How the Web Almanac Reveals Global Digital Inclusion Patterns

Mike Gifford from Civic Actions joins us to discuss the Web Almanac, a remarkable project analyzing 17 million websites and 83 terabytes of data to track accessibility trends across the web. He shares insights about comparative accessibility performance across countries, revealing how European websites generally outperform American ones despite the US having a 30-year head start in accessibility legislation.

• Web Almanac provides annual snapshots comparing accessibility across millions of websites
• European government websites show better accessibility than US government sites due to the Web Accessibility Directive
• Content management systems that prioritize accessibility lead to more accessible websites overall
• Educational institutions (.edu domain) show stronger accessibility performance in the US
• Alt text remains problematic even on otherwise accessible websites
• Website accessibility typically degrades over time as content refreshes and staff changes
• Automation and AI will be necessary to maintain accessibility at scale
• Data can drive competitive improvements between institutions and countries
• Individual contributors can sometimes make more impact than government institutions

Join our community discussions on Blue Sky every Tuesday night to continue the conversation about web accessibility and digital inclusion. Find the Web Almanac at https://almanac.httparchive.org/en/2024/accessibility

Chapters

• 0:08 Welcome to AccessChat with Mike Gifford
• 2:33 Mike's Current Accessibility Work
• 7:22 The Web Almanac Explained
• 13:21 Why Accessibility Must Be Built In
• 21:10 How Legislation Impacts Web Accessibility
• 33:55 Challenges with Alt Text Implementation
• 44:53 Where Do We Go From Here?
• 57:26 The Role of Decentralized Platforms

Transcript

0:00

Hello and welcome to AXSChat. We're really very happy indeed. First time I've not said delighted in a long time in terms of our intros, but we're extremely happy to welcome back to AXSChat Mike Gifford. Mike is long term friend of AXSChat, works for Civic Actions, has been engaged in accessibility and open source for a very long time, but also was one of the catalysts for Access Chat moving our social Q &A to blue sky. So, welcome back, Mike, and thank you. So, for those people that haven't watched the previous three times you've been on, can you tell us about your role, what you're doing right now? sure. So at Civic Actions, an open standards and practices lead, happening both nationally and internationally. So that's partly through Drupal and in Drupal I'm a Drupal core accessibility maintainer. Also involved in the accessibility roles and responsibilities mapping or matrix where it's still in flux at the moment as to what the name is going to be, but there's a community group that WCA3 is set up to try and make it easier to try and... have a sense of like to make WK easier for people to grok in their own particular disciplines. And I also took on the responsibility for developing the accessibility chapter of the Web Almanac this year. So that was a huge undertaking that I wasn't necessarily thinking of getting into until I realized that the guy who had done it the previous... was not going to be able to go off and to do the work. So I stepped up and took on the data roles as well as the editorial roles and did some stuff that helped push it a little bit further than it had been done before. So was very happy about that. So for those of our audience that don't know, can you tell us what the Web Almanac is? So the Web Almanac is an annual snapshot of the web. And it's an opportunity to look at over 17 million websites and 83 terabytes of data and to analyze that to say, how is the web this year different than it was last year? So they look at things like JavaScript and content management systems. They've in the last two years been talking about sustainability because course, we're in a climate crisis. And accessibility has been something that's been part of the Web Almanac from the beginning. Many people will have heard of the Internet Archive, so an archive.org. This is a part of that, it's a part of the same nonprofit and is being developed by the HTTP Archive, which is again trying to aggregate and get a summary of where the web is going and provide that history of the web. Excellent, thank you. Looks like Deborah wants to ask a question. Well, I find it fascinating and I want to thank you, Mike, because even though I've been in the field a really long time and I've done the WCAG groups, it's just really not something that I enjoy doing. I digging into it like y'all do. It's just not what works for my brain, but I am so grateful, so grateful that you are doing it because, but also, and I know that we were all figuring this out, but I like that you said accessibility was included from the very beginning of it because I still struggle with why society would not include humans in technology anyway. I just am really baffled by why we even have to have a word accessibility or any other word because why would we build anything? Technology, the built world. that isn't accessible to humans at every phase of their life. It is so illogical to my brain. And so I'm very thankful that you are doing that. And I hope the audience understands how big of a job these things are that you are doing. And I know people, for example, WCAG is so hard. Well, this is a complicated topic. And so I think it's really wonderful that we were included right from the beginning, but. I was a former programmer. did it for six years, six long years. I was pretty good at it. I just didn't like it. But I learned so very much about things you talk about when you're bringing in the JavaScript, the sustainability. And I'm fascinated by the different steps that are being taken. so first of I wanted to just give you that long. thing just because I'm sure I'm not the only one that feels this way. Some of us will be like, yeah, I know that stuff. It's so boring, but it's so critical to what we're doing. You're the infrastructure so that we don't have to go out and beg to be included. If we make it part of the, yeah, build it into the engine. And so I, that fascinates me. And I was just curious, how did you get involved so early? Thank you. Thank you. But And also, why do we have to get involved so early? Yeah, it's once again, why are you doing anything that doesn't work for all humans? Just don't get it. the web almanac, didn't start, I'm a latecomer. I was actually just trying to go off and be involved. I wanted to help a little bit this last year. And it was only when it became clear that the guy who I thought was going to be leading it wasn't going to lead it, that I was like, okay, if it's gonna happen, I'm gonna have to do it. And I think that the almanac has just been clear that's... that this is one of the important standards for the web and that there's so much legislation around it and there's such a requirement to think about how accessibility is changing over time. it's been lovely to go off and have things like the WebA Million, which does give you that sort of snapshot of how accessible the top million websites are. but the HTTP Archive had a different kind of set of data set and a different engine that they could test against. we're using the Google Lighthouse for scanning these 17 million web pages. And that's actually gives us the information that Axe Core or Deque's Axe Core is behind that. But it allows us to see things a little bit differently because it's a different engine that's designed for avoiding false positives. We mentioned, we know that there's more legislation than ever on the topic of responsibility. There was never a period with so much legislation. So it will be interesting to see how that amount of regulation is impacting. And if there's a correspondence in the web becoming more accessible. all this legislation helping us when you look to the Almanac and to the data to the years actually helping us to progress. Absolutely. one of the things I, there's two pieces of information that's useful. One, as organizations, we're making choices about the content management systems we're implementing or the platforms we're implementing, it's very clear from the data that if you're buying a content management system that is invested in accessibility, that the chances are that your product will be more accessible than not. So... We have data on that that clearly indicates that this is the case. The next, in terms of countries and legislations, it's been interesting. One of the innovations I had is that because I was doing the analytics side as well, is that I could query to pull out government data. So I could try and say, how do governments around the world compare on accessibility and have a... a single spreadsheet that has hundreds of thousands of government websites that are part of this, and then break that down by country to say, okay, how does the UK compare to France? How does Texas compare to New York? And I have that information as part of this, and I was able to provide a leaderboard for governments around the world to see how they're doing. And this is something that's... I was able to demonstrate that the web accessibility directive has an impact because European websites are generally more accessible than the American websites, American government websites. So that's lovely from that perspective. But also it'll be interesting with the European Accessibility Act to see what those changes are in this next year. Now this is again a snapshot so that we can see how accessible governments are and how accessible countries are. in Europe, so when that gets brought over, we'll be able to see that. both the Web Accessibility Directive and the European Accessibility Act, there's going to be differences in how they're implemented in each country. So we'll be able to look at providing some country legislation there. I'm also hoping to be able to next year provide an analysis of how do the states, the US states, are they able to make their websites more accessible? because there's legislation with the Department of Justice that's pushing states to meet Debbie Cague 2.1. So we should be able to see between this year, this past year and this year coming, a significant improvement in the accessibility of states. And I have all of that data for states across the US. We know that there's been a few not so good news about the investments on diversity, equity, inclusion and accessibility. Now this being on the news. So what I would like to ask from you is that with companies somehow under pressure, how can we support those organizations that want to continue to invest because we have data that proves that their investment as broad results and they should continue to invest on this. From looking at what Lainey Feingold has stated, at least in terms of accessibility, it seems to be at this stage that accessibility is not on the chopping block, that there's a commitment still to meet Section 508 compliance. Unfortunately, the data also demonstrates that the US government is not meeting Section 508. that the goal of having an accessible government that any citizen can engage with and work with, there's now ample evidence from the study that the US federal government is not meeting those requirements around accessibility. And there hasn't been a study like this before that really demonstrates that Europe has soared ahead of the US on this front. And the US has done... I mean, the U.S. has 30 years of an advantage or a head start over the rest of the world on digital accessibility. There's really no reason for the U.S. to be behind, except that the U.S. has rested on its laurels and has assumed that since they were early on on implementing Section 508 and looking at these government policy standards, that there wasn't going to be a requirement to go off and to improve and build on that over time. So this also tells us that the fact that there is more culture of litigation in the US, that culture of litigation has not brought to any results. I mean, it has. I if you look at the, I mean, there definitely are improvements, again, doing country comparisons. If you have, if you're comparing the companies in the States, or if you're looking at educational institutions, like educational institutions in the States are quite accessible, know, in general, like the .edu domain is really quite accessible. And it'd be lovely to go off and to see how well does the edu domain compare to to the educational institutions in other countries. We don't have access to lists of every other .edu, but we do have access to the European, or sorry, the American educational institutions as part of that. So that's a nice way to go off and to highlight the work that has been done. And I think lawsuits are part of that, as well as certainly the efforts of the Department of Education, which has... has put a lot of pressure on universities to make sure that they are meeting those accessibility requirements. So, I have a couple of comments based upon the observation that you made before. I don't think we can necessarily blame people for making an assumption that stuff is better. It's something that I come across all the time. You know, we can't always assume that people are ableist because they're not doing something because one of the conversations that comes up all the time is, this is really important. I just assumed that we were doing this already. So there's this assumption that someone else is already doing it. So you wonder whether because it's been around for a while as a topic, people have just assumed that it's being done when it's not, because everybody is assuming everyone else has picked it up, and there's not that level of clarity. I'm also really interested in that comparison of academia between the different countries, because we know that when the public sector web requirements came in in the UK, some of the hardest pushback was from academia. So, if you're able to compare ac.uk with .edu, that would be really interesting because it was really something where a lot of universities and a lot of academics were really reticent to do it, which is disappointing and surprising because you would think that they want people to be able to be educated. But and there were really different opinions between the two, but having that data to be able to compare would be super useful. And to be able to compare, hey, Oxford and Harvard, compete a lot, but if Harvard is a lot more accessible than Oxford, wouldn't it be an interesting thing to bring that leadership together to say, okay, well, we need to go off and at least beat those folks in Harvard so that this is a priority. We can use the Atlantic as a way to go off and to use competition as a way to go off and encourage us to have those best angels and to work towards them and support them. Cause we currently have rankings on the most sustainable universities in the world. So it will be interesting to have one on that. Right. Well, if I could just come in here as the American in the room, I do want to say that we have done some really good things too. We're all pretty shook up over here, but I do want to say that what we do, we did a lot. We did, we tried, there were so many things that we tried to do, you know, and also my beautiful little country here, litigation and legislation, you know. The reality is that's just the way we do it as you all can see. It's really a big old mess. But at the same time, think that what is really good about my country is that we started it. We did our Americans with Disability Act in 1990. We really started updating 508 in 2001. which is is actually a law that's from 1973, but still at the same point, we got along the way, I think we just got so confused about what we were fighting for. I think we just lost, you know, and as nonprofits joined and business to business groups and they started getting funded and funded by the corporation, everything got so messy and so complicated and so out of hand. But I think we should be proud, us Americans, come on, give us our, we can be proud that we started a lot of these conversations and we did have impact. It got real messy, but our litigation and legislation forced a lot of... opportunity now, you can make accessibility great again. Yeah, but I think it's important to highlight that what was done in United States have positively influenced a change in other parts of the world. Yeah, it's going to be great. It's going to be amazing. There's some very fine people working on accessibility. Because it's all about making accessibility great again. Yes. Okay. going back to some of the things you were saying earlier, Neil, it is hard sometimes to make something visible. And so I've been looking at alt text recently and trying to look at the challenges of alt text for websites. And I've looked at some websites in the European Union and also in the... in the US that should be very accessible. I found a bunch of places where alt text is not done properly or it's just missed. And I built up an alt text scanning tool that looks for images and looks for alt text and then sort of provides some examples. And also addresses things like outdated information like long desk was an appropriate form of HTML a few years ago, but it's out of date, it's no longer supported. The WC3 still has links on their, or images on their website that are using long desk, even though it's no longer a supported standard. That's not very good for a standards body to have. But I think it is one of these things where unless it's visible, people are not going to act on it. And there's some things that are also just hard the way that we've tried to approach it, like taking images and looking at it on an image by image or page by page basis, that's a... that doesn't scale well and it allows for things to be missed. by the little open source tool that I created allows us to go up and see that in a different way and also allows us to incorporate AI into that and say if there is bad alt text, how do we provide some better suggestions that authors can see what at least can get an overview of what AI might suggest would be a better description of the image. And it may not be better for the content or for the author, but the level of alt text is just so bad right now. Even in sites that care about accessibility, because it's invisible, authors often overlook it. it still is a real problem, even though it's 1.1.1, it's the basics of web accessibility that's done badly. So, I can see you itching to get into it, but I just have a quick observation is that when the sites are originally put together, quite often you start well, right? The site websites or apps are designed by people that care, and you go through quality processes and everything else, and what happens over time is that the responsibility for refreshing the content gets devolved. and frankly in organisations they go through a process of shifting left and so they want to get more junior people doing this stuff and people leave and stuff doesn't get handed over. So, in the end, you end up with only half of the processes being followed that were originally intended and I see that being quite common in terms of how you end up with this sort of degradation of quality and accessibility on websites over time. So, how do we address that without automation? Because it's going to be very difficult because it's such a big time. we're not going to do without automation. need automation and we probably need AI too. But we also need to understand how the project life cycle works. And you described that really well. And then there's the software life cycle, just understanding how software is built. I've talked here before about open source now, how so often people do not get, governments and organizations don't get involved in the right level if they want to actually influence the project. that there's a real opportunity to, there's one blind Italian student who in the year between finishing high school and starting university contributed more to Drupal 8's accessibility than all the governments in the world combined. And that's not because he was super brilliant. He was, yes. But it's just that the governments in the world weren't working on that. They never looked at the next version. They only worked on the the version of the code they're using now. They were looking at identifying problems in what is here rather than what is coming and looking ahead to that future state. Well, and that's what I would say. do have to say that Neil, especially, and Antonio have been very funny with what they're saying about making accessibility great again. I do. Yeah. But I think it does go back to, I have a couple of just questions. And one of them, I'm curious how all this work is being funded because it's so critical. And then also, OK. Regardless of whether some country made a big old mess of everything, whatever, we need to point finger. yeah. Regardless, where do we go from here and how do we stop this? Because there are so many well-intentioned people trying to make a difference. I know that we got to back out of where we accidentally got into, DEIA, whatever. I still think that so many technologists do want to do this. remember years ago training a big telecommunications brand here in the United States. And so I go in there to train all these engineers and stuff. And they, I swear, almost knew more about it than I did at the time. And I'm like, well, if y'all know this, and they say, Deborah, it doesn't work that way. If the brand does not ask for it, and if not, they do not appreciate us saying, what about, so. It's just, and I've said to Antonio and Neil before, both of them representing a very large global brand, they have a very unique position in what we're all trying to do here, right? And so I just, I wonder where do we go from here? Because yes, my country is in trouble. But... You know, but we do influence a lot of other countries. And so what do we, by golly, what in the world do we do now? And because this isn't about a movement, this is about human beings, right? So what the heck we, you know. So where do we go now? I think that the, I think what you're doing is amazing, but I do wonder as somebody that has really struggled to get work funded, luckily I'm going to move past that. But where do we go? Where do we go from here? Yes, please. I think that actually, everybody's talking about reframing, right? Of course, and so, there's been a sort of seismic shift in how people are viewing certain topics. But, you know, we can actually think about this in a transactional way. You know, because it's, know, when you're making transactions or doing deals on the web, right? You know, what you want is for those transactions to be great and to be very, very, very successful. You don't want failure. And so when you're creating something that is accessible, it creates more great transactions because people aren't failing to be able to do something on the web. And that generates revenue. money talks. those are the sort of things that we can do, you know, to be able to when we look at the projects that we're working on in 2025, to really make them accessible and to really focus on all of the benefits for your, you know, your great country, you know, the benefits of, know, when you're going first, right, you really need to have that. leadership and being able to remove those barriers so that you can have these great transactions. So I think that that's where you can really reframe accessibility to be able to put it front and centre and put it first in the mindset of business leaders right now because they really care about these kind of things. Wow. Wow. This is my life now. This is my life now. Wow. Okay. Mike? I think that this is a really neat, I think there's definitely parts to that, right? And sometimes you've got to talk to the people, you've talked, you've got to talk the language of the people who are there and communicate to them in a way that is useful. And frankly, even if you're looking at things like meritocracies, open source communities are all about meritocracies. It's the best idea wins, but you want to make sure that the right people are in the room. and that the most diverse people are in the room and that doesn't involve just the 25 year olds, right? Like 25 year old guys surprisingly do not have all the best ideas. They might have some good ideas, I'm sure they do. But we know from all kinds of evidence that they don't have all of the best ideas. So how do we try and make sure we're bringing a better picture of humanity? Because that's most is what these projects are trying to serve. And when you're looking at accessibility, a huge part of the demographic that we need to be talking about and trying to serve better are seniors. And now with the baby boomers, we know that there's a huge portion of the population that votes and has, you know, fairly wealthy pocketbooks in comparison to any other demographic that are certainly willing to buy stuff, but they care about accessibility because they want to use the digital services that they've gotten to depend on. so... Again, how do we make sure that those people are listened to and how do you follow the money? Before COVID, I was invited by a very large bank brand in the United States. And they had done a lot of disability inclusion and stuff and accessibility, but they were really freaking out because as their wealthier customers were aging into needing more things, They did not want to use what they had put together for those other people. It was not working. I don't want to hear it. Don't you dare. I will move my money. And they were freaking out over that. I think, and once again, you know, we know that my generations and above are still controlling 60 % of the wealth. You see what some of them are doing. sorry. But it is interesting because there are a lot of neutral organizations that just really do want the customers and they want to fulfill their obligations to the customers and make more money and cross-sell and yeah. So we're dragging them through all of this mess we're all walking, but I just think it's time to rethink. I think part of the problem, even though we have tried, we've been doing access chat for 11 years, but I think people are, especially in the States, I'm just gonna speak for my country, people are so confused about what we're even talking about here. I bring up accessibility to somebody and I hear, I don't know about those LGBTQ, it's like, wow, I am talking about humans, which by the way, the LGBT, they also get to be included. What are you talking about? These are humans. So yeah. amazing how ultimately what so much comes down to is respect and realizing that everyone is walking the world in their own shoes and that they're having their own burdens and trying their best. And we've other people and then that's just such a terrible thing. And then it is so much part of the divide that we're facing. As far as the as again, what we can do from here, I think there's also a lot of opportunity to find ways to collaborate together, to learn from each other. We don't need to reinvent the wheel. We can share our best practices. We don't need to be building sort of internal best practices for accessibility. Yeah, you need that, but we also need to be engaging with the public in the open, trying to share what works for you to learn. And it's been so great to be able to see the work of organizations like the UK's Government Digital Services. and the work that they've shared, which has really helped guide so many governments around the world in terms of best practices. And we need to see a lot more of that type of leadership at a high level where people are sharing their best practices, whether it's design systems or whether it's code-based or whether it's whatever. We need to have more people sharing and... And as you said, Mike, through the entire life cycle, the entire life cycle. In the beginning of the conversation today, we talked about Bluesky. So, and I would like to have your views, Mike, what role the Fediverse and decentralized platforms play in the web going forward. And how can we ensure that those platforms keep the pace and are able to be friendly and accessible? I mean, it is interesting and part of the challenges is dealing with the, is just having how, there's so much social media that is run for corporate interests and profit is the only bottom line. And I don't think that Bluesky is, out of that realm, but they've certainly opened things up and provided more space for it to be an ecosystem, which Twitter was initially aiming to be when it was first started, but then venture capital got in and it became much more about how do we insidify, this is the Cory Doctorow term, how do we suck as much money out of this as possible or get as much political gain out of this as possible. So I do think that there's... a danger with that with any social media platform that is driven primarily by the eyeballs and by advertising. But I also think that there's some really good people behind it and that there's an opportunity to get in early enough to say, we've seen how this has failed in the last decade with other social media platforms that we've trusted in. We need to be able to work together to make sure that there's proper safety mechanisms and oversight involved to make sure that the ones we're investing in going ahead are as accessible as possible, as true to the user's privacy needs and to their needs of having an ecosystem that does not polarize them as much as possible. And so that's what I would hope for. Yeah, so I mean, think that it's refreshing to see the… the lack of control of the algorithm, intentional lack of control of the algorithm on Bluesky, hopefully facilitates those things that you just mentioned.

I need to mention the URL for the Web Almanac, which is https: 30:51

//almanac.httparchive.org. forward slash EN, forward slash 2024, forward slash accessibility. We'll put that on our webpage as well. Mike, it's been a pleasure talking with you as always. We look forward to continuing the conversation on Bluesky with the AccessChat community, which we hope will continue to grow. Everybody come, we're on Tuesday nights, you know. I'm going to do that horrible thing and say, like our video, smash the like button, subscribe to our podcasts, do all of those things, right? And also we need to thank our friends from Amazon and Microtext for supporting us and keeping us on air and accessible. Thank you once again. quickly add that the discussion element on Bluesky is so important because it's one thing to go up and have talking heads like us go and and pontificate about different subjects. But A, how well did it resound to people? Also, what are the things that are missing? And this is the only community I know that has both as that pairing of an online discussion as well as a podcast as well as an online discussion, which is is unique and very wonderful. Yeah, we want that feedback. It's not just for us to say how great we are. We want that discussion and that variety, yes. just real quickly, that we appreciate the community being patient with us, because a lot of people, a lot of our guests that we're having on, they don't have Bluesky accounts. So they're scrambling to get it and sometimes struggling, because this is their first time using Bluesky is when they're going to do the access chat. So we're growing and learning and following each other and, you know. And the platform is growing as much as we like. It's not reaching to all corners of the world. In many places, people don't really know what blue sky is, and that's the reality. Yeah, I had somebody say, are you saying blue? What are you saying blue sky? They weren't even sure. was like, no, it's just blue sky. Is it blue skies? No, no, no. Is it dead blue sky? So. No. So, yes, it's going to be a journey, but yes, thank you once again. And of course, thank you for creating the Starter Pack, which people can then go and find people like yourself and us to be part of our community. So, look forward to continuing this discussion shortly. so much.