Live CEOing Ep 703: Language Design in Wolfram Language [Visualization]

Okay. Hi, everyone. We are doing a design review for some things for version 13.3 of orphan language. Today we're talking about visualization. All right. Static and dynamic highlighting. Is that our main topic here?

Yes.

Okay, so where do I start? So.

Well, maybe you want to start with dynamic highlighting. So if you go to that dynamic. Dynamic plot highlight or dynamic plot highlight.

Is a good place to start.

Okay. There are no examples in the agenda notebook, right?

Correct.

So I'm using. Okay, so.

And it's turned on by default. So, you know, if you do.

Let me tell you something. Let me tell you something. I am fed up with having situations where, for example, something like a histogram and where you have to explicitly enable dynamics in a notebook to get something like histogram or pie chart to work where it comes out gray without dynamics being enabled. So that problem has to be solved. Do you know what I'm talking about?

I know what you're talking about.

Has that problem been solved? Because you can't ship this without that problem being solved.

Yes, that problem has been solved.

How has it been solved?

The necessary functionality is whitelisted.

Okay. Why wasn't it before. That problem for years? Why wasn't that done before in many.

I mean, in many places it is or.

No. But why wasn't that done specifically for the case of, like, pie charts and histograms?

I mean, pie chart should not have that problem.

What about histogram?

Likewise.

Okay, but it did have that problem. It does have that problem in version 13.2. It does have that problem. Or at least 13.1. It has that problem.

I would think there's something else going on.

Okay, but we got to figure out what it is because we can't ship something where that's a possibility. Am I making sense? So how do I make a notebook untrusted?

Go to your. Is it in the preferences dialog?

I can't remember. Go. Well.

Preferences dialog, the security settings.

Okay.

Edit untrusted directories.

Edit untrusted directories. Okay. All right. So I'm going to take. How do I specify that? I'm going to take the directory for this directory. Open, close. I'm going to put that in my map. Whoops. No.

Notebook directory.

Okay, now I'm going to go over here and say, edit untrusted directories. Add. It even has a browse. Okay, go that. That. Do you think it will allow the quotes? It certainly should allow the quotes. Okay. Okay, now let me go and open that. I can't tell. That worked.

Click on the Orange sector. I wonder why that's.

Correct behavior. It's not dynamic now, but what shouldn't it be doing something different UX wise?

If you mouse over one of the histogram sectors. Yeah. It's changing the edge form. Right?

Yeah.

Very simple.

I wonder why the pie chart sectors aren't popping when you click on them in my.

Well, okay, let me say this. We got to fix this problem. Right. We got to fix that problem. I think we should do some kind of UX review. About the stuff. Okay. About. It's really quite important. I mean this.

No, no.

Is can't fail type things.

Agreed. And that's what we had as at the outset and we thought we had it. That was.

Oh, you say it's on by default.

Yep.

I don't really like what it looks like. I like the functionality. I don't like what it looks like.

That's fine.

Do you have anybody from design here?

Don't think we have.

Okay, so we're going to need a meeting which has people from design to talk about this.

Right?

Sure.

You know, one day we should figure out how to get rid of the need for that evaluate. Let's. Let's discuss that. Not today, but okay.

Yep.

What's that nonsense?

What are those?

I don't know what they are. They're self test errors.

Okay.

Math edit cells. You can screen capture them.

Yeah.

Okay. That's cute that those change color. I was wondering what that was going to do. It's a good idea. I think it's too loud, but it's a good idea to have them change color.

One of the things is that our tooltips look very old fashioned. On Windows they have this faded yellow look. It just looks last century. Basically.

Let's fix it.

Exactly. So we're fixing it here, but we want to fix it globally.

Yeah. Well, that's a good thing. I think that they look crummy here as well.

Well, I think they look a thousand times better than the system defines it.

Okay. No, no, no. I mean I'm showing these ones here which are traditional tooltips.

Yeah.

Don't look very good.

Exactly.

Okay, fine. So let's get a design meeting, please. And I don't know when we're going to fit it in, but we need to do it quickly to look at the graphic design of all these elements. They're both default tooltips. And your new tooltips make sense.

Yep.

Okay, So what's the status of this dynamic plot highlight? What the hell is dynamic highlight?

It's a Wrapper. Let's do this first and we'll come to it and we can change the names and all of that.

Yeah. I should be ignoring the red. Yes, yes, obviously. I don't, I don't know because I might have typed it wrong.

It might be a little bit easier to see with fewer curves.

Yeah, Yeah, it's nice. No question it's nice. Something is really wrong. Maybe it's not helped by the fact that it's a non fixed width font. I don't know. It's just look very messy, those numbers. It's nice. Which other ones should I look at here? Which, which other things are. Go ahead and look at drop line and style.

Don't know if it's combining the style.

With the drop line in that case. Well, it's not. I probably should, shouldn't it?

I think so.

Because it combined the ball. What's the name for the ball? Is the one that has the, the numbering. But then style doesn't have the numbering. Correct. Shouldn't it be called thicken or something? I mean style just seems very generic.

I mean it is. I mean it's generic in the sense that. In the sense that's something slightly different in list plot than it does in plot. Yeah. When you have points versus curves.

Okay, look, the thing people really want is also the zoomability stuff, which I don't know whether that's.

I mean that, that's not in here. But yes, that is also.

Yeah.

Okay, so let me look at this. Plus. That's cute. That's nice. So long as it scales well.

I think it's okay.

That's good. It's good performance. Well done.

Yeah. I think there's a threshold where it cuts off, but I don't remember what the number is for it.

Well, it's less than 10 to the 5 right now in a situation like this you could just give. Rather than highlighting an individual point, you could just still give something that is the coordinates of a given location.

Potentially. Yeah.

I mean like you should. Could still do something. In this case it's perfectly reasonable.

Yeah. So you know, when it cuts off we go to like a generic mechanism that doesn't care about the geometry. It's just a coordinate based thing. Nothing in 3D yet. Right now we're just doing curves and surf or curves and points. And so now let's say for example, if you grabbed your plot example that's partially on screen and you.

Yep.

You know, do the, you know, turn it into like, you know, sine and cosine or Whatever.

The sale.

Yep. Yeah. So. So that it's an actual. Yeah, exactly. Right. And so now you can use dynamic highlights. Yes. Instead of the option say and.

So what is the term? Isn't. Don't we have a term highlighted.

We have highlighted, which is the draw with a yellow background thing.

Yep.

To take off the option just. Yeah, it's. Are they still.

What's the default. I mean, it's.

No, That's not. What Looking. Right.

Yeah. I thought that was mint. Do you remember? I thought we were going to have it so that the automatic highlighting was off if you had explicit. Or was it. Are they combining?

Well, I mean, I would have thought this dynamic highlight thing would be a thicken highlighting. We have a kit test. We can't hear you.

You're very quiet men. We're breaking up.

Sorry. We have a problem of that the mouse is hitting the curve to show the style of the point. So. Because they're competing each other. So we need to figure that out. And if you would like to try. Could you try the second argument as a style like red in the dynamic highlight wrapper.

Dead.

It's running into the testing issue, I guess.

Right.

Okay, fine. All right. So it's not working.

Yeah.

Nor's that.

Well, so highlight and so. So the static highlights, you know, the wrapper highlight and the option plot highlights are. Things that generally require a position of where to highlight.

I'm sorry. So that's a completely different idea then.

Although I thought we did have a no argument.

Yes. Could you. Could you check your version of X kernel? I am running today and it's working fine.

Well, I'm running whatever I was given. No, I'm not running today, so somebody should have checked before this meeting if you needed me to be running the one today.

Sorry, I'm just used to you always having the latest look.

I only have the latest if they build correctly. It clearly validated right when I started this meeting. So it's the first time I've used this version and I'm surprised because I was using the external on Friday.

Yes.

So. Okay, let me also make another comment, which is that something I've had trouble with, like for. For histogram. Okay. I really don't want any delay when I have a giant histogram. Go the stupid highlighting thing, which I don't care about for a big histogram. You see what I'm saying? It's absolutely irrelevant for a big histogram and I just don't want to see it. I mean, I really, really don't want to see it. I don't want the thing being grindingly slow because it's running some crazy thing like that. And so how do we deal with that? Again, another, you know, another issue that. Yeah, I mean this highlighting is both ugly and I don't really see the point of it. I just don't see the point for something like this. I mean, it's just pure amusement value.

Is the tooltip useful for you?

Mildly. Okay. Now it's not doing it anymore, which is probably merciful.

Yes. I mean it turns off after.

Okay. There are cases though where, I mean, I don't know, have you tested it carefully that there are no cases where it's like grindingly slowed down by having something dynamic. Do you understand what I'm saying?

I understand what you're saying. And we've done some testing, hasn't been exhaustive yet.

Okay. So we need to make sure to do that. Right, right. And so better to switch it off than to have something grindingly slow.

Exactly.

Okay. Because that hasn't been working so far. By the way, I see on our livestream is asking a question I was going to ask as well. Does it work in the cloud?

It will, it will.

So we.

Perhaps not immediately, but that can be.

A mid release cycle update to the cloud.

Have you talked to the cloud?

Yes, yes, yes, yes.

Is anybody in this meeting from the cloud team?

No.

Okay. Does this require native graphics in the cloud?

Yes.

So it's not. I mean it's on a completely different timescale.

Yes.

Look, it's nice functionality. Okay.

I think that we need it in the cloud, you know, as well as on the desktop.

Sure.

I think that this is, you know, where people's expectation is going.

Yeah, yeah.

You get more and more interactive media and so if we just turn it all off all the time or say that we're not going to have it, then.

Yeah, no, I understand. We need to have it. You know, it's part of modernizing all kinds of things.

You know, matplotlib doesn't have anything like this. People go to other libraries like Plotly and other things to get interactivity, whatever.

The. We should have had it years ago. We haven't. We've had it in wolf mouth for 10 years. You know, an overlay of the stuff Im Alpha.

Nothing this complete.

No, no, no, it is complete. No, no, this is we.

And we wrote a lot of that that sin were from Alpha.

Right.

Oh, who's the we? The.

This team?

No, no, the stuff that Steven's talking about.

They did.

No, or maybe I missed it. But there was a bunch of things we did write like that we did.

Some stuff to help Alpha. Yes.

Okay. Just as well. Okay, so there's a variety of options here for how this works, which is great. And alternate appearances and so on.

And so this, the difference between sort of the use cases, between interactive and stat, you know, dynamic and static is, is that if you're sending it to sort of dead media like PDFs, you know, and articles, you typically want the static things. You want it to remain there.

I know.

You know, and that's part of what you kind of communicating. Whereas in the other cases you're exploring typically.

And I know, I understand that, but I mean, so, so hold on, let's walk through the names here. Okay, this functionality looks nice, except that it needs graphic design, which we need to get that. I mean, is Tim here?

Tim is watching the stream.

He can probably quickly join if you want.

Hi, Tim, join us.

Can we get Jeremy also? I'll grab him, see whether we can. But let's talk about names.

Sure.

Okay, So let's look at what other names we have in the highlight space. Okay, so highlight mesh. What does it do?

Static highlight of, you know, sub mesh or, you know, parts of a mesh.

It's the same. It's similar to what highlight graph does.

But for me, highlight do. What does this do?

Sorry, that's the one that we just introduced.

Oh, okay.

That's the. What is the highlight.

What, what is the. Oh, highlight it. We got to make. These names are not going to work. Okay, so highlighted, highlighted does that right? For a, for a. But highlighted in here doesn't do anything. Is that true?

Correct.

Makes it not render. Okay.

Unknown thing, you know, it's just as if you'd wrapped it in foo.

Yeah. Okay, so. Let's understand the other meanings of highlight, like mesh, highlight, mesh, all those kinds of things.

So highlight, graph, mesh, image are similar in that they give, you know, a big composite object and you specify what in there.

Okay, so let's take a look here. But now in a graph, if I were to say, okay, so I've got a complete graph there. Let's say I, I say edge list of this. And now I wrap. When I make a graph out of that, can I. I can wrap style.

Yes.

Red around that.

Yeah, that's part of the construction.

Yeah, yeah. Okay, nice. Okay, if I want to label, if I want to have a node. So let's say I say style.

Not sure that that works, but if you give the Node list for sure. But let's. Let's probably works.

No, it does.

So you need an explicit node list and you can write.

Okay, right. If I have a node list there. So if I have that be five, and then I have 1, 2, 3, 4, 5, and I say, you know, style, how do I make this thing bigger?

I think you can do large. Can't remember exactly what all the keys are. That definitely works.

Yeah, there should be a way to make it larger.

Maybe point size, large.

Now, see, this is the type of thing in terms of usability that we really need to grind down. It's got to produce some message or something rather than just dumbly not doing it because I just have no idea whether it was not implemented. Somebody forgot, you know, whatever else. Get what I'm saying?

Can you try larger?

Yeah, but I mean, I will, but I'm. No, it didn't work. So you don't even know what. So no. Can you please add this to the list of things that we need to grind down in terms of graphics consistency? This stuff. Fair. Is that a yes? Is that if somebody else will do it? Okay, because I really want to see that list because we're going to want to do a whole giant series of meetings trying to grind all this stuff down. Okay, So I want to come back to the names here. Okay. We simply can't have, you know. So highlight graph is a. Take the graph and highlight it. Yes. Okay, so that's a perfectly. You know, we could have.

Because there is an object. Okay, there, there is an object. Okay. Same thing with mesh, same thing with image.

Right? That's fine. Okay, so let's. Okay, so what is wrong with having highlighted be the wrapper for static highlighting?

I don't know that there's anything wrong with it.

Okay. I mean, that would be consistent with.

Those options have nothing to do with many use cases.

Yeah, I understand that, but I mean, we can probably define that separately. That's okay. I mean, it's like saying style, you know, xxx comma, red. Oops. Right. That works. And so to me, you know, what are you going to do to highlight this? Well, we've got a different form of highlighting, you know, here as in a plot, which I think is okay. I mean, we might, we might want this to inform how you do default highlighting in a plot, but that's a different story. I think that's what we should do for the, for the wrapper, for the static wrapper.

But also, you know, what are the possible settings they're going to be Quite different in a different.

I think, I think this one has to be pulled in with this one here. Okay. We're not going to have a different name. Just too confusing for people. Now the next question is, I think that the other one should be called dynamic highlighting. Arrow true or false or whatever it is, or whatever the. Whatever the option is. The thing you're calling dynamic plot highlight, I think should be called dynamic highlighting.

Okay.

I'm kind of wondering whether dynamic highlighting and highlighting should be the same option and whether there should be a way of indicating. Because I think it's nice to be able to say, I don't know, maybe not. What is highlighting? The non dynamic version. What is that? What are the settings for that? That's what you're pulling.

So you typically often have to open that ref page. One of the things you might want to do is to point out what it is to be highlighted.

Like what point on a curve actually. Yeah, right, exactly. And you want to do what would otherwise be a screen capture, basically.

Right, Sort of. But it's kind of a tough thing to pull off to both do the screen capture and hold your mouse at a particular point.

I know, I know, I know. But I think this is the wrong name for. Okay, so look, one thing is, I think it's a good defensive naming to call this one dynamic highlighting and then whatever it is. No, now this thing here is saying I want a frozen highlight of a particular kind. Correct?

Yes, correct.

Why are you showing dynamic highlight when this plot.

The examples aren't updated.

So. Okay, but, but, but what should it be? How do you specify where you want the mouse to be effectively?

So, so we have two forms. One is a list thing and one is, you know, using placed.

Nobody will ever figure it out. I got a suggestion. Okay. A rule. Well, no, what I'm thinking is this. Okay. That what we're really doing is decorating dynamic highlighting. So dynamic highlighting goes to automatic. I don't know what that is by default. A ball, probably Right. But let's say it's a. A drop line. Okay, but now you need to specify. I actually don't want the dynamic highlighting. I want. Yeah, what I want to do now I've got that dynamic highlighting specified, but I want to. Do you know, really what this is is like it's a freeze highlights, freeze highlighting or something. I guess it's got to be an option. It can't be a plot wrapper. It's got to be an option, right? Yes, but what it really is then.

Is, I mean it could Be a wrapper around the setting.

Yes, it could be. What it really wants to say is where's the mouse? In user coordinates or in scaled coordinates or whatever else. That's really all it's saying is where's the mouse? Am I not correct?

Yes, right now, but sort of for the future, I think that there are other things that we would like to add a vocabulary for like maximum data.

Yeah, okay.

Zeros.

That's nice. Yeah. But now what about multi. Multi highlights. We probably want that as well, right?

Multi static. Yes.

Yeah, Multi static highlighting. Now in this case here, there's only one mouse cursor. But so, I mean, really what this is. So you're. You're suggesting maybe. You know, something like. Something like that. I think this is the one that should be. I mean, it's a little weird to say dynamic highlighting. Well, okay, so the other possibility will be to call this option highlighting. Or plot highlighting.

Min. Does our secret mechanism work or is it broken at the moment?

Wait, what secret? I.

Talking about the default mouse position.

I don't think that's working.

Okay.

Okay, If we called it plot highlighting, is that sensible or is that then inconsistent with all the chart, this, that and the other things. And we got plot range, which. Which works for everything. Right.

We have lots of things that start with plot that work everywhere. Plot range, plot range, padding, etc.

Plot highlighting is a possibility. So then highlighted is the wrapper form. And plot highlighting, which. Where if you say arrow drop line would be a dynamic thing.

Is frozen highlighting better than placed?

Don't be stupid. No, it's no good. But, but, but placed isn't good either.

And also, you know, some things, you know, like. Well, maybe you want the dynamic max. I didn't think of that. But some things are not super dynamic. They're sort of there, you know, and. And this introduces another symbol. Anyway.

Well, okay, but look, one thing you might want is I might want dynamic drop lines, but then I want. Might. Might want a static something else.

Yes.

So it's got clearly got to be a list of. Plot highlighting has got to be a list of things potentially.

Yes, yes.

And then each one. You know what bothers me about placed is that placed is really a thing that says where to place the thing that's inside the placed. Well, maybe that's what you're arguing this is. I think it's a weird, weird thing to say that that freezes it. Well, maybe not. Maybe you're right. So then this would look like. But then we should have only that form we should not have some other form. Okay, place this xy. I mean, I assume you're adding place to place, plot label. Right. So because it's got to work for everything, you know, it can't be the case that you have to go ask you whether it works. See what I'm saying?

Right.

I think I like that, actually. I think this is. This is a workable thing. So plot highlighting, arrow, drop line is the dynamic version. And then placed XY means it's a drop line, but it is in a specific place now, I hope.

Oh, I see. So we are on the same page. You have two options.

No, one option, one option. I don't.

No dynamic highlighting, just plot highlighting.

Okay.

Okay. After you're placed. So do place, whatever, close, comma, ball. Ball.

Yeah.

Right. So that's a fixed drop line at XY and a dynamic ball. Right.

So here's my concern. Are there any forms of annotation that might themselves need to be placed above, below, whatever else?

Not sure I understand the question.

Well, if there's a label, do you want to say where the label goes?

Right. In the dynamic case, do you want to dynamically label always above? If so, you might want to use placed around that drop line thing. You get what I'm saying? Not using placed to mean freeze it.

Right.

But I think, and I mean, I don't know whether the things that you can do with highlighting you can also do with labeling.

And there's, I think there's, you know, some overlap.

Yeah, but. But it's got to be perfectly consistent. The way placed is used has to be perfectly consistent.

Yes.

Between the labeling and this case. Right. Let's very briefly look at graphic design. This is, this is a good start. Whether this thing is placed or something different. I'm not sure.

Okay. I was gonna say the other, you know, possibility that occurs to me, and this may be kind of weird, is, you know, we could use dynamic as a wrapper.

You gotta be freaking crazy. No, absolutely not. Absolutely, totally not.

Okay.

It has a definite meaning and it's not that. Okay. Design folk see that.

Okay.

It doesn't feel quite right. Maybe it needs more space. Maybe it needs. I don't know. What. What's wrong with it? Something's not right. I also don't like the. The decimal wiggling. You see what I'm saying? What do we do about the decimal wiggling? Roger? Brett, you understand what I'm saying?

You mean so, like, you know, in the second place where, you know, it moves a little or.

Yeah, no, the fact that it has different numbers of digits depending on where I am. See what I'm saying? Why is it 0.75? It should be 0.750 so it doesn't wiggle all over the place. See what I'm saying? It shouldn't be puffing in and out as I move on the curve. Am I making sense? I mean, that should just say 0.001 or 0.002 or whatever it is. It shouldn't. I don't think. I think it should be fixed precision. I'm making sense. Fix.

I mean, I think I know how to.

But wait a minute, wait a sec. First of all, we have to decide what we want. What's. What's happening right now is no good. Right? So if. And if it is going to puffer fish. Okay, whatever the. I don't know about puffer fish. Do this. But then what. What's that?

Yes, I got the joke. But keep on going.

Right. They. Then. Then it should go out from the middle.

So I. I know how to do this. When. When we have. When we can look at everything. The thing here. I don't think we can look at everything beforehand. Not realistically.

I think, Roger, all we need to look at is the plot range. I think we can do reasonable heuristics based on, you know, the fact that X is going from 0 to 10 and Y is going from minus 1 to 1.

Fine. By the way, isn't the same issue. I've seen the same issue in, in images, I believe, where. If I go here and I go to, for example, this. God, this thing looks antiquated. Yeah. I mean, that's not wiggling as badly. It is still wiggling. It really shouldn't wiggle. You should have the same heuristics here because you know the pixel values. Right. So leave the space for that extra digit. Am I making sense?

And won't that look weird?

Think so.

Like, you know, because the extra digit digits coming or.

Tim, have an opinion about this, Whether.

It should wiggle or not. I. I agree that we should avoid wiggling as much as possible.

Yeah. But the question is then, if we. If we don't wiggle, then we have a space in this case for that 97. There. There'd be a space before the 97.

I think that would look weird.

In this case. Better than wiggling, I think, in my opinion.

Well, the other possibility to prevent wiggling is to truncate the decimal digits on the right so that you always have a fixed number of digits. So if it's more, you know, on the left. It decreases the digits on the right, which might be okay too. Am I making sense? So it's a number form setting where it's a fixed number of digits.

I think it is a better behavior for dynamic, but for static. Do you still want the extra space?

No, no. And that's okay because there is no scenario in which you're flipperooing, so to speak, between dynamic and static. In other words, it's a programming action to go between those two.

Right. You know, in the static case, your puffer fish has been taxidermied into.

Yes. And it's fully whatevered.

Cooked.

Yes. Okay. What do you think about the look of that actual tooltip? And by the way, whatever that's doing this horrifying thing that, you know, that very minimalist tooltip has to be somehow consistent with this.

I think one thing with that is the. The background of that tooltip is exact same colors the line. And I wonder if it should be lighter.

Interesting. Same hue but lighter.

Because it's, you know, a larger area.

Yeah, yeah, yeah. No, I. I kind of agree with you. I mean, the same hue but lighter is what you're saying.

But then you, you know, is the type going to be visible too?

Well, then I would reverse the. I mean, I would make the type be black.

Right.

Because I don't think there's a guarantee that the line is dark enough that the type can be white anyway.

No.

Jeremy, your comments on this.

I agree. The colors shouldn't match. It's merging with the line in a weird way. It's also following the angle of the line, maybe unintentionally, which bothers me. It's just placed where it's just kind of overlapping that line as much as possible. I wish it was.

Yeah.

Like here, if it was flipped to the right, maybe there's not room. It'd be better.

The flipping is bounded by the.

So, yeah, those are more readable.

Right.

The Philippine current is only bounded by the flower range is trying to avoid going outside.

Okay.

Okay, listen, I think I should be going to another meeting here. Could you guys continue here and just talk about some of the design stuff. I think you need to do an iteration of that. I think you should also talk about whether highlight with its ugly yellow could be slightly modified. Highlighted. Sorry. Could be. Be slightly modified to be a bit more consistent between what it's. What's happening here. Well, the highlighted here, which is this. Wherever it was, the. I don't even know where I had it. I Don't understand that something went horribly wrong here. I mean that, that that dynamic highlight that would be now highlighted on of. Wait a minute. Highlighted is intended to be dynamically or statically highlighted. Highlighted should be statically highlighted. If I explicitly have highlighted in my sequence of curves, I want it to be statically highlighted.

Yes.

Right. Yeah, that's what you want anyway the question.

So the highlighted wrapper would be static and dynamic following the option. Yeah. So one argument is going to be a static thing, you know and then if you go two arguments then yeah. It becomes.

That's fine. So the default is a static thing.

Yep.

That'S fine.

And then we use placed and whatnot in the same way as in plot highlighting for a two argument form of highlighted.

Oh, I see. If you have a ball highlighting you can say that you place it in that particular place and presumably placed allows you to have multiple XY pairs here. Or does it?

You can. We can explicitly make sure that it does.

Well, so that's multi placing where you have multiple things like this. Okay. Anyway, you guys need to work on. And I have no idea what this looks like for a. I mean you haven't done geo plots or anything like that here, but I mean you need to be thinking ahead to those.

Yep.

Also I have no idea what this does on a logarithmic plot.

It works.

Okay, but. But it's got to go to exponents and things and that's going to have a lot of wiggling. Right. If it's. I mean if I do. Doesn't look like it. Oh, 96 there. That's very wiggle for us. Well, I mean that's in my opinion too many digits.

Yes.

And I want to see what happens when it goes to scientific notation. See how do I get this to work? Well.

I think it's current formatting is preventing it from going to scientific notation at all.

It looks like that's pretty crummy.

That's right.

What's that?

It is trying to prevent go to scientific.

Yeah. And by the way, no trailing decimal points please. Ever. See what I'm saying? But you got to figure out what to do with scientific notation and whether it's going to look crazy to up puff, so to speak. It may not look as crazy as the horizontal. Anybody design folk have an opinion about that?

What do you mean by up?

I mean because it's going to have an exponent. Right. So it's going to be taller.

So I mean I personally hate these axis ticks that mix exponential and non exponential and if you think about. Sorry.

Some do, some don't.

I know, I know. So, but, but if you have it in there in the call out that some are not exponential and some are exponential, it will surely wiggle. If you always.

It'll. It'll puff up. It won't necessarily. Well, it might have to. It might have to wiggle. Look, I think there's another solution for wiggling. Okay. I think, I think if you lock the right things, for some absolutely bizarre reason, I was thinking about train tracks yesterday and real and thermal expansion of train tracks and they're a case where, you know, there is severe wiggle. Right. If you compute, you know, it's like a. I forget what it was like a 50 foot. If you had a continuous piece of iron from Boston to New York, it changes in length by 50ft or something. But that's, you know, in practice, you lock the end of each section. Each, you know, whatever it is, 70 foot section of track, you lock the end, you know, you lock the pieces down with those, you know, and then it bows out rather than wiggling. Not that we have that option here, but I'm, I'm just saying I think there might be something where, for example, you maintain the braces in the same place, but you have different, different amounts of space in the middle. I don't know. You got to try it. And I have to go to this other meeting, so I am going to depart here. Okay, but we're going to need to pick this up one more time and after got some more design stuff. I don't know whether you want to try and pick that up at the beginning of the next design review. That's at 4pm Eastern. Or, or you want to make that be the. A two hour block for calculus and algebra.

Yeah, don't mix the two.

This.

That's going to be a disaster, I think.

Okay, fine. All right, so you guys, at top of the next hour, I'm going to depart here. You can keep chatting. Okay, see you later. Bye.