大平/査読結果/MSR2014
をテンプレートにして作成
[
トップ
] [
新規
|
一覧
|
検索
|
最終更新
|
ヘルプ
]
開始行:
[[大平/査読結果]]
*MSR2014 Reviews [#n18b490d]
Dear Masao,
Thanks for your submission to the 11th Working Conference...
We received a record number of 112 submissions (85 full a...
MSR targeted solid research contributions. There are many...
Sincerely,
Sunghun Kim and Martin Pinzger
MSR 2014 Program Co-chairs.
**----------------------- REVIEW 1 --------------------- ...
The paper proposes to use linear integer programming ...
maintainers and thus reduce the time to fix a bug. Th...
actually modelled as a knapsack problem; better, each ma...
a servant with finite capacity/queue. Incidentally, bug...
programming but by an external SVM classifier placed in f...
Pros:
- a nice formulation
- interesting results
Cons:
- the way in which time to fix is computed is obscure
- it compares apples with oranges; there are 2 tasks ...
developer and 2) balancing the workload
- the problem is formulated as a static problem though th...
arriving
The paper is easy to read though the English may need im...
impossible to understand the message. The biggest prob...
unclear how the bug fixing time is evaluated and thus...
estimated. This element is critical as it impacts result...
not clearly say what they did or what they suggest sh...
setting, the time elapsed between opening and changing to...
realistic. See for example the bug (reported but the a...
indeed the time is not measured or relevant and as it s...
the number of bug assigned (parameter L) then I have...
difference with the work:
M. D. Penta, G. Casazza, G. Antoniol, and E. Merlo, “...
through queue models,” in Csmr, 2001, pp. 131-138.
Bottom line, in the above work authors proposed a dynami...
and by plugging the SVM component as a dispatcher one ...
system. A similar task assignment problem has given a st...
M. D. Penta, M. Harman, and G. Antoniol, “The use of sea...
to schedule and staff software projects: an approach a...
pract. exper., vol. 41, iss. 5, pp. 495-519, 2011.
A second serious criticism is that comparing bug triaging...
comparison. Of course an SVM (or any other classifier...
developers workload, it is not a classifier task. To t...
they were not aiming at balancing the workload or reduc...
above papers.
-----------------------
- 問題点１：不具合修正時間の計算方法が不明 -- &new{2014-...
- 問題点２：開発者の作業負荷の計算方法が不明 -- &new{201...
- OpenからFIXEDまでの時間を使っているのが現実的ではない．...
- 負荷分散とタスク割当ての問題を比較するのは妥当ではない...
#comment()
**----------------------- REVIEW 2 --------------------- ...
The paper presents an study on two Eclipse projects, for ...
The paper addresses an interesting and important problem....
The paper is based on the assumption that good bug triagi...
One point that could be weak in the study on how SVM-base...
In addition, it is very difficult to estimate how long wo...
Since the rest of the paper, in some sense relies on the...
In particular, there is one aspect which seems to be very...
In other words, the results of the approach proposed in s...
Another threat to validity that is not mentioned is that ...
Page 2: "one of the automated bug triaging approach" -> "...
Page 3: "and a large number of non-active developers." Ma...
-----------------------
- 論文の想定が現実的なものなのかちょっと議論が必要なんじ...
- 問題１：SVMの割当て方法と時間の計算方法が分からない -- ...
- バグが報告されてからの時間なのか，割当てられてからの時...
- 問題２：修正時間の計算方法が妥当かどうかを議論すべき --...
- 学習データとテストデータで登場している開発者は同じなの...
#comment()
**----------------------- REVIEW 3 --------------------- ...
Most bug triaging approaches are not cost-aware. This cau...
Like:
+ Interesting usage of integer linear programming to solv...
+ Two datasets, from platform and JDT sub-projects of Ecl...
+ Experiments show that the proposed approach can improve...
Negative Points:
+ There exists another approach that is also cost-aware, ...
+ The baseline approach that is used is rather old -- i.e...
+ There is a need to compare with a trivial baseline: Giv...
+ The following assumption used to evaluate the proposed ...
-- Assumption: The bug fixing time for a bug of a particu...
-- Concerns: Two bugs can belong to the same component, b...
-- Assumption: Assigning most of the bugs to a few develo...
-- Concerns: The fact is most bug reports are fixed by a ...
+ The value of L affects the number of bugs that are succ...
+ The constraints used for integer programming seems a bi...
+ Effectively only bug reports from one system is analyse...
+ Are the differences in the time needed to fix a bug sig...
+ During assignment, does the proposed approach take into...
Minor:
+ Section 3.1 can be extended to describe integer program...
+ The description of time limitation (L) can be improved....
+ Many of the related works listed in the paper are rathe...
-----------------------
- 問題１：cost-aware研究との比較がない -- &new{2014-05-0...
- 問題２：Anvikは古い．最新の研究と比較していない． -- &...
- 問題３：バグ修正のコストが最も低くなる開発者にタスクを...
- 問題４−１：平均修正時間を用いるのが妥当なのかを検証する...
- 緊急性の高い（重要な）バグとそうでないバグが同じコンポ...
- あるコンポーネントのバグを1つ修正した開発者と1000個修正...
- 問題４−２：OSS開発の場合，開発者が作業できる時間は人そ...
- 問題５：Lの値は割当てに成功したバグの数とバグ修正に要し...
- 問題６：1つのバグが1つの開発者に割当てられるという制約...
- 問題７：Eclipseだけしか分析していない -- &new{2014-05-...
- 問題８：1つのバグを修正するのに要する時間に有意差はある...
- 問題９：割当て時に修正集の不具合が何割完了しているかを...
- IPについてもっとちゃんとかけ -- [[Minor]] &new{2014-05-...
- Lについて理解できん -- [[Minor]] &new{2014-05-03 (土) 1...
- 最新の関連研究がもっとあるだろ -- &new{2014-05-03 (土)...
#comment()
終了行:
[[大平/査読結果]]
*MSR2014 Reviews [#n18b490d]
Dear Masao,
Thanks for your submission to the 11th Working Conference...
We received a record number of 112 submissions (85 full a...
MSR targeted solid research contributions. There are many...
Sincerely,
Sunghun Kim and Martin Pinzger
MSR 2014 Program Co-chairs.
**----------------------- REVIEW 1 --------------------- ...
The paper proposes to use linear integer programming ...
maintainers and thus reduce the time to fix a bug. Th...
actually modelled as a knapsack problem; better, each ma...
a servant with finite capacity/queue. Incidentally, bug...
programming but by an external SVM classifier placed in f...
Pros:
- a nice formulation
- interesting results
Cons:
- the way in which time to fix is computed is obscure
- it compares apples with oranges; there are 2 tasks ...
developer and 2) balancing the workload
- the problem is formulated as a static problem though th...
arriving
The paper is easy to read though the English may need im...
impossible to understand the message. The biggest prob...
unclear how the bug fixing time is evaluated and thus...
estimated. This element is critical as it impacts result...
not clearly say what they did or what they suggest sh...
setting, the time elapsed between opening and changing to...
realistic. See for example the bug (reported but the a...
indeed the time is not measured or relevant and as it s...
the number of bug assigned (parameter L) then I have...
difference with the work:
M. D. Penta, G. Casazza, G. Antoniol, and E. Merlo, “...
through queue models,” in Csmr, 2001, pp. 131-138.
Bottom line, in the above work authors proposed a dynami...
and by plugging the SVM component as a dispatcher one ...
system. A similar task assignment problem has given a st...
M. D. Penta, M. Harman, and G. Antoniol, “The use of sea...
to schedule and staff software projects: an approach a...
pract. exper., vol. 41, iss. 5, pp. 495-519, 2011.
A second serious criticism is that comparing bug triaging...
comparison. Of course an SVM (or any other classifier...
developers workload, it is not a classifier task. To t...
they were not aiming at balancing the workload or reduc...
above papers.
-----------------------
- 問題点１：不具合修正時間の計算方法が不明 -- &new{2014-...
- 問題点２：開発者の作業負荷の計算方法が不明 -- &new{201...
- OpenからFIXEDまでの時間を使っているのが現実的ではない．...
- 負荷分散とタスク割当ての問題を比較するのは妥当ではない...
#comment()
**----------------------- REVIEW 2 --------------------- ...
The paper presents an study on two Eclipse projects, for ...
The paper addresses an interesting and important problem....
The paper is based on the assumption that good bug triagi...
One point that could be weak in the study on how SVM-base...
In addition, it is very difficult to estimate how long wo...
Since the rest of the paper, in some sense relies on the...
In particular, there is one aspect which seems to be very...
In other words, the results of the approach proposed in s...
Another threat to validity that is not mentioned is that ...
Page 2: "one of the automated bug triaging approach" -> "...
Page 3: "and a large number of non-active developers." Ma...
-----------------------
- 論文の想定が現実的なものなのかちょっと議論が必要なんじ...
- 問題１：SVMの割当て方法と時間の計算方法が分からない -- ...
- バグが報告されてからの時間なのか，割当てられてからの時...
- 問題２：修正時間の計算方法が妥当かどうかを議論すべき --...
- 学習データとテストデータで登場している開発者は同じなの...
#comment()
**----------------------- REVIEW 3 --------------------- ...
Most bug triaging approaches are not cost-aware. This cau...
Like:
+ Interesting usage of integer linear programming to solv...
+ Two datasets, from platform and JDT sub-projects of Ecl...
+ Experiments show that the proposed approach can improve...
Negative Points:
+ There exists another approach that is also cost-aware, ...
+ The baseline approach that is used is rather old -- i.e...
+ There is a need to compare with a trivial baseline: Giv...
+ The following assumption used to evaluate the proposed ...
-- Assumption: The bug fixing time for a bug of a particu...
-- Concerns: Two bugs can belong to the same component, b...
-- Assumption: Assigning most of the bugs to a few develo...
-- Concerns: The fact is most bug reports are fixed by a ...
+ The value of L affects the number of bugs that are succ...
+ The constraints used for integer programming seems a bi...
+ Effectively only bug reports from one system is analyse...
+ Are the differences in the time needed to fix a bug sig...
+ During assignment, does the proposed approach take into...
Minor:
+ Section 3.1 can be extended to describe integer program...
+ The description of time limitation (L) can be improved....
+ Many of the related works listed in the paper are rathe...
-----------------------
- 問題１：cost-aware研究との比較がない -- &new{2014-05-0...
- 問題２：Anvikは古い．最新の研究と比較していない． -- &...
- 問題３：バグ修正のコストが最も低くなる開発者にタスクを...
- 問題４−１：平均修正時間を用いるのが妥当なのかを検証する...
- 緊急性の高い（重要な）バグとそうでないバグが同じコンポ...
- あるコンポーネントのバグを1つ修正した開発者と1000個修正...
- 問題４−２：OSS開発の場合，開発者が作業できる時間は人そ...
- 問題５：Lの値は割当てに成功したバグの数とバグ修正に要し...
- 問題６：1つのバグが1つの開発者に割当てられるという制約...
- 問題７：Eclipseだけしか分析していない -- &new{2014-05-...
- 問題８：1つのバグを修正するのに要する時間に有意差はある...
- 問題９：割当て時に修正集の不具合が何割完了しているかを...
- IPについてもっとちゃんとかけ -- [[Minor]] &new{2014-05-...
- Lについて理解できん -- [[Minor]] &new{2014-05-03 (土) 1...
- 最新の関連研究がもっとあるだろ -- &new{2014-05-03 (土)...
#comment()
ページ名:
![[PukiWiki]](image/wadanyan.png "[PukiWiki]")
