make it clear that Twine.com does follow all the requirements of the DMCA with respect to removing infringing content quickly and efficiently but it must be the frustrating to the person who shared this information to see his hard work disappear. I'm pretty sure he added it to Twine in good faith.
Twine.com like many other sites that are built around user generated content should assist their users to stay on the straight and narrow with respect to copyright. Though I'm a bit foggy on the whole concept of what the semantic web is all about it would seem to me that when you get a URL and a description from a user you might want to try and check automatically to determine if the description is the content. People are generally lazy and hoping they will produce useful summaries of web pages they bookmark is probably wishful thinking. Producing a page with useful summaries of content would help to differentiate twine from Google.com which gives a one line summary and Mahalo.com or delicious.com which provides bare links. I recently join delicious.com so I could make my bookmarks portable and I could see myself using Twine instead if there was some added if this and some other enhancements were added. This would also help reduce the workload on the person who has to responds to DMCA take down notices. At a minimum Twine.com should add some text to their bookmarking tools to remind users that adding all the text from a page to the description will probably result in the removal of their public bookmark.
I wish Twine.com the best of luck in getting through their beta period and working out all the kinks. Though I'm sure Twine would have preferred me becoming familiar with them in different circumstances I'd suggest you check them out and give them a try and help them get better. I'll be checking back with Twine.com and seeing how things develop.
In our next release we will be generating summaries by analyzing the full-text of every URL submitted to Twine.
Going to the URL and grabbing the TITLE text from the page there is a fairly traditional way of grabbing a one-liner that says what the page is about; granted, though, some people create pages with broken, vanilla, or missing titles.