All posts by Peter Mangiafico

Why 2009 is a bit like 1984…

In case you missed this story from the New York Times, Amazon.com has apparently decided that even after you purchase two of their products, they have the ability to connect to your device and “take back” content.  In this case, Amazon realized that a seller of two books that they had sold digitally to Kindle owners did not own the rights to sell the books, and so they decided the best course of action was to stop selling them (fine). They then connected to the devices of individuals who are purchased copies previously and deleted them, providing them with a credit (not so fine).  Ironically one of the books in question was George Orwell’s 1984.

So the seller didn’t have the rights to sell the book, and Amazon is simply protecting the original copyright holder’s legal rights – sounds fair at first glance.  But what makes this a bit unusual is that the users in question had paid for and purchased two products from Amazon: a Kindle and an e-book.  Then, due mostly to Amazon’s poor diligence in vetting sellers on their own store, users had their legally obtained products removed from them.  It’s as if Barnes and Noble came into your house in the middle of the night, and took back a book you bought and the bookshelf it was on.  Is it any wonder that engineering types are nervous about DRM (digital rights management)?  The balance between the rights of users who have purchased content and the license rights of the creators of the content still feels wrong.

Search engine + HTML defeats complexity

I’ve been thinking about the semantic web and I’ve been to many a meeting where discussions of interoperability, standards, and so on have dominated conversations.  I nod my head in agreement and think about how if everyone would just stick with and use standards, all this data would be connected and the Internet would be a far better place.  But there seems to be an opposing current from the likes of search engines such as Google.  Their principle seems to be to lower the need for tight structure and organization of data and just have superb indexing and searching.  Gmail, which pushes tags instead of folders, and lets you search email quickly instead, is an example of this notion.  Recently at eBiosphere, one of the attendees mentioned that he tried in vain to find specific biological data by visiting traditional sources of highly structured data, only to give up, type the phrase into Google and get the answer he needed embedded in an HTML page.  While it would be better if each of those biological projects were sharing today correctly according to semantic web standards, what if Google + HTML already provides some of these answers?  It’s an interesting space, and it’ll be interesting to see how players Wolfram Alpha start to fill it.