Software Engineering discussion

Maybe this chapter should have been in an architecture book instead... or a travel book... or a startup company book... or a sociology book.

I didn't find anything particularly special here. The PGP email client does most of the function described here (and it is "free"). There is certainly room for usability improvement with PGP (I do a similar experiment with my students with similar results), and the screen shots show some improvement in this area. However, the biggest problem with the web-of-trust authentication technique is that it is complicated and no one uses it, yet this technology seems to have been carried over to this project. I totally disagree with the statements that this technique provides better security than certificate-based approaches. The major advantage is that it is free.

I did enjoy reading about how Perl was used in this project, but the author had to recreate many solutions that already exist, like an entire email client. Perhaps there are other plug-in like approaches that would ease the task.

I agree this chapter was all over the place. Some comments on the life cycle of an application rang true.

The web of trust was new to me. Seems like that would be great for distributed computing or federated networks, but I can see how it would be not very user friendly in an email client.