Dr. Brooks
Addison Wesley Longman Inc.
Chapter1-3:
In these chapters we learn about a few fundamental concepts of computer science. First is the idea of the tar pit. The simple idea that many computer scientists who thought they were the best in the world have found themselves skining into the tar and becoming unable to find their way out. They are discouraged when a duo of people in their garage have accomplished something similar to theirs without realizing the degree to which it might work. The idea is put out to not get to overwhelmed by what others are doing and to focus on what you are doign and not everyone else. It is also mentioned to bear in mind the real tiger is always better than a paper one. The next is the premise of the book which is the mythical man month. The idea that man power and months of time can simply be traded based on the number of people doing the task and the amount of actual time is spent working on the gross of the material. As we know this is not true and the author talks about how adding more people can sometimes simply slow down due to having to teach everyone the system, get them caught up, and then divide tasks even further. It might also be that some of these people do not have enough programming experience and can end up only contributing little. Simply put, the idea that you can trade men for months is a poor notion and that it needs to be balanced correctly in orther to work well. Lastly the idea of the surgical team. This is the idea that each team of a group needs to be about 10 people and needs to be setup in a way such that each member has a job on the team and can perform their job well. It all starts with a head surgeon and then you have the co-pilot, the expters, the secretaries, the brains, and so on. Each person does a little part for the team and make sure everything goes out. He doesn't go into great detail about each of the jobs but simply an overview of each and how the team should interact as a whole. He ends by talking about how this can be the best unit for breaking a 100 year job into a 1 year job.
Each of these chapters had something unique about them. It wasn't that the information was new or that it was a clever idea or scheme that was not considered it is the way in which he presented the material that made it very easy to read. The simple idea of comparing the overhead of programming to a tar pit or the idea of having each group work like a surgical team with each person performing a job and everyone working to get a single task done. It is one of the things that is almost directly applicable to computer science where we can say, "yes we can have this person do this and this person this etc..." each one doing a single job and worrying only about their end. That is not to say that if someone is slacking jobs cannot blend but for the most part but each person primarily should stick to theirs. The idea of the man-month was a bit more confusing but i think the premise is correct that having more people does not necessarily get you to finish something faster. You ahve to consider other ideas such as how to train the people and how to get them up to speed on the current system, it is not a cut and dry matter of throw more people at it until it works. All of them are great ideas and as I said were presented in such an easy to understand way that it made the reading enjoyable. When I am less tired I will probably enjoy this quite a bit more.
No comments:
Post a Comment