Essential Craftsman is fantastic. For anyone who doesn't know, the concept is that he is building a spec house and walking you through every step of the process, from surveying the site, to getting permits, all the way through building it.
One of the strangest business practices Greyhound has is the $18 Gift Ticket Fee [1]. Want someone to buy a ticket for you online? That will cost an extra. Think you can avoid it buy going to the station and buying it in person? Nope, the fee still applies.
Last month the network CEO and Mayor of Chattanooga did an AMA where they discussed setting up the fiber network and how they overcame cable operators attempts to block the network.
When online sales first took off, credit card theft was a huge concern. Even though nothing would go wrong for the vast majority of people fear was enough to make users and vendors go to great lengths to protect data. Not a perfect analogy but conceptually similar.
The reason Blackboard doesn't have competition is due to the patents they hold. They have made a commitment not to sue free, open source projects, but as soon as you start charging money they are ready to protect their IP rights.
I phrased that poorly. You are absolutely right, peer reviewers are unpaid.
I also have a big problem with the fact that journals receive a perpetual copyright to the work instead of it becoming an open license 6-months to a year after publication.
Returning to the cost issue. The cost I am referring to comes from paying for the software to manage the peer review process and the time it takes to build the relationships to have enough reviewers available to deal with the first submission and the revision that will almost likely occur.
To put out a single issue with 20 articles can easily involve 50 reviewers and at least 40 authors.
It is the social science and humanities journals that have real problems with getting papers ready for print. It's easy to only think about scientific journals but the reality is a great deal of researchers only have sufficient computer skills. They will write a professional paper and do the best they can to format it but it isn't anywhere near ready to send to the printers.
Large journals can easily cover these costs. Small journals are really struggling to get by. I hear small journal editors talking about how long they will be able to survive. They want to make it work but just don't know how.
arxiv.org has >700,000 open access articles in Physics, Mathematics, Computer Science, Quantitative Biology, Quantitative Finance and Statistics. It's a great resource.
The major cost of a journal is the peer review process, editing, and printing. This can really take a substantial portion of someones time. I don't think it justifies a $25 (sometimes over $40) fee to see an article printed 8 years ago though.
I have seen substantial Federal grants that restrict overhead a university can take to ~20%. These grants were tied to the PI and the money follows the PI if they change institutions.
I was a talk by Stephen Wolfram at MIT in fall of 2010 about Wolfram Alpha. Never spoke to him personally. When Wolfram answered questions he was always direct and gave grounded answers -- not dreams.
More than once he had to remind people that a lot of the shortcomings they point out are not trivial. It's easy to see how someone could interpet him as egotistical for this. But Wolfram seemed much more focused on building a product now and making it better as you learn from customers.
Wolfram is not trying to make everyone happy. It seemed like is he is trying to build something he loves doing and see if others find it useful.
The story comes from a friend who was involved in the project. If you have any specific questions I would be happy to ask them on your behalf.
A quick google search returned a news item posted on the corporate website of Coinstar which claims the machine processes 600 coins per minute. They also have some general information about customer behavior on the page as well. http://www.coinstar.com/us/WebDocs/A3-2-2
The machine is able to calculate the total change deposited almost instantly. Yet, during testing the company learned that consumers did not trust the machines. Customers though it was impossible for a machine to count change accurately at such a high rate.
Faced with the issues of trust and preconceived expectations of necessary effort, the company began to rework the user experience.
The solution was fairly simple. The machine still counted at the same pace but displayed the results at a significantly slower rate. In fact, the sound of change working the way through the machine is just a recording that is played through a speaker.
Altering the user experience to match expectations created trust and met the customers expectation of the necessary effort to complete the task.
In some respects HN is a Gopher server adapted for today. Focused on content, links to information, and just enough detail to keep you going.
As a protocol, Gopher is straight forward. Whenever someone talks about the "semantic web" I'm always reminded of Gopher has been doing since 1991. It is in no way perfect, but worked well.
Caller ID is really important for me. Unless I know who is calling I will probably ignore my phone. Spoofing the phone number of the actual business would be really handy and greatly increase the chance that I take the call.
I can't tell from the site whether one can do this without signing up for an account but if it's not a feature yet I would seriously consider it.
It's going to take him a long time, but anyone who wants to catch up he has a playlist of the first 32 episodes at https://www.youtube.com/playlist?list=PLRZePj70B4IwyNn1ABhJW...