ds4se / chapters

Perspectives on Data Science for Software Engineering
60 stars 34 forks source link

./GunetherRuhe/RuheNayebi.md #43

Closed timm closed 8 years ago

timm commented 9 years ago

After review, relabel to 'reviewTwo'. After second review, relabel to 'EditorsComment'.

hongyujohn commented 8 years ago

Title of chapter

What Counts is Decisions, Not Numbers – Towards an Analytics Design Sheet

URL to the chapter

https://github.com/ds4se/chapters/blob/master/GunetherRuhe/RuheNayebi.md

Message?

This chapter describes a high-level decision-making process for software products. The authors look at this process from the analytics perspective and define what can be provided at each step.

Accessible?

Yes.

Size?

Yes.

Gotta Mantra?

Section 3 "The Analytics Design Sheet" is not very clear to me. It is not clear how each quadrant (Q1 to Q4) is created, and the relationship among them. For example: it says that "Q3 evaluates data features and provides the status quo in terms of quantity and quality of the data for conducting analytics". You may make it explicit how it is done (automatically or manually?).

For the illustrative example, how the results of Q4 are obtained could be described more. For example, for the point about "Clustering of stakeholders based on the commonality of their preferences against stated objectives", how is it derived from Q3/Q2?

Best Points

An interesting idea of making SE decisions based on analytics.

Minor point

A few typos: select-ed => selected sup-port => support

There is a relate work, which proposes the selection of design alternatives for architectural design: Hongyu Zhang and Stan Jarzabek, A Bayesian Network Approach to Rational Architectural Design, International Journal of Software Engineering and Knowledge Engineering, vol. 15 (4), World Scientific, August 2005, pp. 695-717.

timm commented 8 years ago

need a better reference for [6]

i had a look around and maybe these will do better than [6]:

timm commented 8 years ago

This proposal sounds.. familiar. E.g.

So maybe a little more on related work?

Also, the actual connection to data science is a little tenuous. The idea is good (know the goals before collecting the wrong data) but can you expand on that a little.

timm commented 8 years ago

Also, it would be good to see some full example of using these higher goals to drive data collection and interpretration (e.g. is this a theoretical framework or a case study)?

timm commented 8 years ago

mentioned in #124 #23

maleknaz commented 8 years ago

Thanks everyone for the very good feedback. @timm

@hongyujohn