![[spacer]](/schemes/blue/line.gif){kind=small}
|
Genevieve Hudak DLC - Hw 5 Role: Summarizer/Analyzer 1.1 What people found interesting about the article. Most people found the article interesting. In particular, the concept of domain-specific critics interested several people. In general people also found the idea of critics interesting and most of them thought that the critics sounded useful. A few people also liked the idea of linking computers to people in the design process by giving them a useful role. Another interesting point that was made by a few people was that the computer critic has to be filled with knowledge that real people have. Some people addressed this as a question wondering what limits this puts on the computer critic. Personally, I think that computerized critics have the potential of going beyond their hard-coded rules (input by people) and doing a similar thing that Google News does and just go out on the internet and scour for more information relating to the appropriate domain. 1.2 What people did not find interesting about the article. Most people had nothing they found uninteresting about the article. One or two people were not interested in general. Someone commented that perhaps if the critic system were applied to a more useful domain, such as C++ programming, then they would have found it more interesting. 2. What people thought the main message of the article was. A lot of people thought that the main message was that embedded design critics would be helpful in the design process because of the large amount of knowledge in a particular domain. People thought that by putting the burden of knowing everything on the computer, a better design would result, especially when knowledge is properly applied (i.e. the right info at the right time). A similar point was made that if the knowledge of the domain is applied during design, that the design is more likely to be a good one (as opposed to reading about the domain and then trying to do the design). Everyone seemed to agree that this was a good idea and that it would be a useful thing and would ultimately lead to better designs. Many people also pointed out that the critic system still allows for much needed flexibility in design, as well as appropriate feedback. 3. What themes would people like to know more about? Most people wanted to know more. In particular some people were interested in knowing more about how the critics were constructed/implemented. Others were interested in hearing about similar critiquing systems, or what else a critiquing system could be used for. Another question brought up was about how much information a critic should provide and when is it enough/too much? 4. Did people know of other papers, ideas, or systems that were closely related? Pretty much no one had heard of any other DODEs. One mentioned was a home architect program or a home landscaping program. A couple more had heard of other critiquing systems such as Rational Purify Plus, IDEs, and collaborative office systems such as Lotus Notes and Microsoft Office Assistants. 5.1 What did people think the article said about design? People thought that the article said that design could be improved using a critic system, especially in a domain specific environment. This improvement comes from providing appropriate (i.e. related) knowledge at the appropriate times during design. People also thought that the critics helped by helping the designer evaluate their design during the design process as opposed to after it. 5.2 What did people think the article said about learning? People thought that learning takes place during design when critics are involved because they provide instant feedback as well as help along the way. So when someone is not clear how to do something, or if some element of their design is good, the critic provides instant feedback to aid the designer in learning more about their design on the spot. 5.3 What did people think the article said about collaboration? People thought that the collaboration aspect of the critic system occurred during the development/building stage of the actual critiquing system. The collaboration took place between domain experts, developers and users of the system to create the best critic rules. Another point made was that collaboration takes place between the human designer and the computer critic. 5.4 What did people think the article said about innovative media support for these activites? Some people considered the critics themselves to be the media support. Others thought that a different kind of media may have been better for the Hydra system in particular. These included a VR-headset and hyperlinks. 6. Did people have any ideas of how this research could/should be extended based on their own knowledge and experience? One person suggested that this would be useful in circuit design. One question that was brought up was how the critic would deal with the case where multiple rules were broken at the same time. Determining how to present these problems to the user in a way that would help them to resolve it (especially when the rules were interdependent on each other) might need further research. Another person suggested this be applied to software development, in a way that goes beyond IDEs. Someone else suggested that we should see how people feel about already existing critiquing systems such as Microsoft Word Spelling and Grammer. |
