Read mostly for history...
First off, I paid 99 cents plus P&H. I rated based partially on this. Had I had paid that Verilag-Springer monopoly price I would have been greatly miffed when discovered what I bought.
This book contains a summary of how ethernet/tcp-ip works as if you were trying to explain it to a financial decision-maker. I found little in it as far as actually implementing the protocols. This is such an early implementation that port 80 - HTTP, isn't discussed. However, it does illuminate the implementation layer model and what goes on at each layer and mentions the relationships between other similar protocols, such as Xerox XNS or IBM SNA.
I understood that for a while e-books of this were to be placed into public domain, then some copyright-happy folks decided to hold it hostage for a few more years. My advice is if you are trying to get an understanding of what's going on, without getting into details, ( pre-introduction to Networking ), this is worth a buck or two, but the prices I have seen asked for this book could only be justified if it is a collector's item. There are many, many, many other later commodity textbooks on TCP/IP that have much more detail into how to actually implement networking.
But I also note that this book was written when very little was known about networking, and anything at all was better than the nothing that existed at the time. The information in it was timely at the time, but that was 30 years ago. It was the kind of thing an MBA might read, but it wouldn't have done the programmer much good.