need results

[timm]

i'm nearly out of new words to write

still have to describe our oracle etc but that is rewrite of current stuff. so i'm about 1-2 hours away from running out.

then i'll need your quartile changes. and these are going to have to fit on about 1 page (sorry).

[rahlk]

I have the results ready to go. I'm going to have to fit them in a page. If you ping me when edits are done, I'll work on fitting the results. On Mar 9, 2016 18:15, "Tim Menzies" notifications@github.com wrote:

Assigned #15 https://github.com/ai-se/XTREE-FSE/issues/15 to @rahlk https://github.com/rahlk.

— Reply to this email directly or view it on GitHub https://github.com/ai-se/XTREE-FSE/issues/15#event-584087110.

[timm]

i'm stopping at 10:30. care to do a midnight run after that?

[timm]

stopping now. need your results in section 4.5 BEFORE noon tomorrow. improtant.

[rahlk]

Hi Dr. Menzies, I'll have all the results by noon tomorrow. On Mar 9, 2016 10:15 PM, "Tim Menzies" notifications@github.com wrote:

stopping now. need your results in section 4.5 BEFORE noon tomorrow. improtant.

— Reply to this email directly or view it on GitHub https://github.com/ai-se/XTREE-FSE/issues/15#issuecomment-194640864.

[rahlk]

Results are up.

I feel there are way too many comparisons. I'm thinking of finding the median improvements for all changes in Shatnawi and Alves and reporting only those.

So instead of:

We'll have:

What do you think?

[timm]

I concur On Mar 10, 2016 12:12, "Rahul Krishna" notifications@github.com wrote:

Results are up.

I feel there are way too many comparisons. I'm thinking finding the median improvements for all changes in Shatnawi and Alves and reporting only those.

So instead of: [image: image] https://cloud.githubusercontent.com/assets/1433964/13677681/072258d8-e6b9-11e5-9170-5aee8c50fc3d.png

We'll have: [image: image] https://cloud.githubusercontent.com/assets/1433964/13677739/43acd17a-e6b9-11e5-8192-3835dcc915fc.png

What do you think?

— Reply to this email directly or view it on GitHub https://github.com/ai-se/XTREE-FSE/issues/15#issuecomment-194961786.

[rahlk]

Awesome, I'll quickly generate and replace the charts. On Mar 10, 2016 12:59 PM, "Tim Menzies" notifications@github.com wrote:

I concur On Mar 10, 2016 12:12, "Rahul Krishna" notifications@github.com wrote:

Results are up.

I feel there are way too many comparisons. I'm thinking finding the median improvements for all changes in Shatnawi and Alves and reporting only those.

So instead of: [image: image] < https://cloud.githubusercontent.com/assets/1433964/13677681/072258d8-e6b9-11e5-9170-5aee8c50fc3d.png

We'll have: [image: image] < https://cloud.githubusercontent.com/assets/1433964/13677739/43acd17a-e6b9-11e5-8192-3835dcc915fc.png

What do you think?

— Reply to this email directly or view it on GitHub https://github.com/ai-se/XTREE-FSE/issues/15#issuecomment-194961786.

— Reply to this email directly or view it on GitHub https://github.com/ai-se/XTREE-FSE/issues/15#issuecomment-194977997.

[timm]

results!

[rahlk]

They're up

[rahlk]

Working on Tables for Threshold values. I'll have them up in a little while. I'll also put up the graphs.

[timm]

i hear that.

note we have 3 research questions in section 4.1 .

your current results address RQ1 effectiveness and there it looks like EITHER xtree of alves might be useful.

we need the number of attributes changes to address RQ2: verbosity.

finally, we need the changed ranges to address RQ3: co-change

t

[rahlk]

@timm Chart for RQ2: Verbosity. Do the colors look good? I tried to imitate the colours from the Wollongong paper.

[timm]

maybe we need charts with numbers. five such charts would blow our space

btw- u know this is due tomorrow, right?

[rahlk]

You mean like tables, sure. That shouldn't be hard. Shall we show just this one chart and refer the readers to the tables for the rest?

Deadline is tomorrow! I swear I thought it was March 14, I realized it an hour ago. I'll do an all-nighter to make sure everything is ok.

[rahlk]

Rq2 chart is up (needs rearrangement). Breaking for dinner, I'll be back for rq3.

[timm]

I'll do an all-nighter to make sure everything is ok.

u and me both