Lab 6: Projects: demos, cool things, presentation prep!

[ycoady]

Please post 2 cool things about OTHER projects here!

Also, be ready to demo project 1 for David, and prepare for presentations of project 1 for NEXT week!

Focus on:

1. The problem you were trying to solve
2. Your solution (design/implementation)
3. Experimental methodology
4. Results
5. Lessons learned!

You can use a laptop/blackboard, anything you like for the presentation!

[Hoverbear]

I really liked how @eburdon's project was Creative Commons!

I also really liked how @knowlesc's project looked at io.js which is a most interesting fork of Node. It'd be interesting to see if any of the development methodologies changed with the fork.

[guand]

Devin and Andrew 's project: One of the cool aspects from this project was their use of GitHub's API through a Python script to retrieve useful data. The second aspect is their use of Google Sheets to represent the data in a clean and attractive manner as well as creating an average graph for the three repos they selected which was a nice touch.

Jeremy K's project. I really liked how the graphs concretely show that patch notes and forum activities affect each other. Your thoughts on Dynamic Time Warping sounds very cool too.

[BrodyHolden]

The graphs in this project are very clean: https://github.com/devinc13/SENG371-Devin-and-Andrew

A lot of care was put into the README in this project: https://github.com/eburdon/Seng371-SoftwareEvolutionClassProject

[knowlesc]

https://github.com/PolloDiablo/SENG-371-Project-1 Using reddit as a main data source is really cool and stood out from the other projects.

https://github.com/devinc13/SENG371-Devin-and-Andrew Definitely the nicest looking graphs out of everyone.

[chrisjcook]

Chris Cook, Richard Claus, Sarah Nicholson:

1. @devinc13 - We liked that you actually hit the GitHub API and that your script is project language agnostic.
2. @knowlesc - We're intrigued by the tools you used for analysis - namely CodeClimate. We will definitely take your experiences into account when we attempt to refine the metrics we use to analyse growth and complexity.

[PolloDiablo]

Jeremy Kroeker

1. Parker Atkins, Rabjot Aujla, Greg Richardson, Jordan Heemskerk: I like the project layout, the Wiki was a great idea. Its also cool that your code can be executed so easily; just one main file.
2. Mitchell and Tyler: I love the idea of using C++ and OpenGL to make graphs. Very ambitious :P

[fraserd]

Project Examined: https://github.com/eburdon/Seng371-SoftwareEvolutionClassProject

Erika Burdon's project which is all about the Python programming language -- its power, complexity, and evolution -- does not explicitly mention Python version 3, which was the cause of much drama and frustration during the evolution of Python. Perhaps Python version 3 was more the cause of emotional strife than it was the cause of actual problems or perhaps the strengh of Python version 2 floods the results so that any detrimental effects of Python version 3 are not visible.

Project Examined: https://github.com/Brayden-Arthur/SENG-371-Project-1

Brayden-Arthur's project looks at new contributers to a handful of other projects. The charts included show when the new contributer start as well as the size of the codebase over time.

An interesting aspect of this is that, for most contributors, there is a spike in LOC shortly after they joined the project. This might be indicative of the Law of Conservation of Familiarity in action.

[Brayden-Arthur]

@PolloDiablo I thought that showing the relationship between the forum comments and patch notes was a really cool idea. Seeing the graphs shows a neat analysis and it will be cool seeing the project go forward if that's your plan.

@devinc13 I've got to agree with chrisjcook and knowlesc, your project is the cleanest looking with some really nice presentable data.

[mitchellri]

@PolloDiablo I really liked this one. Though it may be difficult to find out what stimulated the discussions on Reddit and whether or not they influenced the patch, it would be interesting to see, and I think that it may be possible. Reddit posts can also be an interesting metric for future projects.

@devinc13 Completely agree with Brayden-Arthur,chrisjcook and knowlesc. Nice work!

[devinc13]

Devin and Andrew: @eburdon Two things: -The comic in your main readme made my day! -In M1, part B, your scrappy crawler script is awesome! :+1:

@Hoverbear Two things: -It's awesome that you learned a new language for the project! -We could totally use this in our project in the future for finding anti-regressive changes, it would probably be more accurate than our issue/pull request keyword searches!

[eburdon]

Project one: @paulmoon, @guand, @iamj1234 1) I really liked the questions, and thought their hypothesis would be very useful in the project planning stages. A manager could potentially plan on adding more developers once a project has reached a certain size (in order to maintain the rate of new features). 2) It was interesting to see how their analysis proved their hypothesis wrong.

Two: @chrisjcook , @sarahannnicohlson, @371student 1) The graph was really great at showing the near-exact correlation between coupling and growth. Really interesting results! 2) The Snakefood tool was a pretty interesting way of showing complexity

[KRollans]

Jason Syrotuck, Evan Hildebrandt, Keith Rollans

@PolloDiablo It is very interesting to see the way that the developers interact with the community, and how much it can influence the evolution of the games. Also, good choice of games!

@eburdon I think it's very interesting to see how the python language gets used over time and how it becomes more power, my first instinct would say that its a snowball effect. Very similar to Matlab, there is ton's of support for a wide variety of projects. and since it's so popular, people develop more librarys and tools for it, and the cycle continues. I'm interested to read your analysis of the language.

[rsaujla]

Parker Atkins, Rabjot Aujla, Greg Richardson, Jordan Heemskerk

@devinc13 : Your repo is really presentable and especially the graphs give a good overview of the result. @eburdon : Very detailed explanation of question and methodologies. plus the harry potter picture makes it even better.

[gregnr]

Parker Atkins, Rabjot Aujla, Greg Richardson, Jordan Heemskerk

(More comments!)

@devinc13 I found it interesting how closely the bugs followed the progressive changes. I think it's cool how you search for different tags/keywords in GitHub issues for each repo to get the most accurate data to represent progressive/anti-regressive/bugs. For us, we just use the same criteria for all repos, but may be interested in changing to follow something similar to your approach.

@eburdon It's interesting how you are able to break such a general question into three methodologies. Looking at how the types of Python projects have changed would definitely show how Python itself has become more "powerful". How the community has changed would definitely influence the language and the direction it goes in the future.

[Bleech94]

Devin and Andrews project (https://github.com/devinc13/SENG371-Devin-and-Andrew)

This appears to have been a very successful and productive project. The data looks great, and it seems like there is a strong correlation between their hypothesis and the results that they obtained. The two sets of data in each graph are well presented, and clear to understand.

Erikas Project: https://github.com/eburdon/Seng371-SoftwareEvolutionClassProject

Harry potter reference for the win. Also, that is a colossal project, Excellent work! The methodologies employed are insightful and hold up when used to generate her results. The effort put in here is obvious, and it seems like there is a lot that can be learned about Python's evolution.

Jorin Weatherston and Brandon Leech

[RyanMcDonald]

@eburdon I like the amount of detail you put into your three methodologies and I think those methodologies were an effectivechoice for getting an idea of the general evolution of a programming language.

@PolloDiablo I liked your use of graphs in illustrating the correlation between keyword frequency in Reddit posts vs patch releases. Especially with the League of Legends data, you can clearly see patches released after a high amount of keyword occurrences.