I understand the kerf issue and based on that, I think that the answer to my question is no, it will not create a snug fit due to the kerf.
I’m trying to confirm whether the scenario I’ve laid out is what Stephanie is describing or not. I’m thinking that it’s not, and that I’ve incorrectly interpreted her statement.
Oh, I know you understand, but I want to include it anyway up top for folks who might be coming in later and reading up on kerf for the first time.
It’s a confusing issue. 
The good news is…there might be a way to optimize that center cut parametrically, and I want to play with that next…
best of both worlds. 
If you were really crazy, you could lay the boxes out so that nonmatching edges nested. But the assembly would be interesting unless you engraved marks to identify each matching edge.
Great job on all of your tutorials!
Giving like that is what makes the community such a rich experience.
Thank you Jules!
Chuckle! Not a problem really…I hate doing laundry. 
The second example will not give you a snug fit… only the first one.
explained here with the last example.
Excellent. I just misunderstood what you were saying previously.
Thanks.
Onshape has that feature in the form of a FeatureScript. A FeatureScript is user-written routine to add a feature that Onshape doesn’t natively support. Most users actively share the FeatureScripts that they write, although some write them to sell.
Bottom line.
Those of us without a GF or even a laser cutter can theorize all we want but a lot is just that, theory.
I have had thought I had several things figured out until they were in my hand and then I really had to figure out what works and what doesn’t.
This is why I am always saying you work to a machines strengths.
Things to consider:
You know Mike, if you know your way around in On-Shape, (which looks like it does pretty much what Fusion 360 does), maybe you could do a quick write-up for kerf in that? It can’t be that much different, and might be easier. (I almost tried On-Shape first, since it’s also free.)
It’s just a few screen caps and a little bit of writing. The worst part is trying to get Discourse to let you format it your way.
Maybe @Hirudin could do one for Solidworks - that one is so pricey, not as many folks are going to have access to it, but still it helps to be able to see if the processes are similar.
It would be nice to have some different software methods posted, so we can see if there is an easier way to do things. I’ve never been one to stick to one program religiously. Some of them are better for one thing and others are better for another.
With all the experts we’ve got around here - we could have one heck of a knowledge base.
Thanks for the vote of confidence, but my tutorial skills are sorely lacking. I’ll have a look at the OS videos though and see if there is anything relevant there.
Sharing how to use the dogbones script would be a good tutorial too. (Even in video form. Video tuts are great, i just don’t have the equipment to create them.)
I love that so many of these tutorials have been coming up lately. I am slowly building a list of stuff I’ll need to know and my bookmarked list is constantly growing
Solidworks does not have to cost your first born anymore:
ok, different suggestion/idea. If you were making multiple boxes, would it be too much work or too complicated to make one box exactly like figure 2 here, and then the second box account for the kerf (also looking like figure 2)? That way you can swap side A from box 1 with side A from box 2 and you would get a snug fit for both boxes with a minimum of waste.
I’m probably missing something obvious, but I’m sure someone will correct me.
This obviously assumes that you were going to make more than one of the box/object in the first place.
I design everything I make 2D or 3D with OpenSCAD which is a scripting language to describe shapes. The example I posted above was made with this code.
kerf = 3;
outset = kerf / 2;
module shape()
difference() {
square([200, 100]);translate([100, -1]) square([40, 60]); }
difference() {
offset(outset) shape();offset(-outset) offset(outset) shape();}
Basically offsetting for kerf is as simple as offset(keff / 2) shape() where shape is your entire 2D design. However I like all my models to represent the actually finished object, not a toolpath. When I 3D print I give the slicer an STL file and tell it the width of the filament. When I mill I give pyCAM a DXF and tell it the tool diameter. I can publish my STL files and DXF files and anybody can make them the right size because there are no machine / material dependent offsets in them.
Bummer…yeah, OpenSCAD is a different puppy…seems to me it has one advantage in that it is a heck of a lot easier to just set up one formula that applies to the shape during the build, without all of the clicking we have to go through with other programs and since kerf is a variable - you can just change the value in one spot and everything adjusts automatically. (Like F360 does.)
But I don’t see a way around your wanting to post a different version for use in CAM applications, other than creating a copy of the file with kerf = 0, and then posting that one for CAM use.
Don’t know enough about how to use OpenSCAD to be able to tell you though.
By the way - you just created a tutorial for how to adjust for kerf in OpenSCAD. 
Pretty that up a little bit with just a bit more explanation and post it as a tutorial.
You guys haven’t twigged to one fact yet - and it’s a scary one…
.
.
.
.
.
.
.
.
WE are the experts!
Right? Kinda like when I am looking for an adult and realize I’m it.
I don’t think I’m ever gonna grow up. Chuckle! 