For those interested, there are 1826 days in 5 years (one will be a leap year, so 5*365+1), which equates to 43824 hours. So, 20000 hours is 2.28 years. Assuming he worked for 8 hours a day, 5 days a week, he spent 10448 hours at work. This leaves him an average of 7 hours, 20 minutes "free" per day. Clearly, most of this would have been consumed by sleeping, but all of this still does not mean he was addicted.
Granted, 1and1 isn't a great host (and I've got my eye on a new host/registrar), but there haven't been any service outages and the services supplied by 1and1 meet the site's meager needs.
(I wasn't about to hack onto their computer and place a text file, I think that'd be worse)
I actually did exactly that about 5 years ago. A neighbor in the apartment building I was living in had an unsecured wireless network. So, I took the time to type up instructions on how to secure their wireless network and saved the text file to their hard drive. The only difference is that I didn't have to "hack" their computer to do this. All I had to do was switch my workgroup to the Windows default WORKGROUP and I could upload files to, and download files from, their computer. Really, in most cases there shouldn't be any "hacking" required. After all, if they are unsophisticated enough to not secure their wireless network, then their computer is not going to be any harder a target.
My neighbor never did secure their network. So, I can only guess they never found the text file.
Why should I write this tool myself? I can sign up for Tynt (and did some time ago) and use a tool that has already been written. The same thing can be said of Google Analytics. Given enough time, knowledge, and incentive, I am sure I could write a similar tool myself. Again, why should I? Someone else has already written it and it does what I need it to do. Since you're such an advocate of writing your own software tools, then I would expect that you've written your own operating system/browser/whatever entirely from scratch. No? Certainly, anyone using an operating system/browser/whatever could have written it themselves instead of giving money to a third party.
Furthermore, I very much doubt that *anyone* who uses Tynt on their site could have written it themselves. I've seen it in use on smaller websites created and maintained by people who are clearly not programmers, but they are able to follow directions on copying and pasting some pre-written code into their site.
Since Google Analytics is also building a massive database of viewer data, then I assume you'll be happy to rant against that, as well?
I use Tynt's service on a cattery website for which I am responsible (www.tabaxipixiebob.com). The primary reason I use Tynt's service is for the attribution link because I need to know if another Pixie Bob breeder copies the site owner's content and uses it on their site. This has happened to at least one other breeder that I know of, where that person's personal information (information about their employment, the name of their spouse, etc) was copied and used verbatim. So, if something like this happens, then I need to be able to contact that breeder and negotiate a change to the content of their site. The attribution service also can help generate backlinks to the site, assisting with search rankings and such.
The other reason I use it is because I want to be able to improve the content on the site. So, if I know what content is being highlighted while being read, then I know what parts of a page are generating interest. A heat map, of sorts, and content that is not highlighted as often probably needs improvement. I view this aspect of Tynt as no different than a writer or movie producer lurking in forums where their work is being discussed to find out what parts are most interesting and what is most criticized/lampooned. I'll grant that I am not *asking* visitors whether they want to share this data, but at the same time, it isn't like I can exactly have a conversation with even a small fraction of the people who visit the site since I have no idea who they are or how to contact them. If I could round up a focus group and get some feedback, then I wouldn't need to use Tynt for this (but I would still use it for the attribution part of the service).
Hope this helps to demystify the use of Tynt a little for you.