There's a lot wrong here.<p>First, there's no such thing as an "abstract TLD". The closest analog would be <i>generic</i> TLDs. The author actually manages to get this completely backwards since the gTLDs are .com, .net, .org and their brethren. The TLDs the author calls "abstract" are all actually country-code TLDs (ccTLDs), and they all have a proper meaning. .ly is Libya, .st is São Tomé and Príncipe, and .ng is Nigeria.<p>Just because these governments don't traditionally care too much about who uses them doesn't mean they don't have a meaning. They aren't abstract.<p><i>The address bar in Firefox will actually redirect you to google.com if you type just google, however what I am proposing is to actually get rid of the TLD for some websites.</i><p>This is pure fiction. What actually happens is if Firefox can't resolve the address, it assumes it's a search term and passes the term into a preset URL (defined by the keyword.URL setting).<p>Back before version 3.0 (I think; don't quote me on that), this preset passed the "I'm Feeling Lucky" flag to the Google search URL. As a result, if you typed "Google" into the address bar, with no TLD, Firefox would effectively search Google for the term "Google" and the "I'm feeling lucky" feature would redirect you to the first result, which is obviously Google.com.<p>Since 3.0, the preset doesn't have the "I'm feeling lucky" parameter, so if you type "google" into the address bar, you'll actually get a page of google results for the term "google". I, and I'm sure many others, have since tweaked the keyword.URL setting to again use the lucky flag, but on a vanilla install, the author is just plain wrong in his assumption.<p>One final nitpick:<p>* After this recent surge of custom TLDs, I am beginning to think, that they are not a limitation, but that we can change them, almost freely, presumed we have the knowledge to do it (Please, correct me if I am wrong, and forgive my ignorance).*<p>Perhaps English isn't the authors first language, and if that's true then I apologize, but I simply was unable to parse that paragraph. Commas just don't work that way.