 I mean, if people can't read it, it's not that big of a loss. Yes, good. The presentation is mostly for me, so I know what I'm doing. Okay, so we'll just get right to it, instead of going through the normal pleasantries. So the first three slides are going to discuss a little bit about the default color management on our platforms on the planet. Well, not all of them, but at least three of them. In no particular order, although you could probably say in the order in which we care about them the most, from least to most, starting with Windows. So the first thing that happens on Windows is the display is always assumed to be sRGB. There's no fancy, well, how does the display actually behave? There's no attempt to understand how it might be different than sRGB by default. You can change that, but you have to go looking for it, and it's somewhat not obvious. Untagged RGB content, so that would be anything without an ICC profile associated with it. That's assumed to be sRGB. Tag content is normalized that is converted into sRGB. So if it's tagged with something other than sRGB, then it is converted into or transformed into sRGB. But only if the application supports Windows ICC. If it doesn't, then it's almost invariably assumed to be untagged, even though it is tagged. So that's a small problem. Print drivers usually are proprietary black boxes. They assume all source content is sRGB. There is yet another black box that transforms the image or the data from sRGB to the print space. So that would be the individual ink droplets. So the summary for Windows would be it's a fairly sRGB world. It's a sRGB-centric sort of a world. On OS 10 things get a little different. The main thing is that there is always an automatic creation of a display profile. So this happens through extended display ID. So the OS talks to the display and says give me some information. And some of that information includes screen resolution. So there's a listing of screen resolutions while there's also some fairly basic color information that's included in EDID. And the system has a service that will build a display profile based on that rudimentary color information. And that's an ICC profile, international color consortium profile. It's set as default and applications can discover this in a fairly easy manner. Untagged content. This is complicated and I'm not going to get into it right now. But it's either assumed to be the same as the display space or referred to as display RGB or it's assumed to be sRGB. There are some legacy behaviors where it can be Apple RGB, which is I won't talk about that either. But that's also another possibility that makes this a little bit complicated to talk about. Tagged is also normalized to sRGB or display RGB. So if you happen to have an Adobe RGB image, you double click on it, it displays, then in that case it will convert it to display RGB because it's meant for your display. If you print it, it'll probably get converted to sRGB because the assumption is in this last dash point, again, print drivers are proprietary most of the time. That's less common on Windows than it is or on OS X that is on Windows. But it's still essentially proprietary black boxes and you don't really have a whole lot of control or understanding or any real ability to find out how it might behave or make it behave differently. It's just either you accept it or you don't use it. Now, on Linux, I have to add a little qualifier because Linux world is actually quite vast in terms of color and color management. So I have to qualify this particular size systems using color D. But what happens on such systems most of the time, or at least the ones that I've been using, automatically creates an ICC profile from EDID. So it's not an sRGB world. It is a display oriented, real world oriented. And this is actually kind of brave, I think, because at least in the Apple world where they control the displays so much, they don't have to do things like blacklist displays. So most of the displays are fairly well behaved and they don't lie about EDID information. And the rest of us were kind of under the gun when it comes to manufacturers. Manufacturers sometimes, often, don't include the right color information in EDID. So there are some special things that color D can do that will do sanity checks on the color information in EDID and won't let you, well, hopefully won't let you put a loaded gun to your head if your display manufacturer hasn't done the right thing, which is to tell the truth. But if they do, then great, you get a decent display profile built for you and you didn't even have to do anything. Untagged RGB content. This is tricky because I have to say unknown at the moment. It could be device RGB or SRGB. It depends on the scenario. It depends on the components. That could depend on the distribution that's being used. So, yeah. There is a plan to normalize to destination. So what that means is to convert content to a specific destination based on device mappings. And that's one of the things that color D does is it helps with the device mappings. And by that I mean printer particular model. It could even get to the granularity of particular media types that you happen to choose. So that you aren't just saying, okay, here's an ECI image or an SRGB image or whatever. And I need to print it on this, this and this or display it and print it. And it's just going to handle those conversions so that it's not just an SRGB world. It's not just a generic, okay, we're going to get the best that we can. There's actually a really decent plan to treat different devices as being different and do as good of a job as is possible on each of those devices. That has some tricks involved. So one of the problems in printing on Linux right now is that the application and the print driver settings must be manually set. So you have to understand those. Those are a little bit non-obvious. And I'll get into that in a later session today. This is like a difficult slide. I might even just skip this one. What did I really want to get out of this? The main thing I guess I wanted to say for this slide was, yeah, let's just skip this. We'll talk about this in the two-hour session. I'm trying to kind of help condense things rather than, because I tend to do these 20-minute sessions and they take 47 minutes rather than 20 minutes. And I don't want to do that. That's inconsiderate. So this also follows with the previous slide. Since I skipped it, I should probably skip this. But there are different versions of the ICC file format, profile format, and there is ICC version 4. That's the new current one that's been around for 10 years, which is actually not that new. Version 2, the one right before that, version 3 was skipped. Version 2 has been around forever. That's the one that's largely being used. And we're already discussing a version 5 and I'm thinking, okay, while version 4 has been around for 10 years and no one's really adopting that, what are we really, we need to solve some other issues apparently if we're going to move to version 5. So this is kind of an interesting, typical user problem that I will talk about more in detail in the two-hour session later today. I like to present this one so that people have some ideas what I'll be talking about later today and tomorrow. So this typical user problem is that prints are too dark, why does this still happen? And what's going on is people go to the store, they look at a bunch of displays, they see the brightest display and what do they do? They buy the brightest display, which is exactly the wrong thing to do. If you want color accuracy or print accuracy in the home where you have photographs of your kids or someone and or something, and you put it up on the wall, right? And usually this is in the hallway, maybe it's in your bedroom. It's not very well lit, it's somewhat dark and subdued, and yet you have this really bright display. Well, that's a problem and that's not something that's perfect, I'm actually doing reasonably well. That's not something that is congruent. You can't have a really bright display for source imagery and have a dark environment for viewing the prints. This is not something ICC Color Management per se technology-wise helps us out with. So there are actually some things that aren't even really platform specific that can help us out and understand some of these sorts of problems. I'll just leave the rest of that text alone on its own. This is a new interesting problem in the state of color management today. So this is an old problem that has become an increasing problem because of the brightness of displays. But it's an old problem. New problem. And this is an interesting problem in that it's sort of a, okay, we got a little ahead of ourselves with standards. So the standards are supposed to help us and should be vetted and this one wasn't particularly well vetted before it ended up as an end user recommendation. So what happens here is we have these things called optical brightening agents. Optical brightening agents are fluorescing materials. So what they do is they take invisible light, typically invisible light. It could be violet or slightly bluish. Usually it's around UVA and it excites a material, a fluorescing material and then it re-emits a photon at a longer wavelength or a lower energy. So it takes invisible light and makes it visible. It makes paper look brighter and cleaner and all sorts of supposedly really good things. The problem is is that when you change your light source, all of a sudden papers that look like they ought to be the same aren't the same. They visually look different and the images that are on them look different. And so this is something that doesn't probably affect a lot of photographers. Most photographers are probably sensitive for good reason in avoiding optical brightening agents in their media because they tend to reduce print permanence ratings. But for people who are improving, this is a particularly sore problem because, yeah, thank you. Because if there is a mismatch in the UV between the proofing sheet and press or the thing that's the final output and then the proofing medium, things won't look the same. So it ends up not being a proof. So that's an interesting problem. And so we can talk about the causes and some of the solutions for that. There is a solution that's been published. I don't know how well it works yet or not, but there's that. So my sessions of interest related to color. Now, there are a lot of really interesting sessions that I've seen, but the two with the red X's are the self-serving ones. Those are the ones that I'm talking in or speaking in. So the other, I'll mention the other three mostly, the multi-toner one is interesting because that's not ICC color management. That's something different. And that's fascinating to me because there are a lot of things that ICC color management helps us out with, but there are actually a lot of use cases that it does not. So that's fascinating and I recommend that one. I'm going to be there. I think the open hardware color hug spectro session that's today, tonight, that I think is really important for us to understand because, you know, we've got a lot of proprietary hardware devices that are out there. And a big part of the reason why we don't even have particularly great driver support or application support is because the hardware is closed. And this doesn't just apply to specters. It also applies to some of the best displays and some of the best displays that are available only communicate with proprietary software. So that's a bit of a challenge. Magic lantern is on the other side of this. This is on the capture side with kind of an open source firmware approach for cameras. And so that's also really deeply fascinating because that's a deeply, deeply proprietary world with the camera manufacturers. So the only thing I'll add just to end on the end to end workflow is mostly going to be this. Does it work? How should it work? What are the problems that we're finding with it? Can we actually do an end to end workflow? It's not so much like a step by step or demos because I think you guys can probably figure out how to do those things. So I'm going to talk about the stuff that's harder to figure out. That's it.