Fred Brooks 12-24
Everyone of the mainframe manufacturer snow white and seven dwarfs missed the mini computer because their business was really good. Micro computers every one of mini computers missed the micro computer, personal computers instead of mainframe, both were fueled by sociological aspects. Beginners make baby mistakes, but experts make huge mistakes, talks about his experience by talking about a language that was designed underneath his manage. Didn't understand schedule time programming language on designing this language. The desiderata is also whats wrong on this model it is always changing. As one wrestles with the design the desiderata changes also. He talks about where he studied payrolls, and how it is always changing in big companies. Discusses his house design and how he had went through so much hassle to meet the requirement and eventually decided he will just have to buy land from his neighbor to meet the requirements it's thinking outside the box. The rational model is fundamentally wrong, it doesn't really describe whats going on. Talks about Germany and United States adopted a design model, and has been modified over and over, and how they have been trying to change it because it doesn't have how real design experts actually work.
Design of Design Chp 6
pg 63- Collaboration in Design, quote from Bernard Baruch- a meeting is a refuge from the dreariness of labor and loneliness of thought.
64- Talks about the two big changes in design recently, most of it is done in teams and that the teams now collaborate using telecommunications. That it is a good thing they are in teams but its not universally true. Have to worry about conceptual integrity.
65 - Saying that most great works over our history have been made by one mind maybe two and that two is the magic number, has a picture of Nautilus Sub which we dont know who the chief designer is.
66- Discusses why engineering design has shifted from solo to teams. Because technology sophistication is the main driving reason for teams. Everything is so made so complex now days to keep up.
67- Talks about how sophisticated it has become for even farmers, baker, and mill workers. That we need the know how of masters from various different fields. Also, driving for for sophistication is need for new products.
68- Costs of Collaboration- partitioning of a design takes more time. Since there is more in design team there is a cost of teaching everyone and bringing them up to speed on goals and details of the design.
69- Also must develop a way to control the change of the design so it is specific for one user and his work. Design integrity and Wren's St Pauls's Cathedral with picture.
70- Discusses in more detail conceptual integrity what it could be and why one or two mind designs usually have more integrity.
71- Discusses briefly why people think his argument is wrong and then discusses the beginning to getting conceptual integrity with a team design. Having one system architect is best way to ensure conceptual integrity.
72- Further discussion of the system architect and then another way to ensure integrity is to have one user-interface designer.
73- Further hitting on why their needs to be one key designer who understands all of it and starts to discuss the benefits of collaboration.
74- Discussion of collaboration working because the designers ask different questions and how their contradiction in opinion can help, and they brainstorm together.
75- Competition as a alternative to collaboration is discussed by having design challenges with known constraints, and discusses Brunelleschi and the Santa Maria del Fior with his radical concepts.
76- Discusses how he has been a judge in a design competition at IBM, and how in Civil Engineering they are more readily to have design competitions for a particular job.
77-Unplanned design competitions happen in product fights or in the market when two products become similar in certain market goal. He also begins to discuss design review
78- Further discussion of design review and why it is important because it involves so many people beyond just the designers and gives and example for a submarine being built.
79- They "fantasy" concept for design collaboration and how it works, and then discusses how it ultimately doesn't work when the collaborators do design.
80- Discusses design control and ultimately relays his opinions back to the example of nuclear sub, and how it is difficult in a virtual reality design to do some things, and conceptual integrity is out the window.
81-Conceptual design definitely cannot be collaborative, and he discusses why two person teams are great throughout the process and relates to other jobs in the world where two people are required.
82- Gives more backing of two person teams iron sharpens iron and how well they did in a study. Also as to why so many computer scientist tools have ever made it into every day use.
83- All of this explains why they should be leery of placing graduate students with no real world design into collaborative design projects, and the rest is notes and references.
No comments:
Post a Comment