2007-04-27

Reflections: In and Out

• Min Young Nam
• Mu Sun
• Pi-Cheng Hsiu

Red rules:

• No collaboration is not resaerch anymre
• parallelize your lists
• "5 times so far" -- I thought this meant that your message had been said by 5 different people
• "No Visual is No Fun"

• visual information is more effective than words on a slide.

2007-04-11

Weaknesses in my wrtiring

• brevity
  • make the thesis obvious
• balance
• benefit

Steps toward Readable Writing

introduce new terms and jargon at the end of a sentence.

"handbook for techincal writing"

"style toward clarity and grace"

2007-03-28

robustness of wireless networks

• still jargon based, adverse, robustness,
• will cyber vision also be low bandwidth?
• I would like more justification for why this work is important, and how it relates to other work.

• The craft of scientific writing.
• to show something is better, quantify the conditions for success and demonstrate the improvements.

2007-03-16

• I'd like to see more about what you are going to have to do to target this audience? What do you expect them to know/expect?
• answer the mary shaw questions, what came out of that
• graphics are too busy.

2007-03-07

• Beyond current web search

Good

• good job justifying your research.
• first half of presentation much better at

Bad

• how does second speaker relate to entity ranking?
• screen scraping practical
• how are you going to improve entitiy ranking
• give examples when talking about improving an resaerch area.
• definition of entity.

What I can learn

• re-use advisor's ideas and slides
• simple and straightforward example to understand problem.

• Computer aied design for composite systems
• Min Young Nin

• evaluation and automated metrics for system architectures?

Good

• much much more organized, easier to follow
• fits together in a bigger whole, you have obviously thought about the subject more.
• good effort in explaining what your evaluation criteria are.

Bad

• don't apologize.
• What about things like Inferno? Plan 9? Thin clients? Distributed systems?
• talk slower.

• at least go over the evaluation criteria for the final topic you pick
• more research on what has been done before
• give an example for your final selectoin, explain it better.

2007-02-28

• Hoang ngyuen

Good

• beter graphics than last time.

Bad:

• Patter needs to be better rehersed.
• don't be too verbose.
• too much time on the second slide.
• talk slower! I'm lost!
• Make your point clearly and concisely.
• jargon.

run on sentences

Smart Survalence (renamed from Magic Campus)

• minimal sesor configuration that will yeild data for facial recognition in real time.
• senso planning
• facial recognition
• real time middleware

The Good:

• showing where they can get research from
• highlights of what areas you are working on, and how they interact.
• good format -- preview

The Bad:

• hand offs are rough
• second speaker, speak up!
• hard to see bars in lower right corner, better description of what the columns mean.
• derivation of research is interesting, but focus is different from first talk.
• Where do the facial recognition and real time sensors fit in?

good presentation, but the first half had differnt statements than the first half, and those should be reconciled.

2007-02-21

Cyber physical systems

good

• got back on track when you displayed the evaluation chart.

bad

• buzz words, not explanation, just lists of what you would like to see.
• "If you can't explain it, skip it."
• You have to explain what these categories require. It's not coming across. I don't have the knowledge to judge how well your dependent theory validates your requirements.
• explain the dependencies in words

Final summation

• good evaluation chart.
• good job voicing his opinions.

• too many topics
• "If you can't explain it, skip it."
• explain in words what things mean
• difficulty !!= familiarity
• impact evaluation
• biting off more than you can chew -- too many topics, all medium-difficult and high impact.

2007-02-09

• Beyond Current Web Search
• tao cheng, shui lung chuang, Hanna Zhong
• http://parrot.cs.uiuc.edu/entitysearch/large-demo.html

The Good

• useful introduction and definitions at the beginning.
• good explanations of problems.
• good summary of current research.
• good demo

The Bad

• make sure your demos work.

2007-02-02

• Magic campus

The good

• well laid out, very focused.
• good layout of what is involved.
• good definitions in the beginning, introduction to the area
• good pacing per slide

The bad

• privacy issues
• is this a software issue or a litigation issue? A police issue?
• I'm unsure of what is possible w/r/t kullman filtering.
  • really, filtering is side issue, you're focused on the service. There should be more information based on that.
• change in interest between speakers.
• "magic" or "smart" campus?

Good

• communication illustrations were good...

Bad:

• Starting out, NO IDEA what you guys are talking about.
  • first speacker, speak up and don't hurry to get out of the limelight.
• things on slides laid out at random. Very unprofessional.
• desperately needs an outline
• need to answer why?
• living inside a computer -- what are you tellins us?
• The transition slices give us no inforation, nor do they tell which direction you guys are going.
• what is your new idea?
• where are the research challenges? What are you proposing?
• to many workds on later slides "Technolocical benefits"
• what has been done in this area?
• english is off - "not so familiar people"

• not convinced of energy efficiency
• can you guarantee uptime?
• security when every network packet can be sniffed.

Blank title slides are a signal to a viewer that you are changing topics, so they can set what they just learned aside and start learning anew.

2007-01-31

• I like the fact that you laid out what the bounds of what you were talking about in the beginning.
• First speaker - very well articulated, very insightful and interesting guide that gets the audience up to speed on the terminology
• Second speaker - I'm often lost in why you are talking about certain things. Your motivations aren't clear.
  • use more inflection to keep audience interested, highlight points.
• good approach in presenting your strengths directly.
• Third speaker - need more aims of where yous goals interact with the infrastructure you present. It's not obvoius from the diagram.
  • I take it back, good example there at the end

Good:

• I like that summarrized things in the beginning
• good job introducing teminology
• the fact that we are able to comment on feasibility is a very good sign.

Bad:

• middle section was the weakest,
• open/closed problems.

• move example before concise definitions, shorten -- end in catastrophic disaster.
  • revisit when you present your approach
• I missed the motivations during the second speaker. Need more connections in the venn diagram.
• not always clear where the open problems are.
• I'd like to see where your approach falls in the venn diagram you presented.

2007-01-26

research in cyber physical systems nd real time networks

• soft systems, sensors, temperature, chemicals, alams
• hard systems, remote operation of robotics
• wired or wireless

• industrial applications
• medical - remote surgery and assisted living
• safety , defense

Need more examples of how resource planning, etc help your project.

• good list of possible applications. Comment: give us some figures or statistics or results that connect this resarch to the real world? How many people could benefit from asssited living? How much money could be saved through hard real time systems in dangerous industrial environments?
• I did get a laundry list of metrics we can use to judge a cyber physical system, but at the end I'm still not sure what exatly you are doing to solve the problem. I think part of that is due to the having to hurry at the end.
• Comment: I need a connection between time considerations of real time systems to cyper-physical systems. Is this an improvement on what exists? Groundwork research? What's already been done?
• good policy of citations, perhaps give them less focus on the slide.
• Comment: good motivation for why you are working in a particual area, perhaps you could be more specific with your interests?

Good

• Aaron
  • good examples of what exists.
  • good examples
  • good show of challenges.
• pritish
  • good composure working though a cell phone going off.
  • good exploration of what areas you are moving into.
• Abinhav
  • good concise description of the problem and what you are trying to solve.

Bad:

• Need outline of 3 things you are going to be looking at in the beginning.

Ugly:

• I really like your summary of where the challenges are, and why you aren't considering certain areas of research.
• consider time!

• good summary of where the challenges are, and why you've picked a particular area.
• good examples, and good justifications, but I think they need to be more concrete, lecturing to people not in parallel computing. Aaron did a good job of making things concrete.

• I understand that both of the first groups were taking a chance presenting first, and I think the presentation requirements were fuzzy enough that there was alot of interpretation of what you guys should do. Kudos for going first.
  • I support the idea of dividing up work, but I think everyone needs to work together to present a more cohesive package.
  • overlap:
    • aaron and pritish
  • contraditions
    • Abinhav thinks there is too much emphasis on domain specific languages, but pritish thinks we need more domain specific architectures, who's features must transmit up to the language level. How do you resolve these two opinions?
    • Abinhav thinks that it's better to work on scalability, Aaron disagrees.
  • lack of approach