./etbarr/naturalness.md

timm commented 8 years ago

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

andymeneely commented 8 years ago

Title of chapter

The Naturalness of Software

URL to the chapter

https://github.com/ds4se/chapters/blob/master/etbarr/naturalness.md

Message?

What is the chapter's clear and approachable take away message?

Source code has an emergent, predictable structure to it not unlike natural language. So techniques that analyze natural language can be used to understand software and assist in its development.

Accessible?

Is the chapters written for a generalist audience (no excessive use of technical terminology) with a minimum of diagrams and references? How can it be made more accessible to generalist?

Yes, although when it gets to bits toward the end of the introduction you might lose the audience. I would cut some there while you're ahead.

Size?

Is the chapter the right length?

Yes, although it gets close to wearing out its welcome so cutting down a little bit would be nice

Should anything missing be added?

Not that I can think of

Can anything superfluous be removed (e.g. by deleting some section that does not work so well or by using less jargon, less formulae, lees diagrams, less references).?

As above, I'm thinking the intro could be tightened up.

What are the aspects of the chapter that authors SHOULD change?

Just do some general revisions - this is a nice chapter.

Gotta Mantra?

We encouraged (but did not require) the chapter title to be a mantra or something cute/catchy, i.e., some slogan reflecting best practice for data science for SE? If you have suggestion for a better title, please put them here.

Software is as Natural as Language

Best Points

What are the best points of the chapter that the authors should NOT change?

The opening is great.

mdipenta commented 8 years ago

Title of chapter

The Naturalness of Software

URL to the chapter

https://github.com/ds4se/chapters/blob/master/etbarr/naturalness.md

Message?

What is the chapter's clear and approachable take away message?

This chapter discusses to what extent code has the repetitive characteristics of spoken languages, whether and how we can exploit this to help developers

Accessible?

Is the chapters written for a generalist audience (no excessive use of technical terminology) with a minimum of diagrams and references? How can it be made more accessible to generalist?

Overall, at the level it is written, the paper is accessible enough. Clearly the reader won't be able to get full details of n-grams theory and how to apply it, but I believe it's fine

Size?

Is the chapter the right length? Yes

Should anything missing be added? The paper contains several references to papers describing applications (which is ok, though you could reduce a bit, see below). Instead it would be useful to add a reference on n-gram theory and natural language analysis (probably reference [1] from your paper, i.e. K. Sparck Jones, “Natural language processing: a historical review,”). When talking about assistive technology, another example of application could be mobile device keyboards. To some extent, it looks like similar models are being applied already, though I'm not sure how sophisticated they are, and I know there is little about software development in that example. At least such a metaphor could be possibly referred in the intro as it is a (non-software engineering) example of practical application of such a theory.

Can anything superfluous be removed (e.g. by deleting some section that does not work so well or by using less jargon, less formulae, lees diagrams, less references).? Some references are almost duplicate, just choose the extended version each time, e.g. of A.Hindle, E.Barr, M.Gabel, Z.Su, and P.Devanbu. On the Naturalness of Software you could just choose one (BTW the URL of the extended version is currently not working, it should be http://macbeth.cs.ucdavis.edu/natural.pdf and not http://macbeth.cs.ucdavis.edu/nature.pdf ) and only A. T. Nguyen, T. T. Nguyen, and T. N. Nguyen. Lexical statistical machine translation for language migration. In Proceedings, SIGSOFT FSE, pages 651–654. ACM, 2013. removing A. T. Nguyen, T. T. Nguyen, and T. N. Nguyen. Migrating code with statistical machine translation. In Companion Proceedings of ICSE, pages 544–547. ACM, 2014.

What are the aspects of the chapter that authors SHOULD change?

I liked the "Transforming Practice" a lot. Mybe the "Studying the "Natural Linguistics" of Code" paragraph could be made closer to practical applications/implications. Above all, I would avoid a title with "Studying" because here we're talking about applications.. it could be something like "Helping project newcomers to learn coding styles" or better "Monitoring code styles to better promote team working". "Analysis and Tools" is also maybe a bit generic.. it could be something like "Inferring Code Semantics"

Gotta Mantra?

We encouraged (but did not require) the chapter title to be a mantra or something cute/catchy, i.e., some slogan reflecting best practice for data science for SE? If you have suggestion for a better title, please put them here.

_I believe so, this is the title for such a kind of chapter_

Best Points

What are the best points of the chapter that the authors should NOT change?

The introduction. Despite, as I said, the approaches behind this work might not be trivial to understand, the introduction clearly explain the naturalness methaphor, highlighting similarities and differences between spoken language and source code

Minor issues:

$n-1$ -> n-1 Write extended conference names in your references, e.g. In ICSE 2012 -> In Proceedings of the 34th ACM-IEEE International Conference on Software Engineering (ICSE 2012)

lauriew commented 8 years ago

@barr I'm quite late in giving you feedback on this chapter. You have two reviews. Tim sent out an email back in December asking for revisions by January 13. Both author and editor (that would be me) are late on this chapter. I'd ask that as quickly as possible you revise your chapter and get us the new version as we are already working with the publisher on finished chapters.

In addition to the reviewer feedback, I offer up the following:

rather than a footnote from the title, I'd suggest you create a reason to cite the NIER paper. If the verbatim text is extremely close, maybe you can make an effort to make it more different as you respond to reviewer feedback.
"Of all that we humans do, it is our use of language that most sets us apart." From what? From each other? From animals? Since it's the first sentence of the chapter, it would be good for the reader to jump on board with you. The whole first paragraph didn't clarify my question for me.
"...wrought by an expert hand to stand forth as an exemplar." Exemplar of what?
"...can we use statistical methods from natural language processing to assist programmers?" Assist programmers do what?
"Typically, English comes in between 7 and 8 bits..." If I'm understanding it right, should it be "English tokens contain in between 7 and 8 bits..."?
I found the "Analysis and Tools" less accessible. Can you provide some practical and understandable examples of what can come out of this.

ds4se / chapters