 Jeläämmäinen ei ole laajikin, mutta tämä on ... aivan koko nousee. Ja tämä on kyse pyrä Gracele omin vierään ... pensiapäämät, joita kuin on ... alla kylät. Laitakonnon on ennakkolin kärsteys. On ennakkolin... Ennakkolin... Ennakkolin... Emäksi... Ennakkolin... Ennakkolin... Ennakkolin... Ennakkolin... Ennakkolin... Ennakkolin... And it's important to understand the history of the written word, like in the beginning we were writing with tools, with a pen or a brush. And then currently we are like using type which is used as typefaces. So we design typefaces, they are like typified forms. This is what a font is. But how the future will look like we are not sure because with verbal fonts everything can change. And to understand that perhaps a bit better we could also look at society how this developed the last years. So in the beginning thousands of years ago when people were still writing with a pen I think it was a highly personal world. And this I think changed dramatically with the invention of Gutenberg. Because then it changed to this rational society which is based on knowledge, which is based on the printed books what we know. But how the future looks like we don't know yet. Only thing we can know is that it's gonna be variable. So in that way I think it's important to look at this change what happened when fonts become dynamic. And because probably the change is much bigger than we can perceive or understand at the moment. Because as long as fonts were static then what we actually designed were contours and those contours defined the letters. But at the moment when we work with dynamic fonts then it's probably not this contour what we design anymore. But perhaps it's much more this relation of this different contours. And that's perhaps the core of the design. And the whole design space. We have to have a control of the whole design space. Not only like static fonts inside this design space. So in that way in the 20th century our thinking when designing fonts were very much based on this contours. And we believe that in the 21st century this will change into this new perception of interpolation. Because again that's I think the core what like a variable font is about. It's a font in time with dynamics. And therefore those interpolations are very very important. And yeah you could question like why we are talking about interpolation. Is it really so important? And I think it really is because right now the font editors are able to make a linear interpolation. But with Hoi we are going to prove that actually what we really need is curved interpolation. What we call as Hoi. So in that way I think the problem of the 20th century in the beginning like in the 1960s when we first made the digital typefaces was how can we make curves on the computer. And you know like today it seems to be like totally clear to us and logical to us like of course we have the BCA curves. But we also have to remember that there was a development. There was a moment in the beginning when computer engineers said like yeah we cannot make curves. Because the computer can only think in zeros and one and so those are straight lines and how can we make those curves. And in the same way like that it's totally natural for us today that we need the curves to design our contours. I think the challenge of right now is that we have to realize like how can we get our interpolations which are actually at the moment straight. How can we get them into curves. And there seems to be this kind of wicked typographic problem within like this interpolation of funds. Because like we use interpolation already like for the last 20 years also to generate our static funds. And there the workflow works like that that first you design your masters then you do the interpolation. And quite often people have to fix it afterwards. And once the funds are fixed you will ship them. That's I think the normal workflow for most people. So like if you have a 8 family you have 8 static funds and you can still fix stuff. You can like do some optical corrections for the counter forms or stuff like this. But when we go into a variable fund workflow we cannot fix those little details anymore. We just have to accept what is there. So the workflow actually for available funds will be like first you design the fund and then you ship it. So there is no chance for fixing stuff. And this brings up this kind of wicked problem. Lukas Techot. He was like writing and thinking and talking about interpolations for quite a while. And he talked about this kind of wicked problem just a couple of months ago in Berlin at Teipel Labs. And we want to show like a short fragment of this talk. Funds are not always possible. I had to make this stencil fund for a customer last year. And I found a lot of letters with incredible interpolation problems. For instance this K. We see that in both masters the lines are all nicely parallel. But in the interpolated letter it's awful, right? And it's unsolvable. And a lot of letters had this problem because of the simple geometrics. So it's really unsolvable like he says in the current fund editors. There's no way to overcome this trouble. You will always have this distorted two points in a K. But at the same moment it's very fascinating like what's really going on. Like where does it go wrong? It's fascinating to try to understand the real problem behind it. So on the one hand Lukas said like it's impossible. So there seems to be like this kind of wicked problem. But at the same way perhaps it's just a matter of looking differently at the situation. And once we can look at it and we can actually see the problem. Then it's also possible perhaps to find the answer. So in that way if you look at it, this K problem. It's like to keep those lines parallel. And to understand it we try to abstract it a little bit. This is a problem so we only have like these two lines. And those are interpolated to these other two lines. And then when we interpolate them with 50% we get those four points. And when we connect them we see that actually those four new interpolated points are not on the line anymore. Which seems to be strange. Like we expected that they should be on a line but they are not. And again it's still hard to understand what's really going on there and what's the problem. So actually you see that there's points which should travel in different speeds. But in a linear interpolation they like every point is traveling on the same speed. And that's why you get distorted. You don't have any more slinner or good shapes in between shapes. To make it a bit more clear we try to abstract it one step further. We just have three points on a line. And so we interpolate those three points. And then we get this. And then if we go halfway again 50% we get this. And then it's even more extreme, the mistake, because this white point is not on the line anymore. So where it should be is actually here on the right. That's how we want to have it. And it seems like a wicked problem. But again you look at it and you have to look like how can you solve it. And once you look at it from a different perspective, once you drop this idea of these straight lines, the solution is actually quite simple. What we have to do is that this point which actually moves on the line, also this middle point, it should not travel on a straight line but it should travel on a curve. And once it's interpolating not on a straight line but on a curve, we can solve the problem. This is like the original with the linear interpolation. We see that the white is not in balance with the red dots. But once it's traveling on the curve, everything stays on the line. Let's do this one more time. So you see this line which is straight there between three dots. So in that way once you incorporate that in your interpolation, so once you make sure that these points are not interpolated on a straight line but on a curve, actually this is then this kind of interpolation cloud, it will show you that the problem is solved. And then again once the problem is solved, it's not a problem anymore. So basically there are no problems. This is what we got with the linear interpolation. It's more obvious the problem when you want to rotate something. Sometimes the type designers want to rotate something from the bowl to the light. And we never do that because we cannot achieve that. And that's actually what's happening here. If we want to rotate this form, this will happen. It will not rotate but because of linear interpolation, we get this kind of strange result. And then sometimes we try to fix this with intermediate masters and then we have something which looks a little bit okay but also still looks like not really okay because still the points are not moving on the curve. So if you want to rotate a form, it has to move on a curve, otherwise you cannot rotate that. And that's like basically a limitation with which we somehow learn to live along. But still I think it's now a moment also to question this limitations because of this missing moment where you can fix something. You could think that it doesn't make any sense for a normal text typeface but even like with one axis wave thing like Lucas had, there was this problem and it's solvable with this tool. And another situation is if you for example want to bend something, then again with linear interpolation there is no chance to get this done but once you use like a non-linear interpolation so higher order interpolation and interpolation where the points move on a curve, everything can be achieved. So that's all. We call it Hoi. So with Hoi the problem what we had with like impossible interpolations is solved because on the left you can see like a kind of hack where you try with the intermediate forms to make something get written but still it looks very dull. You see the mistakes there. So linear interpolation goes on there. And on the right this is like a form which is interpolated with the help of this higher order interpolation. So on the right that's a variable font O which is then interpolated with this higher order interpolation. So this leads us to make more complex results. So if you want to write something, for example this, we can also do this in a total perfect way with this higher order interpolation. So things which we could not done in the past with the tools or the strategies which we used until now. So this is still, it's text. It's not like animation. It's a bi-frame thing. But it's a variable font and it's a working applied generated font what you see. Kind of our dream since 20 years which is coming true now which makes us very excited. And another example of the same thing but a little bit more complex. Two glyphs saying one C. And of course like sometimes people say how do you do that? It's not possible to define it or to explain everything in 20 minutes. So therefore what we want to do is actually we want to show you for the first time a demo of a whole typeface, a bellow which is using this technique to get it perfectly animated so that if you want to use a website or for films and so on you can use this to have it really nicely written. Let me see. So this is generated font working in a browser with two sliders. So basically what you see is that it has an endless resolution like no delay and everything is perfectly written. Do I come? It's a dream coming true and I'd be more happy than this. So basically what you see is that perhaps it's even more beautiful written than if you would write it yourself with a pen and a paper. Of course some people will be curious how to use this technique and we were thinking about that quite a while for the last couple of months and then when you think about the tools you are using to make fonts there's always this discussion I think at the moment going on between native and extension. So we were talking to some people and it would be the perfect extension for a font editor but I'm not sure about that because if you look at it like what is a font editor then a font editor is more than that like this is just the simple the most simple font editor you can imagine it just has a tool to select a point but we learned that this is not enough we always have to use also like we have to have a tool to move the points like that's the inner logic of a font editor but I think in the future when we will do more variable fonts when we will design those interpolations then this tool I think is as logic as this tool like a tool to define or design those higher order interpolations which are using curves Yeah, a way to connect the dots in a more sophisticated way inside the design space because at the end I think what a font editor is will change in the future with the variable fonts until now it was a tool to design outlines primarily but in the future it will be as much also a tool to design those interpolations and in that way I believe that this functionality should be really at the core of a font editor and not as an extension So yeah It's like a challenge for the font editor guys here in the audience Please come back to us We hope that font lab 7 will be like a Hoi font editor and Cliffs free perhaps it will support this interface to make higher order interpolations or perhaps even a Robo font will come as a Hoi editor Yeah, there's another guy who can have better closing words than us So we can just press play here Okay If I had never dropped out I would have never dropped in on that calligraphy class and personal computers might not have the wonderful typography that they do Of course it was impossible to connect the dots looking forward when I was in college but it was very very clear looking backwards 10 years later Again, you can't connect the dots looking forward you can only connect them looking backwards so you have to trust that the dots will somehow connect in your future you have to trust in something your gut, destiny, life, karma, whatever because believing that the dots will connect down the road will give you the confidence to follow your heart even when it leads you off the well-worn path and that will make all the difference Thank you