📚 : Final tweaks.

artifact-evaluation.md

@@ -86,15 +86,15 @@ We made the following manual changes to the images for inclusion in the paper: w
 
 ### Step 3
 
-The raw ingredients for Figure 13 (see above) are also generated by the web app; we assume the web app has already been built using the instructions above. The matrices shown in the lower half of `http://localhost:1234/` contains the raw ingredients of Fig. 13; the first row of matrices, headed "Needs relation (and round-trip)", were combined to produce Fig. 13a; the second row of matrices, headed "Needed-by relation (and round-trip)", were combined to produce Fig. 13b.
+The raw ingredients for Figure 13 are also generated by the web app, which we assume has already been built using the instructions above. The matrices shown in the lower half of `http://localhost:1234/` contain the raw ingredients of Fig. 13; the first row of matrices, headed "Needs relation (and round-trip)", were combined to produce Fig. 13a; the second row of matrices, headed "Needed-by relation (and round-trip)", were combined to produce Fig. 13b.
 
-The Fluid source code used for this example can be found in `fluid/lib/convolution.fld` and `fluid/example/slicing/conv-emboss.fld`. This example is also verified by the function `test_bwd` in `test/Main.purs`, which runs the test case `fluid/example/slicing/conv-emboss.fld` with the output expectation `conv-emboss.expect.fld`, found in the same folder. This can be run by following the instructions for running the test cases above.
+The Fluid source code used for this example can be found in `fluid/lib/convolution.fld` and `fluid/example/slicing/conv-emboss.fld`. This example is also verified by the function `test_bwd` in `test/Main.purs`, which runs the test case `fluid/example/slicing/conv-emboss.fld` with the output expectation `conv-emboss.expect.fld`, found in the same folder. This can be run by following the instructions above for running the test suite.
 
 We made the following manual changes to the images for inclusion in the paper: we added the "what do I need?", "what only needs me?", "what needs me?", and "what do only I need?" arrows; the vertical bars separating inputs and outputs; and the headings "input selection A", "output selection B", etc.
 
 ### Step 4
 
-The raw ingredients for Figure 16 (see above) are generated by the test function ```testBwd``` of `src\test\Main.purs`, running the test called `section-5-example`. To verify, follow the instructions above for running the tests inside a container shell. The relevant Fluid source file is `fluid/example/slicing/section-5-example.fld`, with the output expectations in `fluid/example/slicing/section-5-example-1.expect.fld` (for left-hand side of Fig. 16) and `fluid/example/slicing/section-5-example-2.expect.fld` (for right-hand side of Fig. 16). In these and other test expectations, underscores are used to indicate "selected" data values.
+The raw ingredients for Figure 16 (see above) are generated by the test function ```testBwd``` of `src/test/Main.purs`, running the test called `section-5-example`. To verify, follow the instructions above for running the test suite. The relevant Fluid source file is `fluid/example/slicing/section-5-example.fld`, with the output expectations in `fluid/example/slicing/section-5-example-1.expect.fld` (for left-hand side of Fig. 16) and `fluid/example/slicing/section-5-example-2.expect.fld` (for right-hand side of Fig. 16). In these and other test expectations, underscores are used to indicate "selected" data values.
 
 We made the following manual changes to the images for inclusion in the paper: we removed the underscores indicating selections and used appropriately shaded backgrounds instead.