The new multi-tab editor doesn't make sense when you switch back to a previous tab because it doesn't draw the view that that tab generated. One has to hit F5 to make the view match the tab each time you switch tabs. That is tedious and can be time consuming if the preview takes significant time.

It would be much better if the view was saved in an off screen buffer and restored. If that is not

The OpenSCAD script and ordering guide specify 3.2mm MDF for the laster-cut parts. Unfortunately, it looks like Ponoko no longer carries 3.2mm. In some commentary in the ordering guide @scottbez1 suggests that it's possible that the material was always 3.0mm and that the Ponoko was just listing it incorrectly as 3.2mm.

So, I tried ordering a parts using 3.0mm MDF with the latest release but thi

at: http://rwb27.github.io/openflexure_microscope/docs/2a_basic_optics_module.html

Step 7: It is not clear which side of the lens should face inside the dark chamber.

Issue to track progress on writing a CONTRIBUTING.md :-)
A rough, initial plan has been written by ubitux already in #2 which was:

Here is a quick & dirty howto:

• create boards/.scad following other .scad files in the same directory. This .scad is directly open-able with openscad.
• if necessary, add the electronic components in electronics/.scad. The .scad files are dir

Is your feature request related to a problem? Please describe.
Sometimes custom cuts into a lid for cable access, swtich insertion and such are easier to do with a pocket knife than modifying the design in the computer. Thus a lid style which is easy to carve was suggested by a lead user.

Describe the solution you'd like
Horizontal and vertical lines that are easy to cut with a pocke

Extend tests/test_table.awk to verify that each internal thread has identical pitch as its external counterpart.

There is a bunch of CSG APIs out there and we need to catch up with it:
https://github.com/uetchy/jsxcad/blob/39d5a86072282ab06de4d150f0884dea414bba4d/src/jsxcad.js#L4-L9

The user guide says "OpenSCAD like functions support the fn parameter, which is the numer of segments to approximate a sphere (default 32, total polygons per sphere fn*fn)."

I really doubt that this is consistently being followed in the function parameters. A few test cases should be created to verify.

Here's an example from the user guide.

`cylinder({start: [0,0,0], end: [0,0,10], r1: 1,