By Sarah Elliott, Andrea Neeb, Jessica MacPhee, Maygan Coulombe, Colette Dilkes ALCOHOL POISONING.
Master Defense Moritz Neeb 9.02
Transcript of Master Defense Moritz Neeb 9.02
![Page 1: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/1.jpg)
Tracking Innovationin Open Source Software Projects
Master DefenseMoritz Neeb
9.02.17
![Page 2: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/2.jpg)
Innovations in Open Source Software
● OSS is under a license that allows redistribution and modification
● OSS is developed in an open, “asynchronous” way by developers around the world
● Innovation Introduction in OSS is an orchestrated (i.e. active) change to the development process
![Page 3: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/3.jpg)
Research Goal
![Page 4: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/4.jpg)
Motivation
● The field of OSS is not explored enough compared to its relevance
● Understanding OSS developers can be extrapolated
● Knowledge about Innovation helps new OSS projects
![Page 5: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/5.jpg)
Research Questions
● How are innovation episodes of 2007 continued during 2008-2016?
● What constraints are given for the search for these episodes?
● How can the new data be integrated into existing theory?
![Page 6: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/6.jpg)
Do continuations of2007 episodes exist?
● Around 130 “innovation episodes” were identified for the year 2007
![Page 7: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/7.jpg)
Roadmap
1) Background Information
2) Search Process
3) Results on Concept Level
4) Conclusion
5) Discussion
![Page 8: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/8.jpg)
Background
![Page 9: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/9.jpg)
Definitions
● Innovation: Either the process/system/tool under discussion or the one in place.
● Episode: An attempt by one or more participants to change an innovation
![Page 10: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/10.jpg)
Success
An innovation is successfully introduced, when:
1) it is used on a routine basis and it has solved the problem it was designed to solve, or
2) it attained the goal it was designed to attain.
![Page 11: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/11.jpg)
Introduction Lifecycle
Discussion Execution
Discussion Barrier Execution BarrierUsage and
Adoption Barrier
Introduction Success
Abandoned Dead-ended
Rejected Killed
Postponed
Failed
No Adoption No Goal Attainment
No Execution
Unknown???
Usage
Source: Dissertation C. Özbek
![Page 12: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/12.jpg)
EpisodeTopics
![Page 13: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/13.jpg)
Data that was used
● Mailinglist data● Annotated mails (coding)● Second channel information
– code repositories
– projects website
– wikis
![Page 14: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/14.jpg)
The Search Process
![Page 15: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/15.jpg)
Motivation for a search method
2007 2007-20160
100000
200000
300000
400000
500000
600000
emails
![Page 16: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/16.jpg)
emails @ GRUB
![Page 17: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/17.jpg)
Keyword Search
● Signal-to-noise ratio● Difficulty to find innovations● Evolution of SCM innovation
![Page 18: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/18.jpg)
Identified continuationsper innovation
● SCM – all (8)● GSoC – all (6)● Bug Tracking – 4/6● License – 2 / 5● Version Naming – 1 / 3
![Page 19: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/19.jpg)
Signal-to-noise ratioof different innovations
SNR (qualitative)
low
gsocgit
bzr
svngpl
version name
bug tracker
high
![Page 20: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/20.jpg)
Finding continuations
● Easy to check success, e.g. via– “git still in use?”
– “license still the same?”
● Hard to find unprecedented problems, e.g.– “problems with license/git?”
● But with evidence, hit-rate is very high, e.g.– “there was a chance of version number scheme
since 2007 – when?”
![Page 21: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/21.jpg)
More evidence: developer activity
![Page 22: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/22.jpg)
SCM switches
![Page 23: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/23.jpg)
git is becoming popular
![Page 24: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/24.jpg)
The search process
Summary● Keyword search works fine for most of the
relevant cases● Evidence of change helps a lot
– via emails: e.g. author activity
– second channel: e.g. project's infrastructure
● Pressure from “outside” helps a lot for innovations
![Page 25: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/25.jpg)
Results on the conceptual level
![Page 26: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/26.jpg)
Continuation “capability”
● There were two groups of innovation continuations:– core episodes
– peripheral episodes
● These groups have very different “barrier-heights”
● The visibility on the mailinglist is also different
![Page 27: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/27.jpg)
Core Episodes
● Innovation is used on a daily basis● If there is a problem, it will be discussed
(urgency high)● high numbers of people affected
→ continuation is likely
![Page 28: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/28.jpg)
Peripheral Episodes
● “One-shots” or minor changes● Can still be important for progression● Continuation is unlikely (did not find any)● Probably hard to find on their own
without knowledge of their existence
![Page 29: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/29.jpg)
Innovation Lifecycleis a circle indeed
Discussion Execution
Discussion Barrier Execution BarrierUsage and
Adoption Barrier
Introduction Success
Abandoned Dead-ended
Rejected Killed
Postponed
Problem? / Idea?
No Adoption No Goal Attainment
No Execution
Unknown???
Usage
"State/Perspective Changes"
Continuation
modified version ofsource: Dissertation C. Özbek
![Page 30: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/30.jpg)
Continuation:From problem to solution
● to let this happen, an innovator has to step up again
● evidence of repeated innovation attempts– autotools @ Xfce
– SCM @ GRUB
![Page 31: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/31.jpg)
State change / perspective change:From a solution to a problem
● A successful episode can turn into a problem or a new idea to improve the situation
● evidence:– bug tracker @ GRUB (process is “rotten”)
– SCM @ GRUB (data loss motivates innovators)
![Page 32: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/32.jpg)
Innovation Concepts
Innovation Introduction
Adapters
Enactment Scope
Partial Migration
Radicality
SignalingGarbage Can
ControlIdentifi- cation
Cost
Capability
Effort
Innovation Decisions
Forcing Effect
Voting
Maintainer Might
Representational Collective
Tool Inde- pendence
Time
Source: Dissertation C. Özbek
![Page 33: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/33.jpg)
Results summary
● Keyword search on mailinglists revealsenough material to evaluate the situation
● SCM management systems progression● Lifecycle revision● “Concept” validation
![Page 34: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/34.jpg)
Conclusion
● The high-level “innovation processes” have not changed much
● Mailinglist communication reveals the controversial topics, but “success” is difficult to see (i.e.: silence means a good thing)
● The “concepts” of 2007 are still valid, but might need a restructuring
● The “outside progression” keeps projects evolving
![Page 35: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/35.jpg)
Thanks for your attention.
![Page 36: Master Defense Moritz Neeb 9.02](https://reader035.fdocuments.us/reader035/viewer/2022072917/62e3b3181d16e4395d455948/html5/thumbnails/36.jpg)
Discussion
● Firefox WebExtensions as outside pressure?● Are peripheral episodes important?● Relationship to Q&U research:
the projects' “develop-ability”?
● Your questions?