WEBVTT 00:00.000 --> 00:12.280 Welcome, everyone. I'm Francial, the Community Manager for MDN at Mozilla and I'm here 00:12.280 --> 00:19.760 to run a panel which is called Browser Priorities from a developers standpoint and here 00:19.760 --> 00:25.000 we've got four lovely people who would you like to introduce yourself? 00:25.000 --> 00:28.000 Hi. It's on. 00:28.000 --> 00:30.000 Yeah, it's on. 00:30.000 --> 00:40.440 Hello, everyone. I'm Kadir. I'm a web platform product manager at Google. But I was at Mozilla 00:40.440 --> 00:45.520 before for over decade. I was actually here in 2004. I think it might be the first 00:45.520 --> 00:51.000 time that we had this dev room. So it's nice to be in this room again. 00:52.000 --> 00:57.000 I'm Silverest. I'm a director at Mozilla. I've been at Mozilla for only 12 years. 00:57.000 --> 01:00.000 And I run a lot of different things like organization. 01:00.000 --> 01:01.000 Only. 01:01.000 --> 01:03.000 I'm a newbie. 01:03.000 --> 01:07.000 I'm Tina. I work in Netapurah for last eight years. 01:07.000 --> 01:16.000 I'm working mainly with some QA staff, like role-end features, like VPN and blocker releases, etc. 01:16.000 --> 01:22.000 And also working with user reports. So I'm a person who read the box, actually. 01:22.000 --> 01:27.000 And answer it sometimes. So yeah, nice to see all of you today. 01:27.000 --> 01:32.000 And my name is Patrick. I work at Microsoft on the edge web browser. 01:32.000 --> 01:36.000 Used to be at Mozilla before for a few years. 01:36.000 --> 01:42.000 And then at Microsoft, I work on the web platform and the inter-approject and a little bit of dev 01:42.000 --> 01:46.000 work. So I'm a bit over the place. But yeah, excited to be here. 01:46.000 --> 01:54.000 Awesome. All right. So before we start, I wanted to lay the groundwork on what this panel actually is. 01:54.000 --> 01:58.000 It's not a feature discussion. It's not a standards negotiation. 01:58.000 --> 02:02.000 We're not here for the head. We don't want to start any fights. 02:02.000 --> 02:09.000 But we really want to help you understand how do browsers ship. 02:09.000 --> 02:15.000 And how can you play a more influential thing part in that? 02:15.000 --> 02:22.000 And yeah, to start with, I think let's start with the big picture and inter-op. 02:22.000 --> 02:29.000 For Kiddier, from browser engineering perspective, what has changed about inter-operability 02:29.000 --> 02:31.000 say in the last five years? 02:31.000 --> 02:37.000 Yes, really good question. About five years ago, we had our first big developer survey. 02:37.000 --> 02:44.000 And one thing that came out of that was the MDN web developer needs assessment. 02:44.000 --> 02:48.000 The MDN web DNA was 30,000 developers answering questions. 02:48.000 --> 02:52.000 The biggest issue facing developers was interoperability. 02:52.000 --> 02:57.000 So that's where we really saw for the first time at scale that's really something that we have to solve. 02:57.000 --> 03:04.000 And since then, we've started a bunch of things. One big thing is the inter-op program, 03:04.000 --> 03:09.000 which is where, for the first time, since we've started shipping the web, 03:09.000 --> 03:15.000 browser vendors are actually coordinating on what to ship in the next year. 03:15.000 --> 03:21.000 So many people haven't heard of that probably, but the major browser vendors and the ecosystem 03:21.000 --> 03:30.000 coming together, looking at proposals and then deciding what's going to be interoperable by the end of the next year. 03:30.000 --> 03:41.000 So that's new. And what we are also now doing is that we're looking at making the web platform easier to understand with baseline. 03:41.000 --> 03:46.000 So that's something that didn't exist five years ago, but the states, if you want to know, 03:46.000 --> 03:48.000 doesn't feature actually work in Toronto for aably? 03:48.000 --> 03:53.000 The answer to that question is usually, is it baseline or is it not baseline? 03:54.000 --> 04:03.000 And those are things that, yeah, we've only had for a few years, but they've changed fundamentally how interoperability works on the web platform. 04:03.000 --> 04:12.000 I would actually like Dina's opinion on this, because something working in theory, something interoperable in theory, 04:12.000 --> 04:20.000 versus actual real testing, how has that changed in the last few years? 04:20.000 --> 04:29.000 Regarding how it works inside our teams now, it for the last five years probably over the pandemic, like the years, 04:29.000 --> 04:38.000 we rely mostly on analytics team now. So before it, it was a kind of bug fixing just when bug appears. 04:38.000 --> 04:44.000 So we start to fix it. And now we try to analyze the gaps. 04:44.000 --> 04:54.000 Using different and analytical tools, usually it's look outside. I don't know, to deep diver like into details, but it helps us a lot, 04:54.000 --> 05:00.000 and it also changed our culture inside the company. So yeah, in this way. 05:00.000 --> 05:08.000 I would say that browsers broadly agree today, right? And I mean, yeah, we do. 05:08.000 --> 05:18.000 But are there any interoperability problems that remain difficult to fix? Maybe, so let's check it on to this. 05:18.000 --> 05:24.000 Yeah, I mean, sure, there are always interoperability problems. One is the platform above all the time. 05:24.000 --> 05:32.000 And if the web is to win, it needs to absolutely continue evolving, and because it does, and it does in a messy way, that's not linear. 05:32.000 --> 05:40.000 There are always differences between browsers. So although we do agree on the massive amount, like the majority of the things, 05:40.000 --> 05:49.000 at the fringes is where it breaks. And I guess another example of this is different oasis, different devices, hardware, 05:49.000 --> 05:56.000 means that even if we all agree on the same things, it's not going to work the same everywhere all the time. 05:56.000 --> 06:07.000 WebGPU recently launched any supported by augmented browsers, but it doesn't actually work on exactly all of the different combinations of oasis and browsers. 06:07.000 --> 06:16.000 And so there are some areas where it's just like that, and we have to take it into account, and it's messy, but eventually we get there along the way. 06:16.000 --> 06:28.000 Well, we're at it. How do you balance that developer expectation, whether it's enterprise, whether it's an individual user? 06:28.000 --> 06:39.000 So enterprise, excuse me, enterprise is a big thing for Microsoft, which I work for, obviously, because we have a lot of partners and there are our customers, and we need to make sure that they have a great experience. 06:39.000 --> 06:51.000 And I think it's actually, they are usually the first ones to upgrade to the latest version of the browser, because they want the latest security fixes, security is super important to them. 06:51.000 --> 07:04.000 So not everything is a problem, but we also do realize and we understand and we work with them on apps that they have that are business critical, right, and they can't really always change. 07:04.000 --> 07:16.000 And also not all markets in the world are on this evergreen kind of web thing that most of us are on, I guess in this room, where we always have the latest browser versions. 07:16.000 --> 07:28.000 So we have to be conscious of that, and we have to give IT admins that enterprises, the way to set flags, to turn off some stuff, or to remain in a sort of a legacy mode for a while. 07:28.000 --> 07:40.000 Like we do this at Microsoft Edge on like a browser basis, or a per-site basis, we give them the tools essentially to stay, to keep on having their business critical application running. 07:40.000 --> 07:51.000 So it's nuanced, there's no like one answer, but it is something that we have to care for pretty much eternally, because the web is here to stay. 07:51.000 --> 08:02.000 Because the end I were talking about the scale of the web and how many users does the web platform actually cover at the end, it's more than 5 billion maybe. 08:02.000 --> 08:09.000 And yeah, but these are the good things when we align, let's talk about a bit about when we don't align. 08:10.000 --> 08:15.000 To sell faster, right, when browsers drift apart, does that make? 08:15.000 --> 08:23.000 When browsers drift apart, is it usually because of lack of resources, issues or general technical disagreement? 08:23.000 --> 08:29.000 I think the comments are when we talk about browser, it depends, it's sometimes we are going to align. 08:29.000 --> 08:40.000 Sometimes we disagree, so some of the typical things are, which has been a lot of big subject of discussion, it seems like web USB web serial. 08:40.000 --> 08:52.000 Google has been pushing both things for a long time, and we started also to change our mind, for example, for web serial, we realize that we have better way to make it and to make it more secure. 08:52.000 --> 08:57.000 So sometimes we disagree at the beginning, and sometimes we are going to change our mind. 08:57.000 --> 09:07.000 Another big subject that we have seen for a long time is web RTC, some of our competitors, we are not implementing web RTCs exactly the way they should do. 09:07.000 --> 09:16.000 So we have been discussing with them in various forum, like the USB 3C and so on to say, can we align on those things? 09:16.000 --> 09:20.000 We are so complex and so intertwined, so sometimes it takes forever. 09:20.000 --> 09:30.000 We also, on the positive side, we collaborate a lot on the technical level, so for example, Google is now using Rust inside Chrome, which is initially a material project. 09:30.000 --> 09:40.000 We share a lot of libraries and code, we often contribute, it's very common that, well, not that common, but sometimes we find security issues on Windows and we are going to report them to our friends. 09:40.000 --> 09:46.000 We have private mailing list where we can share the kind of information, so we collaborate a lot. 09:46.000 --> 09:52.000 We are competitors, but we are also partners in many subjects, and we like to improve the security of the web. 09:52.000 --> 09:57.000 Well, say, anybody else wants to answer that? 09:57.000 --> 10:01.000 Where about this email? 10:02.000 --> 10:11.000 So, while we broadly agree on standards, where do standards sometimes give the impression that a problem is solved, right? 10:11.000 --> 10:15.000 Even though developers still don't see a different behavior in their browser, right? 10:15.000 --> 10:20.000 So I think I'll pass it on to Kadeen, Kadeen, that. 10:21.000 --> 10:40.000 I think the issue is, oh, yeah, so, you know, yeah, sometimes it looks like a browser's implemented feature and they have, but the feature still doesn't work in general probably everywhere, and that is a big issue for developers. 10:40.000 --> 10:48.000 So we have two ways to address that. One is, and we've heard about that today already, web platform tests. 10:48.000 --> 10:56.000 So we have, I think at this point, I wanted to like, over two million tests running in web platform tests, the web platform is huge. 10:56.000 --> 11:04.000 And so what we're trying to do is get features to ship with high quality test suites. 11:04.000 --> 11:11.000 And that allows us, hopefully, to catch those things before they actually hit users and developers. 11:11.000 --> 11:15.000 But sometimes, that's not possible. So they do get out. 11:15.000 --> 11:21.000 And at that point, it is always a prioritization issue of like fixing issues. 11:21.000 --> 11:26.000 Sometimes that's difficult because there's already an expected behavior that you can't easily change. 11:26.000 --> 11:32.000 But one thing that we're now trying to do and have been doing over the last few years is within the intro project. 11:32.000 --> 11:41.000 Those also are things that we try to align on as browser vendors and try to ship things truly interoperably. 11:41.000 --> 11:45.000 So we don't just use interoperable new features that we align on. 11:45.000 --> 11:51.000 But we also look at what are features that we've already shipped, but developers can't use them because they are too busy. 11:51.000 --> 11:55.000 So the first few years of the intro program were actually focused on that. 11:55.000 --> 12:03.000 So lots of CSS Grid and CSS Flexbox issues that we fixed in the first few years of that program. 12:03.000 --> 12:11.000 There is still more. But the way to actually do that is we're trying to get better at web platform tests coverage. 12:11.000 --> 12:17.000 And I think interoperable is common for everyone to come together and collaborate. 12:18.000 --> 12:28.000 What does that look like on Google's and how does Google talk with other browser vendors and collaborate when it comes to interrupt? 12:28.000 --> 12:41.000 Yeah, so that is the main, we do it all the time with browser vendors because that's the only way to really ship things to the web platform. 12:42.000 --> 12:53.000 So we try to get things to consensus because developers often times rely on that interoperability before they can start using a feature. 12:53.000 --> 13:00.000 So even if we don't initially have maybe buy in for things, we will still experiment with those things. 13:00.000 --> 13:10.000 But we'll go back and try and show whether if there is developer interest, we show that there is developer interest and try to build consensus around that. 13:10.000 --> 13:21.000 And again, the way we have been doing that in the last few years has been through the intro program which anybody can send proposals to. 13:21.000 --> 13:26.000 So I think this year we had more than 150 proposals. 13:26.000 --> 13:31.000 Some of those things, chrome and already implemented, some of those things other browsers already implemented. 13:31.000 --> 13:37.000 And so that was the way to maybe get convinced again or look at the data again. 13:37.000 --> 13:40.000 JPEC Excel is maybe a good example of this. 13:40.000 --> 13:46.000 So originally we were convinced that there was a need. 13:46.000 --> 13:51.000 And we thought that maybe there is already a good solution to that problem. 13:51.000 --> 14:00.000 And over time, by hearing from developers about that issue, that convinced us that there is something here that we have to address. 14:00.000 --> 14:04.000 And so the intro program is really the way where we coordinate these things. 14:04.000 --> 14:08.000 And JPEC Excel is actually one of the highest rated months from the community right? 14:08.000 --> 14:13.000 Yeah, it's one of the new things that we've launched called developer signals. 14:13.000 --> 14:19.000 You can find us on can ius.com or if you go to webstates.dev. 14:19.000 --> 14:29.000 There is an upload that gets you to get up repo where we collect information about what are developers using these things for. 14:29.000 --> 14:36.000 Where are the issues, what are the work around so people are using and what are the use cases. 14:36.000 --> 14:39.000 So that gives us lots more context for those things. 14:39.000 --> 14:46.000 And yeah, in the case of JPEC Excel, if you use that information to change our mind of that one. 14:46.000 --> 14:47.000 That's awesome. 14:47.000 --> 14:52.000 Let's get into the decisions that developers really see. 14:52.000 --> 15:01.000 So, Sebastian, what kind of trade-offs do blousetings make that developers really see? 15:01.000 --> 15:05.000 So, I think even Google is actually dealing with that content. 15:05.000 --> 15:09.000 There are so many things that we want to do on the web, but limited time and resources. 15:09.000 --> 15:11.000 So we already have to make trade-offs. 15:11.000 --> 15:15.000 So for example, in my team in my org, we do a lot of different things. 15:15.000 --> 15:16.000 So we do engineering work flow. 15:16.000 --> 15:20.000 So we provide the tool for the Firefox community to work on Firefox. 15:20.000 --> 15:24.000 But we do a sort of translation. So we always have to make trade-offs. 15:24.000 --> 15:33.000 Of course, at the end, the user Firefox or our Chrome, they don't care about the number of resources in the engineering teams. 15:33.000 --> 15:35.000 They want to get new features. 15:35.000 --> 15:38.000 So we always discuss among ourselves. 15:38.000 --> 15:40.000 We define what is a priority. 15:40.000 --> 15:43.000 We get feedback from everyone inside the organization. 15:43.000 --> 15:49.000 And we also look at interrupt and what our competitors are doing and what the web devs they want. 15:49.000 --> 15:51.000 So we pay a lot of attention. 15:51.000 --> 15:53.000 But you can see that I have gray hair. 15:53.000 --> 15:54.000 It's probably because of that. 15:54.000 --> 15:57.000 It's all the trade-offs that we have to make all the time. 15:57.000 --> 16:03.000 Because the web platform is where more complex and the operating system and everything is moving super fast. 16:03.000 --> 16:04.000 So it's very hard. 16:04.000 --> 16:05.000 That job. 16:05.000 --> 16:07.000 And it's impossible. 16:07.000 --> 16:12.000 But then when push comes to shove, how do you prioritize? 16:12.000 --> 16:14.000 How do you decide what to prioritize was? 16:14.000 --> 16:19.000 It's developed experience, consistency across browsers or scalability. 16:19.000 --> 16:21.000 So we're seeing at once. 16:21.000 --> 16:23.000 So you have to prioritize. 16:23.000 --> 16:27.000 It's not like you cannot say web compact is not going to be a focus easier. 16:27.000 --> 16:30.000 You always have to do the things on Firefox. 16:30.000 --> 16:32.000 We have thousands of developers here. 16:32.000 --> 16:36.000 So we need to make sure that the process is the CI works and so on. 16:36.000 --> 16:39.000 So you cannot say we're not going to do the things this year. 16:39.000 --> 16:41.000 You can de-priorize some things. 16:41.000 --> 16:44.000 But you always have to invest in all those fields. 16:44.000 --> 16:46.000 You cannot slow down. 16:46.000 --> 16:48.000 That's awesome. 16:48.000 --> 16:51.000 When once we've shipped something. 16:51.000 --> 16:54.000 What trade-offs to browsers? 16:54.000 --> 16:55.000 Brow it. 16:55.000 --> 16:56.000 Brow the things make. 16:56.000 --> 17:00.000 Between test coverage, shipping speed, catching real world compatibility issues. 17:00.000 --> 17:02.000 That's what you know. 17:02.000 --> 17:03.000 Okay. 17:03.000 --> 17:07.000 Depends on the area probably in which we will dive deeper. 17:07.000 --> 17:11.000 Because if we're talking about, let's say, prominent decks. 17:11.000 --> 17:12.000 Yes. 17:12.000 --> 17:16.000 Here we need to rely heavily on some automated tests. 17:16.000 --> 17:22.000 And because we need to speed up, we need to deliver this. 17:22.000 --> 17:25.000 Let's say security updates very fast. 17:25.000 --> 17:30.000 So here it's testing quality, of course. 17:30.000 --> 17:35.000 And we also need to keep a high level of internal testing. 17:35.000 --> 17:40.000 Like we need to get sure that Opera Core tested well. 17:40.000 --> 17:45.000 Because those things that we had and not connected to Chromium Core. 17:45.000 --> 17:50.000 It also should be delivered and tested well together. 17:50.000 --> 17:53.000 Like together with a Chromium Core. 17:53.000 --> 17:59.000 So we need to know how it will impact real browser behavior. 17:59.000 --> 18:05.000 And also, like, for example, such kind of things that may be developed. 18:05.000 --> 18:07.000 It didn't think about. 18:07.000 --> 18:10.000 But for example, we have some web.com tests. 18:10.000 --> 18:14.000 But then we can't rely on them in 100 percent. 18:14.000 --> 18:18.000 Because we have such feature as sidebar setup. 18:18.000 --> 18:21.000 And in sidebar, let's say we have some messenger. 18:21.000 --> 18:24.000 I don't know, Facebook for messenger. 18:24.000 --> 18:31.000 And we also have to test how it will work together after an update. 18:31.000 --> 18:34.000 After an after release and you have this from Chromium. 18:34.000 --> 18:36.000 How will work? 18:36.000 --> 18:38.000 Different context menu inside this panel. 18:38.000 --> 18:44.000 Because let's say less, like, we're excited. 18:44.000 --> 18:45.000 And that's it. 18:45.000 --> 18:48.000 So it's how it works. 18:48.000 --> 18:50.000 How it works. 18:50.000 --> 18:51.000 That's awesome. 18:51.000 --> 18:54.000 And how does scale come into picture here? 18:54.000 --> 19:04.000 Do you worry about scale on how many devices the regions oftentimes when you're testing those features? 19:04.000 --> 19:09.000 Like internally, we have we rely on a special lab. 19:09.000 --> 19:10.000 Like we call it. 19:10.000 --> 19:15.000 So we tested the huge amount of devices. 19:15.000 --> 19:20.000 But if we're talking about desktop, we shouldn't rely on such amount of devices. 19:20.000 --> 19:21.000 Of course. 19:21.000 --> 19:25.000 So for the desktop, it's just operational systems. 19:25.000 --> 19:29.000 It's Windows, Mac, and Linux, they should be tested. 19:29.000 --> 19:30.000 And it's all. 19:30.000 --> 19:32.000 Cool. 19:32.000 --> 19:35.000 Next one is for Patrick. 19:35.000 --> 19:47.000 I would love to know which engineering trade-offs most strongly shaped browser priorities in ways that developers tend to underestimate. 19:47.000 --> 19:55.000 I think it's easy to underestimate the fact that the browser is just the window to everything else. 19:55.000 --> 20:00.000 The browser is the most used app on any desktop for sure. 20:00.000 --> 20:02.000 And it's the portal to everything else. 20:02.000 --> 20:08.000 And so you can literally access any untrusted code on your computer at any time. 20:08.000 --> 20:14.000 And I think it's easy to underestimate the trade-off that that causes to us as a browser team. 20:14.000 --> 20:17.000 That means privacy and security are paramount. 20:17.000 --> 20:19.000 Like they're super important. 20:19.000 --> 20:28.000 And like recently I was looking at the local font API, which is an API that's proposed to access the list of fonts on your system. 20:28.000 --> 20:34.000 And we can't really do it actually because if you do it's a way for potentially bad actors to fingerprint your device, 20:34.000 --> 20:37.000 just by knowing the list of fonts that you have. 20:37.000 --> 20:43.000 So there's all of these things are trade-off that we have to constantly make. 20:43.000 --> 20:47.000 Because we run untrusted third-party code, we just can't open up everything. 20:47.000 --> 21:02.000 And that's also why sometimes we don't agree between browser vendors because some things are seen as more difficult or complex security-wise or privacy-wise by certain vendors and less by others. 21:02.000 --> 21:05.000 So yeah, that's a big trade-off that we have to make all the time. 21:05.000 --> 21:08.000 Government regulation as well comes into play. 21:08.000 --> 21:10.000 We are different per geographies. 21:10.000 --> 21:13.000 Obviously we have to keep all of that into account. 21:13.000 --> 21:15.000 We can't just do anything we want on the web. 21:15.000 --> 21:18.000 We have to take all of these into account all the time. 21:18.000 --> 21:23.000 And can you talk about a bit about backward compatibility and how much that matters? 21:23.000 --> 21:25.000 I mean it matters a lot. 21:25.000 --> 21:33.000 And if you take the first website ever created, it still runs spacejam.com still runs exactly the same as before. 21:33.000 --> 21:35.000 So it's very, very important. 21:35.000 --> 21:43.000 But I think one thing that developers and even me didn't really realize until recently is that backwards compatibility is not guaranteed. 21:43.000 --> 21:49.000 We do take stuff out of the web on a regular basis more often than we probably realize. 21:49.000 --> 21:55.000 Where SQL I think was the biggest most recent one that was just removed. 21:55.000 --> 22:01.000 We obviously need to do this gradually because otherwise we break the web and no one likes to break the web. 22:01.000 --> 22:05.000 But there is something that can happen every now and again. 22:05.000 --> 22:07.000 That's awesome. 22:07.000 --> 22:13.000 While we've talked about backward compatibility, I think we need to look ahead in the last five minutes. 22:13.000 --> 22:15.000 This is an open question for everybody. 22:15.000 --> 22:19.000 So I would love for each of you to answer this. 22:19.000 --> 22:25.000 What is one realistic improvement, browser vendors could work on together in the next year or so? 22:25.000 --> 22:33.000 In practical terms, what is good enough interoperability mean? 22:33.000 --> 22:36.000 I mean, I didn't start. 22:36.000 --> 22:38.000 I would flip it around. 22:38.000 --> 22:41.000 I would ask the audience, what do you want us to work on? 22:41.000 --> 22:46.000 What is the most important and that's why we have all of the surveys and repository to capture feedback. 22:46.000 --> 22:51.000 Because we want to stay honest and do the things that matter the most for everybody. 22:51.000 --> 22:56.000 I think something you said before Sylverse which is it's a game of consensus. 22:56.000 --> 23:02.000 We have to agree on stuff together in the same rooms and we discuss all the time. 23:02.000 --> 23:05.000 But we all come with our own perspective. 23:05.000 --> 23:12.000 We come edge, we come with like a bit of enterprise perspective and other vendors may come with other perspective and it's great. 23:12.000 --> 23:13.000 It's how we push forward. 23:13.000 --> 23:15.000 We get all of these different voices. 23:15.000 --> 23:17.000 We bring them all to the table. 23:17.000 --> 23:23.000 We have these use cases from them and together we can shape the web into what it is. 23:23.000 --> 23:24.000 Awesome. 23:24.000 --> 23:25.000 You know? 23:25.000 --> 23:31.000 Yeah, like what come to my mind it may be some kind of protocols or regarding some concerns. 23:31.000 --> 23:40.000 For example, after some operational system updates, 23:40.000 --> 23:49.000 if you choose browser, then it should be the same browser as the user should choose and something like that. 23:49.000 --> 23:57.000 So maybe more working together regarding those kind of protocols. 23:57.000 --> 24:00.000 So initially I was going to continue what we have been doing. 24:00.000 --> 24:05.000 There are the places where we can improve and we haven't talked about AI at all. 24:05.000 --> 24:08.000 And I have to talk about AI but not seriously. 24:08.000 --> 24:14.000 We see that we are all trying to do different things in our own way and we need to be smarter. 24:14.000 --> 24:16.000 It's not going to go away. 24:16.000 --> 24:22.000 All our company are investing in those spaces and we don't want to fragment the web with AI. 24:22.000 --> 24:24.000 So we'll have to coordinate more in the future. 24:24.000 --> 24:25.000 What does it mean? 24:25.000 --> 24:26.000 How it is relevant to the user? 24:26.000 --> 24:28.000 How is it safe? 24:28.000 --> 24:30.000 And a lot of unknown. 24:30.000 --> 24:34.000 It's going to be exciting to see what we can do on the do. 24:34.000 --> 24:36.000 That sounds good. 24:36.000 --> 24:37.000 Absolutely. 24:38.000 --> 24:39.000 Let's do something. 24:39.000 --> 24:40.000 Yeah, yeah. 24:43.000 --> 24:50.000 You know, one thing that we've been doing with the intro program that's been really great is we've identified solutions. 24:50.000 --> 24:55.000 And that we're already in flight and needed introoperability. 24:55.000 --> 24:57.000 And that's been working really great. 24:57.000 --> 25:00.000 But that's of course very late in the process. 25:00.000 --> 25:06.000 So one thing that I would like to see more of is as identifying spaces of opportunity like web UI. 25:06.000 --> 25:12.000 Where we see styleable form elements, for example, is like an issue. 25:12.000 --> 25:15.000 That's not just one feature. 25:15.000 --> 25:18.000 That's the whole area of improvement. 25:18.000 --> 25:21.000 So aligning on those things is way harder. 25:21.000 --> 25:24.000 Getting data around the needs in that space is harder. 25:24.000 --> 25:27.000 And I'm hoping that we can do more of that. 25:27.000 --> 25:30.000 But also, if I'm allowed a second one. 25:31.000 --> 25:36.000 I would love for more. 25:36.000 --> 25:42.000 More people, developers, companies on the web to get involved in shaping the platform. 25:42.000 --> 25:46.000 Because it's like it's the one platform that really truly doesn't belong to anyway. 25:46.000 --> 25:49.000 It's the decentralized development really. 25:49.000 --> 25:53.000 And for the most part, it's the big vendors that invest. 25:53.000 --> 25:59.000 But there are lots of others who live on the web platform. 25:59.000 --> 26:04.000 And I'd love for them to actually get more involved in expressing their needs, 26:04.000 --> 26:10.000 but also in contributing to engines like all of the engines are open source. 26:10.000 --> 26:13.000 So a huge opportunity that doesn't really exist anywhere else. 26:13.000 --> 26:16.000 So I'd love to see more of that from the community. 26:16.000 --> 26:20.000 Absolutely. That's actually a great segue into our channels. 26:20.000 --> 26:22.000 Let me get to the next slide. 26:22.000 --> 26:25.000 Yes. 26:25.000 --> 26:26.000 Right. 26:26.000 --> 26:28.000 Get heard. 26:29.000 --> 26:35.000 So we've got four ways that you can get heard. 26:35.000 --> 26:36.000 Get involved. 26:36.000 --> 26:40.000 I think the 2026 proposals will be out soon. 26:40.000 --> 26:43.000 But please, what's opinion? 26:43.000 --> 26:48.000 Oh, yeah. I mean, if you go through this one, interrupt 2026 is now going to be announced. 26:48.000 --> 26:54.000 But every year, once a year in the third quarter, there is a proposal that we are opening it up for proposals. 26:54.000 --> 26:56.000 So get ready for that. 26:56.000 --> 27:03.000 That's we really look at that state of service. That's the state of HTML state of jazz state of CSS. 27:03.000 --> 27:05.000 We look at the results of those. 27:05.000 --> 27:07.000 We go through them together. 27:07.000 --> 27:10.000 So absolutely worst taking those surveys. 27:10.000 --> 27:11.000 Can I use feedback? 27:11.000 --> 27:15.000 So this is not just on can I use, but what Patrick mentioned. 27:15.000 --> 27:17.000 Developers signals repo. 27:17.000 --> 27:19.000 But it is exposed on can I use. 27:19.000 --> 27:23.000 If you're in the hurry, just leave a up vote that you need that one. 27:23.000 --> 27:25.000 If you have more time, let us know about your use case. 27:25.000 --> 27:28.000 And you can do that throughout the year, any time. 27:28.000 --> 27:33.000 And then the last one there is that that's where the feedback actually goes. 27:33.000 --> 27:37.000 Yeah. That's the repo where you can get involved with what did everybody else say? 27:37.000 --> 27:39.000 Yeah. That's perfect. 27:39.000 --> 27:41.000 We are already up. 27:41.000 --> 27:44.000 In terms of time, I'm sure you have questions. 27:44.000 --> 27:48.000 Please find, I actually request you to stay around for the next 30 minutes. 27:48.000 --> 27:50.000 Please find them around the bedroom. 27:50.000 --> 27:53.000 Be a huge applause for our panel today. 27:53.000 --> 27:55.000 Thank you so much.