That doesn’t sound right, let me take a look at that and see what is going on there!
There isn’t a built in reflect tool is there? I don’t see one
1 Like
No, there isn’t right now, but I think that would be a good thing to have!
1 Like
There should be github molecule called Mirror which might be what you are looking for
2 Likes
Ok, so i think the copying and pasting not taking into account the mouse position sounds right. Currently, the pasted atoms are just offset from the original atoms. The new connections do seem like a bug but I think i might know what’s causing it and will try to fix it this week.
2 Likes
If you have something connected to the top level output and you toggle on the Top Level Wireframe check box, you should be able to see your whole design even if you’re deep inside a nested molecule
2 Likes
Thank you very much. Mirror looks cool. Now that I have connected to the output I can see the wireframe. Cool tool.
1 Like
So as far as I can tell the abundance atoms are logged as being in a position in a grid that is scaled to the screen that you are opening them on. I would really like a feature where I could open that grid as a constant window that I can scroll within. My projects are getting pretty complicated and I understand that there is a good natural boundary based on the available space but I would like to check connections sometimes and be able to adjust if I needed to. I have been working on a small laptop. A big screen would be great but I don’t have access to one at the moment and it was useless when I opened it up on the phone and things were caught along the edges. Even if it was a limited canvas it would be good to be able to open it to a standard dimension especially when looking at a project made by someone using a different dimension screen. Still having fun. appreciative of it’s flexibility and fun tools.
1 Like
Are you using molecules (Shapes → Molecules) within projects to help divide up the space?
They work a bit like folders to contain a subset of a project to keep things tidy on a smaller screen.
Edit: Once you have placed one double click on it to enter it and then you can use the up arrow on the right or the nav bar on the left to go up levels
I have been using molecules but for somethings I need to use several calculations to place things relative to each other and it eats up space. I could code more efficently as well but I try to group things as I work and make meaningful steps so I can follow back and pull what I need. If the experience changes drastically based on the size of the monitor that is looking at it and there is no way to change that it feels like people may have problems opening each other stuff. I have been reluctant to cut the pieces up more because I am using the same calculation at different levels and don’t want to repeat it.
I tried playing with zooming in the browser but it just changes the size of the font. I don’t need a huge canvas but I want it to be readable and functional on different devices.
1 Like
The way it works now is that each atom remembers it’s position on the screen from 0-1 so if it’s at .4 on your screen it will also show up as .4 of the way across my screen regardless of screen size. Similarly the size of the atoms will take up a percentage of the screen size (although there is a setting where you can customize that).
The goal is to force folks to break things up into more manageable bite size pieces and split things across levels. I do a lot of my modeling on my phone and so it needs to be reasonably simple at each level.
If you are finding that you have elements which you want to repeat across multiple levels those might be a good candidate for becoming their own projects which get imported when you need them. If you used them across multiple levels in this project they might be useful in other projects too.
I’m not disagreeing with you, it is a good suggestion and excellent feedback! That’s why it is the way it is and maybe it could be better if the canvas could pan
For instance in this project All three layers are basically the same but I need to do the dogbones, three sets of drill holes and resize things according to the wood thickness and other variables. This is one of 8 molecules that I have already made. Each layer is pulled from a different place on the chain of nodes. Parts of the calculations I can erase at the end but most of this is useful.
Here it is on the phone, not that i would do a lot of editing there but it feels like there should be a way to have a standard view so people can communicate clearly across devices and platforms.
1 Like
Woah, yeah that is getting complicated!
Is it possible to do an architecture like this where the shared common shape is passed to each layer to be customized?
I like the idea of a natural limit but i have been happy working at this complexity. some things can’t be broken up. I am having fun with making it a good procedural thing where one could adjust the major variables and it all would still work. Right now people could adjust
Wood thickness
Beam length
Beam thickness
Hardware and bolt sizes
Leg length
waste board size
Cnc bit diameter( I want some things dogboned but not all things)
I have left myself places to adjust some thicknesses of pieces as I don’t know if they will work well until I try.
1 Like
It sounds like a really really cool test case, I’m excited to see how it turns out and I really really appreciate the feedback along the way!
1 Like
Not so much as the customization all uses the same set of dependent formulas and other parts in different places. That’s basically what I am doing. I could pass all of those piece into the smaller molecules but i want to make changes all at once.
1 Like
(sort of solved) Oh no, I just opened the project again and all of the connections are gone. Eeek. Just reloaded everything an hour later and they are all gone. Sadness. Will check tomorrow.
My copy of your spinny wheel is still ok? Don’t know what happened here. I may make a copy and rebuild it. Is there a way to locally back things up ? Or walk back changes? I tried control z. I tried reloading on different machines.
Ok I opened the github and it still has the commit that wiped the project. How do I bring it back at this screen? I copied the full code text that had gotten wiped as a file on my computer but I am flummoxed as to what happened. Can I paste it back? It went from 1500 lines to 500. Kept the atom locations and equations but erased all of the connections and numbers that I had entered. I had opened abundance on my phone to take a screenshot and then closed that and opened back here and then it was gone.
It wiped anything that said io value and the connectors section.
So I figured out how to paste the text back into the document and then opened it and saved properly and it seems to be working. Two questions though. Why did it happen in the first place? and how do I roll back a github repo?
Bar wrote:
The way it works now is that each atom remembers it’s position on the screen
from 0-1 so if it’s at .4 on your screen it will also show up as .4 of the way
across my screen regardless of screen size. Similarly the size of the atoms
will take up a percentage of the screen size (although there is a setting
where you can customize that).The goal is to force folks to break things up into more manageable bite size
pieces and split things across levels. I do a lot of my modeling on my phone
and so it needs to be reasonably simple at each level.
it’s a worthy idea, but as a hard requirement I would oppose it. especially as
people age they need larger fonts to read and fitting everything on a single
screen just doesn’t always work
I’m not disagreeing with you, it is a good suggestion and excellent feedback!
That’s why it is the way it is and maybe it could be better if the canvas
could pan
I think the hassle of panning around will encourage people to break things up.
but as a project gets more complicated, the number of inputs and outputs related
to a level can get ugly.
David Lang
1 Like
I have truly no idea. I have never seen anything like that happen before. Super weird. That is 100% a bug though, I’ll investigate, but if you can find a way to make it happen consistently that’s often like 90% of the work in fixing something like that.
The good news is that because everything is saved on GitHub nothing is ever really lost. The bad news is that we don’t really have a good way to access that history right now.
I would like to add a timeline feature that lets you easily roll back to any earlier state in the UI, but we don’t have anything like that right now.
The easiest way to do it right now might be to go to the save before the bad one and click on “Browse Files”
That will basically take you back to that point in history.
From there you can download the maslow.yaml file from before the bad save, and then upload it again in the current version.
Its a bit of a pain in the butt tho and I think we need to make sure that this doesn’t happen
