Review Request: Rougier #28

Closed
wants to merge 40 commits into
from

Conversation

@rougier
Member

rougier commented Feb 26, 2017

AUTHOR

Dear @ReScience/editors,

I request a review for the following replication:

Original article

Title: Weighted Voronoi Stippling
Author(s): Adrian Secord
Journal (or Conference): International Symposium on Non-photorealistic Animation and
Rendering
Year: 2002
DOI: 10.1145/508530.508537
PDF: http://mrl.nyu.edu/~ajsecord/npar2002/npar2002_ajsecord_preprint.pdf

Replication

Author(s): Nicolas P. Rougier
Repository: https://github.com/rougier/ReScience-submission/tree/rougier-2017
PDF: https://github.com/rougier/ReScience-submission/blob/rougier-2017/article/Rougier-2017.pdf
Keywords: Python, Stippling, Voronoi, Computer Graphics, Blue Noise
Language: Python
Domain: Signal Processing

Results

  • Article has been fully replicated
  • Article has been partially replicated
  • Article has not been replicated

Potential reviewers


EDITOR

  • Editor acknowledgment (@ThomasA) February 27, 2017
  • Reviewer 1 (@rth) February 28, 2017
  • Reviewer 2 (@almarklein) February 27, 2017
  • Review 1 decision [accept] April 19, 2017
  • Review 2 decision [accept] April 28, 2017
  • Editor decision [accept] May 9, 2017
@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier Feb 26, 2017

Member

@khinsen Can you handle this submission (from me) ? I think @ThomasA can edit it but I'll let you decide. I'll now restrict myself to my author role for this submission.

Member

rougier commented Feb 26, 2017

@khinsen Can you handle this submission (from me) ? I think @ThomasA can edit it but I'll let you decide. I'll now restrict myself to my author role for this submission.

@rougier rougier changed the title from Rougier 2017 to Review Request: Rougier 2017 Feb 26, 2017

@rougier rougier changed the title from Review Request: Rougier 2017 to Review Request: Rougier Feb 26, 2017

@khinsen

This comment has been minimized.

Show comment
Hide comment
@khinsen

khinsen Feb 27, 2017

Yes, I'll take over the EIC role for this submission!

@ThomasA can you supervise the reviewing process?

khinsen commented Feb 27, 2017

Yes, I'll take over the EIC role for this submission!

@ThomasA can you supervise the reviewing process?

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Feb 27, 2017

Yes 👍
Just a moment while I read up on the procedure as this is my first.

ThomasA commented Feb 27, 2017

Yes 👍
Just a moment while I read up on the procedure as this is my first.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Feb 27, 2017

@rth and @soolijoo can you review this replication?

ThomasA commented Feb 27, 2017

@rth and @soolijoo can you review this replication?

@soolijoo

This comment has been minimized.

Show comment
Hide comment
@soolijoo

soolijoo Feb 27, 2017

Contributor

Hi, Sorry this is outside my expertise.

Contributor

soolijoo commented Feb 27, 2017

Hi, Sorry this is outside my expertise.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Feb 27, 2017

@soolijoo OK, I will try someone else.

ThomasA commented Feb 27, 2017

@soolijoo OK, I will try someone else.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Feb 27, 2017

@almarklein can you review this replication?

ThomasA commented Feb 27, 2017

@almarklein can you review this replication?

@almarklein

This comment has been minimized.

Show comment
Hide comment
@almarklein

almarklein Feb 27, 2017

Sure! Is this ready for review at this point?

Sure! Is this ready for review at this point?

@rth

This comment has been minimized.

Show comment
Hide comment
@rth

rth Feb 28, 2017

@ThomasA I would be happy to review this.

rth commented Feb 28, 2017

@ThomasA I would be happy to review this.

@khinsen

This comment has been minimized.

Show comment
Hide comment
@khinsen

khinsen Feb 28, 2017

@almarklein Submissions are supposed to be ready for review, so please go ahead!

khinsen commented Feb 28, 2017

@almarklein Submissions are supposed to be ready for review, so please go ahead!

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Feb 28, 2017

Thanks @rth and @almarklein, you can go ahead with the review.

ThomasA commented Feb 28, 2017

Thanks @rth and @almarklein, you can go ahead with the review.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Mar 15, 2017

Is everything going OK with the review, @rth and @almarklein?

ThomasA commented Mar 15, 2017

Is everything going OK with the review, @rth and @almarklein?

@rth

This comment has been minimized.

Show comment
Hide comment
@rth

rth Mar 15, 2017

@ThomasA Sorry about the delay. Will submit a review within the next few days..

rth commented Mar 15, 2017

@ThomasA Sorry about the delay. Will submit a review within the next few days..

@almarklein

This comment has been minimized.

Show comment
Hide comment
@almarklein

almarklein Mar 15, 2017

Dito; its on my list, I even have the tab open since yesterday.

Dito; its on my list, I even have the tab open since yesterday.

article/Rougier-2017.md
+
+We applied the method proposed in the original paper with some variations due
+to the non-use of the GPU for computing the Voronoi diagram. In the original
+article, author computes the Voronoi diagram by taking advantage of the graphic

This comment has been minimized.

This comment has been minimized.

@rougier

rougier Mar 23, 2017

Member

Corrected.

@rougier

rougier Mar 23, 2017

Member

Corrected.

article/Rougier-2017.md
+card (GPU) and the fact that a set of cones seen from above actually represents
+a Voronoi diagram (see [@Hoff:1999]). We use instead
+the [QHull](http://www.qhull.org) library (through
+the [Scipy](https://scipy.github.io) Python package) for computing the Voronoi

This comment has been minimized.

@almarklein

almarklein Mar 15, 2017

Can you explain the reason for using QHull instead of the GPU approach? Same for why you "... did not use the proposed optimization" below.

@almarklein

almarklein Mar 15, 2017

Can you explain the reason for using QHull instead of the GPU approach? Same for why you "... did not use the proposed optimization" below.

This comment has been minimized.

@rougier

rougier Mar 23, 2017

Member

GPU approach would require opening a GL context and I thought it would be too much trouble to make the script work on OSX/Linux/Windows. Plus, the GPU suffers from lack of precision because it works at image level (even if authors use higher resolution to cope with such imprecision). Since the qhull library is directly accessible from within scipy, I prefer to use this.

I will add a justification in the text.

As for optimization, I implemented the optimized version that is less readable in my opinion. I'll leave the "old" code for informational purpoose.

@rougier

rougier Mar 23, 2017

Member

GPU approach would require opening a GL context and I thought it would be too much trouble to make the script work on OSX/Linux/Windows. Plus, the GPU suffers from lack of precision because it works at image level (even if authors use higher resolution to cope with such imprecision). Since the qhull library is directly accessible from within scipy, I prefer to use this.

I will add a justification in the text.

As for optimization, I implemented the optimized version that is less readable in my opinion. I'll leave the "old" code for informational purpoose.

article/Rougier-2017.md
+plants (figure @fig:large-plant & @fig:small-plant), the output of our
+replication is clearly at a lower quality without having identified the
+cause. Most probably, the limited resolution of the input image may be a
+critical factor and it is not clear if the author used these small resolution

This comment has been minimized.

@almarklein

almarklein Mar 15, 2017

How likely is it that the cause is related to the differences in the method, e.g. the omission of the optimization mentioned above? (Or was that a purely performance optimization?)

@almarklein

almarklein Mar 15, 2017

How likely is it that the cause is related to the differences in the method, e.g. the omission of the optimization mentioned above? (Or was that a purely performance optimization?)

This comment has been minimized.

@almarklein

almarklein Mar 15, 2017

At the bottom of this page it seems to be claimed that these are the original images (which I interpret as being the ones that he used): https://cs.nyu.edu/~ajsecord/stipples.html

@almarklein

almarklein Mar 15, 2017

At the bottom of this page it seems to be claimed that these are the original images (which I interpret as being the ones that he used): https://cs.nyu.edu/~ajsecord/stipples.html

This comment has been minimized.

@rougier

rougier Mar 23, 2017

Member

Optimized and non-optimized are supposed to give the exact same results and I checked this is the case.

For the image, they are accessible but you cannot use them without permission.

@rougier

rougier Mar 23, 2017

Member

Optimized and non-optimized are supposed to give the exact same results and I checked this is the case.

For the image, they are accessible but you cannot use them without permission.

article/Rougier-2017.md
+We contacted the original author asking for permission to re-use
+the
+[original images](http://cs.nyu.edu/~ajsecord/npar2002/StipplingOriginals.zip)
+but did not obtain any response. We thus display here only the output of our

This comment has been minimized.

@almarklein

almarklein Mar 15, 2017

I am slightly confused by what you mean here. I suspect that you did use these images to produce the results shown in this paper. Do you mean that you'd like to have re-published the original (color) images in the paper, or that you'd like access to the output images produced by the original algorithm?

@almarklein

almarklein Mar 15, 2017

I am slightly confused by what you mean here. I suspect that you did use these images to produce the results shown in this paper. Do you mean that you'd like to have re-published the original (color) images in the paper, or that you'd like access to the output images produced by the original algorithm?

This comment has been minimized.

@rougier

rougier Mar 23, 2017

Member

Access is ok, but I would have liked to ship the original picture in the data directory of this submission (using CC0) but for this, I need permission from the author. Same for side-by-side comparison of original and replicated results.

@rougier

rougier Mar 23, 2017

Member

Access is ok, but I would have liked to ship the original picture in the data directory of this submission (using CC0) but for this, I need permission from the author. Same for side-by-side comparison of original and replicated results.

This comment has been minimized.

@rougier

rougier Mar 23, 2017

Member

Get permission in the end. I added the original pictures.

@rougier

rougier Mar 23, 2017

Member

Get permission in the end. I added the original pictures.

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 9, 2017

Member

@ThomasA Typos have been corrected. For the merge conflict, it is not really a problem since this PR is not supposed to be merged anyway.

Member

rougier commented May 9, 2017

@ThomasA Typos have been corrected. For the merge conflict, it is not really a problem since this PR is not supposed to be merged anyway.

@ThomasA ThomasA self-assigned this May 9, 2017

@khinsen

This comment has been minimized.

Show comment
Hide comment
@khinsen

khinsen May 9, 2017

@ThomasA A partial replication is a replication of only some of the results of the original paper. My understanding of the conclusion is different: all results were replicated but for one image the agreement with the original paper is not as good as for the others. Since the reviewers are happy with this, I conclude this is a complete replication within the state of the art of the field.

So unless the author (@rougier) or the reviewers (@rth and @almarklein) disagree with my interpretation, I'd say we can classify this as a full replication.

khinsen commented May 9, 2017

@ThomasA A partial replication is a replication of only some of the results of the original paper. My understanding of the conclusion is different: all results were replicated but for one image the agreement with the original paper is not as good as for the others. Since the reviewers are happy with this, I conclude this is a complete replication within the state of the art of the field.

So unless the author (@rougier) or the reviewers (@rth and @almarklein) disagree with my interpretation, I'd say we can classify this as a full replication.

@ReScience ReScience locked and limited conversation to collaborators May 10, 2017

@ThomasA ThomasA referenced this pull request in ReScience/ReScience May 10, 2017

Closed

Publication support #52

@ReScience ReScience unlocked this conversation May 10, 2017

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

@rougier I have trouble rebuilding the article. I have installed pandoc and pandoc-crossref. make gives me the following output:

pandoc-crossref: Error in $: mempty
pandoc: Error running filter /home/tha/.cabal/bin/pandoc-crossref
Filter returned error status 1
Makefile:10: recipe for target 'rescience-template.tex' failed
make: *** [rescience-template.tex] Error 83

Any idea what is wrong?

ThomasA commented May 10, 2017

@rougier I have trouble rebuilding the article. I have installed pandoc and pandoc-crossref. make gives me the following output:

pandoc-crossref: Error in $: mempty
pandoc: Error running filter /home/tha/.cabal/bin/pandoc-crossref
Filter returned error status 1
Makefile:10: recipe for target 'rescience-template.tex' failed
make: *** [rescience-template.tex] Error 83

Any idea what is wrong?

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

@rougier it seems it was just the newly installed pandoc-crossref that does not play nicely with Ubuntu's supplied version of pandoc.

ThomasA commented May 10, 2017

@rougier it seems it was just the newly installed pandoc-crossref that does not play nicely with Ubuntu's supplied version of pandoc.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

@rougier I am stuck trying to build the PDF. I cannot use the 'fontawesome' package on my system. XeLaTeX seems to try to call some font-generating stuff that fails. I have TeXlive 2016 installed in Ubuntu (installed directly from TexLive - not Ubuntu repositories). The fontawesome package is installed, but somehow not useable.

ThomasA commented May 10, 2017

@rougier I am stuck trying to build the PDF. I cannot use the 'fontawesome' package on my system. XeLaTeX seems to try to call some font-generating stuff that fails. I have TeXlive 2016 installed in Ubuntu (installed directly from TexLive - not Ubuntu repositories). The fontawesome package is installed, but somehow not useable.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

@rougier if I comment out \usepackage{fontawesome} from the template, make just seems to stall and never get further than this:

$ make all
[1/4] Processing Rougier-2017.tex (pass 1)

ThomasA commented May 10, 2017

@rougier if I comment out \usepackage{fontawesome} from the template, make just seems to stall and never get further than this:

$ make all
[1/4] Processing Rougier-2017.tex (pass 1)

ThomasA added some commits May 10, 2017

Article template updated from submission repo
rescience-template.tex was updated from ReScience/ReScience-submission.
@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 10, 2017

Member

If you compile using directly the latex command, what is the output ?
The fontawesome package is needed only for putting the github glyph into the red box on the right.

Member

rougier commented May 10, 2017

If you compile using directly the latex command, what is the output ?
The fontawesome package is needed only for putting the github glyph into the red box on the right.

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 10, 2017

Member

For the fontawesome installation, what is the error in the font generating stuff ?

Member

rougier commented May 10, 2017

For the fontawesome installation, what is the error in the font generating stuff ?

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

@rougier By the way, I just by mistake pushed some updates to your repo. Now I do not understand how I actually had the permission to do so? I guess this is not a major problem - the updates are supposed to go in the archive repo anyway.

ThomasA commented May 10, 2017

@rougier By the way, I just by mistake pushed some updates to your repo. Now I do not understand how I actually had the permission to do so? I guess this is not a major problem - the updates are supposed to go in the archive repo anyway.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

Regarding fonawesome, making the paper says:

$ make Rougier-2017.pdf
 [1/4] Processing Rougier-2017.tex (pass 1)

kpathsea: Running mktextfm FontAwesome
/usr/local/texlive/2016/texmf-dist/web2c/mktexnam: Could not map source abbreviation F for FontAwesome.
/usr/local/texlive/2016/texmf-dist/web2c/mktexnam: Need to update /usr/local/texlive/2016/texmf-dist/fonts/map/fontname/special.map?
mktextfm: Running mf-nowin -progname=mf \mode:=ljfour; mag:=1; nonstopmode; input FontAwesome
This is METAFONT, Version 2.7182818 (TeX Live 2016) (preloaded base=mf)


kpathsea: Running mktexmf FontAwesome
! I can't find file `FontAwesome'.
<*> ...our; mag:=1; nonstopmode; input FontAwesome
                                                  
Please type another input file name
! Emergency stop.
<*> ...our; mag:=1; nonstopmode; input FontAwesome
                                                  
Transcript written on mfput.log.
grep: FontAwesome.log: Ingen sådan fil eller filkatalog
mktextfm: `mf-nowin -progname=mf \mode:=ljfour; mag:=1; nonstopmode; input FontAwesome' failed to make FontAwesome.tfm.
kpathsea: Appending font creation commands to missfont.log.

I am whining about it here as well: https://tex.stackexchange.com/questions/369068/how-can-i-use-the-fontawesome-package-with-xelatex

ThomasA commented May 10, 2017

Regarding fonawesome, making the paper says:

$ make Rougier-2017.pdf
 [1/4] Processing Rougier-2017.tex (pass 1)

kpathsea: Running mktextfm FontAwesome
/usr/local/texlive/2016/texmf-dist/web2c/mktexnam: Could not map source abbreviation F for FontAwesome.
/usr/local/texlive/2016/texmf-dist/web2c/mktexnam: Need to update /usr/local/texlive/2016/texmf-dist/fonts/map/fontname/special.map?
mktextfm: Running mf-nowin -progname=mf \mode:=ljfour; mag:=1; nonstopmode; input FontAwesome
This is METAFONT, Version 2.7182818 (TeX Live 2016) (preloaded base=mf)


kpathsea: Running mktexmf FontAwesome
! I can't find file `FontAwesome'.
<*> ...our; mag:=1; nonstopmode; input FontAwesome
                                                  
Please type another input file name
! Emergency stop.
<*> ...our; mag:=1; nonstopmode; input FontAwesome
                                                  
Transcript written on mfput.log.
grep: FontAwesome.log: Ingen sådan fil eller filkatalog
mktextfm: `mf-nowin -progname=mf \mode:=ljfour; mag:=1; nonstopmode; input FontAwesome' failed to make FontAwesome.tfm.
kpathsea: Appending font creation commands to missfont.log.

I am whining about it here as well: https://tex.stackexchange.com/questions/369068/how-can-i-use-the-fontawesome-package-with-xelatex

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 10, 2017

Member

Yes, I can see your changes in my repo. That's totally weird.
In the meantime, once you've made all the changes, I'll be able to build the PDF and push it.

Member

rougier commented May 10, 2017

Yes, I can see your changes in my repo. That's totally weird.
In the meantime, once you've made all the changes, I'll be able to build the PDF and push it.

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 10, 2017

Member

Just to be sure, you've also installed the FontAwesome.otf somewhere ?

Member

rougier commented May 10, 2017

Just to be sure, you've also installed the FontAwesome.otf somewhere ?

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

Yes, it is in the Texlive tree

ThomasA commented May 10, 2017

Yes, it is in the Texlive tree

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 10, 2017

Member

Oh, the font is bundled with the tex package ?
For the push permission, I'll ask github support how this is possible.

Member

rougier commented May 10, 2017

Oh, the font is bundled with the tex package ?
For the push permission, I'll ask github support how this is possible.

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

Yes, fontawesome is included in Texlive. I am getting some help here: https://tex.stackexchange.com/questions/369068/how-can-i-use-the-fontawesome-package-with-xelatex, but none of it seems to work...

ThomasA commented May 10, 2017

Yes, fontawesome is included in Texlive. I am getting some help here: https://tex.stackexchange.com/questions/369068/how-can-i-use-the-fontawesome-package-with-xelatex, but none of it seems to work...

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

OK, I finally made it work.
One remaining oddity is that xelatex does not seem to interpret `` '' correctly the way pdflatex does. Is that normal?

ThomasA commented May 10, 2017

OK, I finally made it work.
One remaining oddity is that xelatex does not seem to interpret `` '' correctly the way pdflatex does. Is that normal?

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

I have been looking at this: https://tex.stackexchange.com/questions/193412/what-is-happening-to-the-quotes and this: https://tex.stackexchange.com/questions/278251/ligatures-stopped-working-in-xelatex.
I can fix the quotes if I insert \defaultfontfeatures{Ligatures=TeX} after line 33 in 'rescience-template.tex', but only if I run \setmainfont{[some font]} afterwards to set a font. Without it it has no effect. If I attempt to insert the \defaultfontfeatures line before line 32 in 'rescience-template.tex', make stalls in the first processing of Rougier-2017.tex (doesn't err - just never finishes).
I suspect that fontawesome is somehow messing with this Ligatures feature...

ThomasA commented May 10, 2017

I have been looking at this: https://tex.stackexchange.com/questions/193412/what-is-happening-to-the-quotes and this: https://tex.stackexchange.com/questions/278251/ligatures-stopped-working-in-xelatex.
I can fix the quotes if I insert \defaultfontfeatures{Ligatures=TeX} after line 33 in 'rescience-template.tex', but only if I run \setmainfont{[some font]} afterwards to set a font. Without it it has no effect. If I attempt to insert the \defaultfontfeatures line before line 32 in 'rescience-template.tex', make stalls in the first processing of Rougier-2017.tex (doesn't err - just never finishes).
I suspect that fontawesome is somehow messing with this Ligatures feature...

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 10, 2017

Should we simply ignore that `` '' quotes look ugly with this compile workflow?

ThomasA commented May 10, 2017

Should we simply ignore that `` '' quotes look ugly with this compile workflow?

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 15, 2017

Member

Could you make a PR for the main rescience-template and fix it on this repo then ?
I go the explanation for your commit in my branch, it's a new feature of the PR where you can allow edits from maintainers (there is a tickbox for allowing or not).

Member

rougier commented May 15, 2017

Could you make a PR for the main rescience-template and fix it on this repo then ?
I go the explanation for your commit in my branch, it's a new feature of the PR where you can allow edits from maintainers (there is a tickbox for allowing or not).

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 18, 2017

Sorry, I got stuck last week trying to compile the paper for publishing. This week I am swamped with a course I am giving. I hope to pick this back up tomorrow or during the weekend.

ThomasA commented May 18, 2017

Sorry, I got stuck last week trying to compile the paper for publishing. This week I am swamped with a course I am giving. I hope to pick this back up tomorrow or during the weekend.

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier May 29, 2017

Member

@ThomasA Any chance to publish it this week ?

Member

rougier commented May 29, 2017

@ThomasA Any chance to publish it this week ?

@khinsen

This comment has been minimized.

Show comment
Hide comment
@khinsen

khinsen May 31, 2017

@ThomasA a gentle reminder...

khinsen commented May 31, 2017

@ThomasA a gentle reminder...

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA May 31, 2017

ThomasA commented May 31, 2017

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Jun 2, 2017

@rougier could you please rebuild the PDF on rougier/ReScience-submission on your rougier-2017 branch (including the changes it turned out I was able to push to your repo)? I have no XeTeX experience and seem unable to get the fonts set up correctly on my system to make it compile.
Then I will pull it from there and archive it.

ThomasA commented Jun 2, 2017

@rougier could you please rebuild the PDF on rougier/ReScience-submission on your rougier-2017 branch (including the changes it turned out I was able to push to your repo)? I have no XeTeX experience and seem unable to get the fonts set up correctly on my system to make it compile.
Then I will pull it from there and archive it.

@rougier

This comment has been minimized.

Show comment
Hide comment
@rougier

rougier Jun 2, 2017

Member

Done.

Member

rougier commented Jun 2, 2017

Done.

@ReScience ReScience locked and limited conversation to collaborators Jun 2, 2017

@ThomasA

This comment has been minimized.

Show comment
Hide comment
@ThomasA

ThomasA Jun 2, 2017

EDITOR

This submission has been published and will soon appear at http://rescience.github.io/read/

DOI

ThomasA commented Jun 2, 2017

EDITOR

This submission has been published and will soon appear at http://rescience.github.io/read/

DOI

@ThomasA ThomasA closed this Jun 2, 2017

@khinsen

This comment has been minimized.

Show comment
Hide comment
@khinsen

khinsen Jun 7, 2017

This looks good! Thanks to @ThomasA for handling the review, to @rth and @almarklein for reviewing, and to @rougier for doing the replication work!

khinsen commented Jun 7, 2017

This looks good! Thanks to @ThomasA for handling the review, to @rth and @almarklein for reviewing, and to @rougier for doing the replication work!

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.