Brian Kardell, Eric Meyer and Ujjwal Sharma discuss ECMA, W3C, and WHATWG — how they work, how they differ, and how they change over time. Standards, politics, budgets, and browser reality, all in one quick episode.
Brian Kardell: Okay. Hi, I am Brian Kardell. I am a developer advocate at Igalia.
Eric Meyer: I'm Eric Meyer, also a developer advocate at Igalia.
Ujjwal Sharma: And I'm Ujjwal, also a developer advocate at Igalia.
Brian Kardell: Woo-hoo. Trifecta.
Eric Meyer: Right.
Brian Kardell: So Ujjwal has been writing a couple of pieces on ECMA. Do you want to talk about what those are talking about?
Ujjwal Sharma: Sure. So ECMA, even by the name, sounds scary. I can't imagine how people who have never worked with it perceive it, but I just wanted to give people sort of an insight into how it works, why it's important, and like-
Brian Kardell: Yeah, so you've been writing those pieces, and then for various reasons, some of which we'll get into, we've been having among ourselves a lot of discussions about W3C and WHATWG and standards organizations in general, like SDOs, and comparing them and contrasting them and thinking about how they work, and how they've changed over time and how they've stayed the same. So we thought that could be an interesting show to talk about. One of the things that I thought was really interesting when we were talking about this is that we were talking about SDOs, standards development organizations. Did you read Tim Berners-Lee's Weaving the Web?
Eric Meyer: Long time ago.
Brian Kardell: Yeah. I'm going to paraphrase, but in it, W3C has this idea of REC. I don't know, can you explain that, Eric?
Eric Meyer: Yeah. I mean, all of the standards that we think of as the standard etched in stone are first of all, not really etched in stone, but second of all, are recommendations. There is no requirement or mandate or anything that comes out of the W3C. So recommendation is kind of the end of the process and you start with various stages, but eventually, you get to recommendation, which requires, well, at least in the CSS working group, generally requires two interoperable implementations at a minimum and some other stuff.
Brian Kardell: Yeah. So it's an interesting thing when you look at HTML, it says the HTML standard. It says right there, the HTML standard.
Eric Meyer: The WHATWG HTML standard?
Brian Kardell: Yeah.
Eric Meyer: Yeah.
Brian Kardell: And when they're originally setting up the W3C, they consciously did not call it standards. And in his book, he explains that they didn't think that they wanted some kind of to claim some top down authority on the web or risk being seen as some kind of regulatory body and fall into different kinds of traps. They thought that it might create this tension where the companies that were most involved would see W3C as a gatekeeper rather than a help.
Eric Meyer: A facilitator. Yeah.
Brian Kardell: A facilitator, yeah. And so he explained that the recommendation was chosen as the term because it emphasizes consensus and signals that W3C is offering guidance and not imposing rules and it was meant to reflect the web's architecture of voluntary adoption and interoperability. It still is thought of as a standard, right? Regardless of what they wanted to do. So it would be interesting then to also compare ECMA and ECMA's history because ECMA was around pre WHATWG, early standards. I think 1998, I think, is the first. Do you happen to know, Ujjwal, what is the-
Ujjwal Sharma: Yeah, I think the first published version was '98 and I think it was written a bit sooner than that.
Brian Kardell: It was weird though, the first one, because I don't know, did you have any experience? I don't mean to date you or anything or date myself, but Flash had a language that it used called ActionScript. Did you ever use this?
Ujjwal Sharma: Yeah. I mean, it was the best way to make video games at the time, right?
Brian Kardell: So ostensibly, that was also ECMA script, which is weird, because they were pretty different, and there's interesting stories about that, but I'm curious, I don't want to put you on the spot about asking historical questions for when you weren't there or anything. So when did you join and what amount of the history do you know?
Ujjwal Sharma: Oh, I joined TC39 around 2019. So around the time that Temporal and Interop were taking off, but-
Brian Kardell: Okay.
Ujjwal Sharma: Yeah.
Brian Kardell: So what I can remember from firsthand experience is there was this period where sort of the web is cool, it's new, it's something we've never seen before. It's really taking off in ways that are unbelievable, but we don't... Kind of nobody really knows where it's going and even most of the people in W3C are sort of imagining there might be something else. And what if JavaScript but typed and with classes? And they imagined this JavaScript 2 thing, I think it was mainly Adobe. Adobe was leading it because they wanted it to power Flash's future and Mozilla actually was sort of behind it because they had the Tamarin engine donated to them, which supported this. So Mozilla was like, 'Cool, let's do it.' But Microsoft and... Well, it was basically Microsoft was the other player in the game that had any real weight, but also Yahoo and some big sites that were involved were opposed to that. And that period dragged on for a long time. So there is this long period in history of the web where both in JavaScript and in HTML and CS, the world was just kind of stalled, right? Part of it was political, part of it was pragmatic, but yeah, anyway, they ended up eventually brokering this much smaller thing that didn't include any of that stuff and they called it harmony because they got everybody to come back to the table because there were doubts that maybe even anything would happen. And so there was this harmony era from maybe 2008 to 2014, I would say. And then in 2015, this is ES6, right? Is it ES6 is what we called it?
Ujjwal Sharma: Yeah.
Brian Kardell: I don't know that anybody uses the ES terms anymore, right? Do they?
Ujjwal Sharma: I mean, it's getting more and more phased out these days, but people still use ESNext as a stand in for the newest JavaScript.
Brian Kardell: But ES6 was the biggest set of changes to JavaScript. It also, I think, formalized some of the structure of things. Then I think after 2015, things began to change to not take big things generally, right? Not the whole language as a big thing at least. Is it right?
Ujjwal Sharma: Yeah. The process evolves to sort of avoid any such big changes because it's impossible, I guess, to make up your mind about this, which is why harmony was so difficult. But yeah, I think apart from Temporal, which is unique in a way that people were motivated uniquely for Temporal, I think any changes of that size are just infeasible and politically impossible at this point.
Brian Kardell: But I think it's more than just whether the feature is big. What I'm emphasizing is that there isn't a like, 'We're going to draw a box around these and this is exactly what we're going to have in it.' And like, 'Well, if we can't get it done until 2029, oh, well.' That's just because... The whole thing sinks or swims that way, right? But I think part of that is a realization that that's not really how it works, right? Individual features sink or swim. I think that's an interesting change to just recognize the reality of that. And I think even the stages model is interesting because you also gave these abstract terms for progression, but they're just simple numerics and they don't... Maybe they're nicely ambiguous is what I'm trying to say. Eric, maybe you get what I'm trying to say. CSS is a thing we could compare against, right? So CSS follows W3C REC track process and so you get, first, you get an editor's draft, a first public working draft, right? There are several aspects of stages. And at what point is it real? Do you know what I mean? I don't know that-
Eric Meyer: Well, my experience, and this is not going to be universal, different things will have different experiences, but my experience in the CSS Working Group is that you have the editor's draft, which is sort of the, it's the first cut. It's the, 'Hey, we have an idea. Let's scribble some stuff down to more or less degree of detail, depending.' And then you go through a number of working drafts. There's the first public working draft, but then there's sort of an iterative process. It's like, 'Okay, well, we've been looking at this and the working group needs to decide how this thing interacts with another CSS spec.' Once it starts to get close to candidate recommendation, I think is when specs tend to start to become real because you do eventually have something go up for candidate recommendation. And that's really the point where interoperability is invited. Well, implementation is invited and then interoperability is invited. There may already be implementations by the time you hit that point, in which case, it tends to go faster, unless of course in the candidate recommendation phase, it's discovered, 'Oh, man, there's a major flaw in this design and we have to go back to the drawing board and overhaul whole parts of this.' Sometimes candidate recommendations go back to the working draft phase. It's not a one-way process necessarily. So that's, I think, where CSS specs at least start to get real is when you're either at or getting close to the candidate recommendation stage because within the working group, it'll be said part of the discussion will be, 'I feel like our next working draft is the last one before candidate REC unless something comes up.' And so that-
Brian Kardell: The thing that I like about talking to people is that sometimes somebody says something that you weren't expecting and then you're like, 'Oh, mental note, mental note, mental note, and we're going to maybe fork this into three different ways I wasn't anticipating going into this.' So just to tie it back to ECMA quickly, is it possible? I don't know that... I've never even thought of this, but can you go backwards in stages?
Ujjwal Sharma: Yes. And as a matter of fact, in this upcoming meeting, a notable proposal, decorators, is going back in stages. So yes, you can. As I mentioned, it's a sort of public symbol or a publicly shared indicator for how stable or how mature or how far along a proposal is. And fortunately, sometimes we push something along and realize a bit too late that it's not finished. So it's safer to just push it back a little bit and revisit the topic than keep going with it.
Brian Kardell: Yeah. So going back to CSS, the other thing that you made me think of that I totally had spaced on, I mean, not spaced on, but was not in any of my process of thinking, was the thing that I was trying to say is say selector's level, pick your level, that as a spec, doesn't sink or swim as a spec. It's the individual selectors that individually sink or swim and that's the sort of comparison I was trying to make, is that very often, whether something is editor's draft or REC or whatever, is ultimately not the important thing. The important thing is what is implemented. And then I was thinking, well, we do our best to keep making smaller and smaller, more discreet specs, which is also a change since we're talking about these changes because once upon a time, there was just CSS1 and CSS2, right?
Eric Meyer: Yeah. CSS specifications used to be monolithic, but that's because they were much smaller. CSS does have 'monoliths', it has yearly snapshots, which is basically, here's a list of all of the specifications. And the level that we regard as being part of this year's snapshot. So selector six might be a candidate recommendation phase, but it's not going to be in the yearly snapshot, the latest basically recommendation usually. Again, things are a little fuzzy because humans are involved, but yeah, that's as close as we get. But yeah, if you tried to take all of the CSS modules, which is what we call them, the individual specifications, and put them into a single tone, it would be enormous. It would be probably several thousand pages. This is a lot of work for just one working group to do, let alone an entire standards development organization.
Brian Kardell: Yeah. This goes back a little bit to this era when the W3C was new and definitely didn't know what it was doing. It was making recommendations that weren't standards, but people began treating them as standards, and it was very consensus driven and there was definitely this impression, it seems to me, that to an extent, if we come to these groups, if the group agrees, then it's going to get done. And this is another aspect of mental note in the things that you were saying, there are things that are REC or things that are candidate recommendation or that have implementation in zero browsers. I think that's something that's confused people a lot over the years and I think it's something we're still learning how to deal with. I do think that people are still confused by what is real and not. And I guess this is probably possible in ECMA as well. Are things ever marked as stalled or is it just like, just has no energy right now, let's just keep waiting.
Ujjwal Sharma: Yeah. There's an explicit sort of state for that kind of signaling this is stalled for real, not just slowed down, but the way... We call it the archive though, and unfortunately, as it is, it is a bit more of a graveyard than the icebox.
Brian Kardell: There is definitely this thing that I've seen happen in ECMA where it's like, 'Ooh, this thing is coming because it's stage one or two or maybe even three.'
Ujjwal Sharma: And then the problems start. Yeah.
Brian Kardell: And then it doesn't. It doesn't, yeah. And it's not clear to people if something is coming or if it's done or if it's like, where is this at? I think we could deal with more things like baseline, which focus on reality, which kind of ties me back to this era that we were talking about where nothing was happening, W3C had this impression that you bring people together, and if we talk about it in this group, we can make it happen. They were down this other XHTML, XForms, XSLT, everything was XML based. And meanwhile, the web was... Web developers were making incredible things with the same web that hadn't changed in years, and that led to this meeting in 2004, which led to the split and creation of the WHAT Working Group to continue an iteration of HTML in the browsers. And what's interesting though is it seemed like it was maybe not a permanent thing, but a this is how we get it back on track. And then in 2007, there was an effort to bring it back to the W3C and from 2011 to 2018, there was this two Popes. Do you know about the history of the two Popes? It's very much like that. There's the W3C HTML spec and the WHATWG HTML spec. And they had different editors and they sometimes would disagree. And so one of the things that was interesting about WHATWG and the split was that they were like, look, you got to convince implementers. They have budgets and they have all these practical constraints and everything. And if two implementers agree we can get it done, let's write down what people really do, not the thing we hope that they will do. It's very loose though, the WHATWG. They didn't even have any formal meetings. Everything was just done online, the editor was kind of superpower. It was very rough consensus and running code oriented like IETF. And in 2017, Apple, Google, Mozilla, Microsoft all finally helped put together a real structure with a steering group and IPR policy, which is wild to think about that we didn't get that until 2017, right? And I think part of that is because in 2019, then we came to a memorandum of understanding with W3C and now, the WHATWG is the official home of the HTML and the DOM standard. Didn't even realize, but do you know that the WHATWG now lists 22 standards?
Eric Meyer: I did not.
Brian Kardell: And also, just in the past year or two, they started having actual meetings, triage meetings where you can go and talk to, get the implementers together to talk to each other. It's basically like a working group meeting except that it's just like, well, it's about HTML and anything that touches HTML. And they started doing crossover meetings too where like, well, HTML and CSS touch each other, so we should make sure that HTML and CSS have a joint meeting every now and then. And doing that with a couple of groups, OpenUI especially, which is another development along the way.
Eric Meyer: So how does WHATWG do their thing? How do they keep it going? Because my understanding, it's not really an official... I mean, official, but it's not a formal organization from what I can tell.
Brian Kardell: Yeah. I mean, that's fascinating as well because even W3C, until two years ago or so, I might be wrong on that because sometimes time blurs together, but until very recently, was not... Actually, there wasn't an organization W3C, a legal entity. It was just a collective fiction of agreements between, basically, colleges that hosted people for W3C and collected money on behalf of the joint thing. It was not actually entirely dissimilar, except the collected money part. ECMA also collects money, right?
Ujjwal Sharma: Yeah. So ECMA has a fixed fee structure for all of its members and it hasn't been increased for a while, so I don't want to jinx it, but yeah, there's different member levels, and based on the involvement that you need in sort of the non-technical issues, you can pay appropriately.
Brian Kardell: Do you know what the max rate is?
Ujjwal Sharma: 70,000 Swiss francs. So this is approximately a little less than 90,000 US dollars.
Brian Kardell: Okay. W3C also collects money. Their top rate is I think 77,000. WHATWG doesn't collect any money, so what do they do with the money? I mean, this is a question I think a lot of people are curious about, right? What does ECMA do with the money? I think ECMA has just a couple of employees, right?
Ujjwal Sharma: ECMA has a number of employees, including the secretary general, and a few others like people doing the administrative stuff. I think most of the rest of their money is spent either on infrastructure and sort of organizing it. But yeah, it's a good question and I think unfortunately, we don't have a lot of transparency regarding the exact sort of the specifics of how they spend their money, but this is my understanding.
Brian Kardell: Yeah. I don't know. And to be clear, it's not a criticism of any of these organizations, but W3C spends its money on all kinds of things. I don't know that it's necessarily all the stuff that membership would like it to spend money on, but it has team contacts for each working group, although that's not universal, which is really interesting. So it's not obviously required. I also think it sort of sends a signal about how important anybody thinks your working group is if it doesn't have a team contact, but then it has a travel budget for like, whoa, well, this team that you have a person on a working group, they have a face to face, so you have to go to the face to face. And CSS Working Group has four face-to-face meetings a year, ideally. W3C has a lawyer, it has infrastructure, it's not great, but they have very nice integration that people wrote over the years for bots that do useful things that they maintain. They have their own mailing lists that people do still use, they have a Slack channel, they have all of the voting infrastructure, they have the website. There's actually quite a lot there. What's the thing that we all used to use a lot and now I haven't used it in a long time? The validator, right? And they have a CEO and a CFO now because they are a legal entity. It's interesting because you say ECMA has infrastructure, but who hosts Test262? Is it ECMA?
Ujjwal Sharma: Well, I mean, so now, a little bit better, but no. So originally, Test262's infrastructure by and large was owned by Bocoup who are sort of our allies in this space and have worked in this for a long time. Then we took on this work after they left TC39 for the most part and yeah, it's basically co-maintained by folks who care about it and that's pretty much it. This is something that ECMA could contribute to, but unfortunately, it's too specific to TC39 for ECMA to intervene directly, if that makes sense.
Brian Kardell: Yeah. I mean, just for point of comparison, W3C also doesn't host web platform tests, but they do spend money on helping maintain and build things that have to do with web platform tests, but they don't pay for the hosting or the infrastructure or any of that kind of stuff. W3C also, I forgot to say, has DevRel that do a good job and they have lots of community outreach. The whole thing built with the host models was in part also to be a center of gravity. There's stuff in... So you have Asia, Europe, US currently are kind of the big three, but they also don't do. There was an attempt for at one point, but they don't host the documentation for the web. They contribute, I think, sometimes to the writing of the documentation. They contribute to BCD-oriented stuff, baseline, those discussions, helping organize them and stuff like that, but nobody pays for studying the web. We do have HP Archive and Web Almanac and all that kind of stuff, but nobody pays for any of that.
Eric Meyer: Well, I mean, with those, those are Google projects, right?
Brian Kardell: Yeah. I mean, obviously, these things all exist so somebody is paying for them, but the point is it's not one of the standards development organizations, it's not collectively funded. It's a champion funds it, but then what happens is what happens if that champion suddenly can't afford to or-
Eric Meyer: Changes priorities or...
Brian Kardell: A good example is we let Mozilla just have it and we'd be like, 'Okay, well, I'll do it at MDN because Mozilla is the only foundation here. They're trustworthy and they employ staff writers. So, cool. We'll also help out, but great.' But then we learned to depend on them to host it and then to have the staff writers and then suddenly, they lay it off the staff writers and it was like, we had to create Open Web Docs, which is a way to collectively fund it, which there could have been a way to collectively fund it if that's the sort of thing that these organizations did.
Eric Meyer: Yeah. And Open Web Docs is not an SDO, is not developing standards. It's more of an SDO in the sense of a standards documentation organization, but it's trying to fund salary for three or four people and takes it a lot less than, for example, ECMA or the W3C, which we talk about how much money those organizations take in, but like Ujjwal said, ECMA's membership fees haven't changed in a very long time, literally a generation. They haven't.
Brian Kardell: Oh, yeah, 25 years in the case of-
Eric Meyer: Of ECMA.
Brian Kardell: ... ECMA.
Eric Meyer: Yeah. And the W3C hasn't raised them in 13 years. And so when you run the non-constant dollar calculation when you do the adjusted for inflation calculation, those organizations have taken huge pay cuts over the period of not changing their fees. Even though they're still taking in the exact number of dollars, those dollars don't buy as much anymore.
Brian Kardell: The one thing that we left off of what W3C does is TPAC, which is actually a really huge deal in my opinion. A bunch of these things, WHATWG kind of free rides on. The same people are involved in both in a sense and so WHATWG has meetings at TPAC. You can use the fact that W3C organized all that stuff.
Eric Meyer: Yeah. I presume somebody's donating their infrastructure as well for WHATWG, like hosting for example.
Brian Kardell: Hosting for the website?
Eric Meyer: Yeah, the website, the standards. I mean, they got to live somewhere on the web.
Brian Kardell: I assume that it is actually just GitHub-hosted at this point. But it was, for many years, famously because this was a thing where Hixie was very proud of the fact that he had a dream host account that costs $20 a month and that was he just paid for the whole thing out of his own pocket as opposed to joining, paying, all these big companies paying into one pool. But it was interesting the thing that you were just talking about, about how they've taken a pay cut and I wonder, it would be interesting to look at the membership over time and stuff like that and see how much money they took in 13 years ago versus how much money they've taken now.
Eric Meyer: Yeah, that's a good point.
Ujjwal Sharma: Yeah. On the ECMA side, I can say for sure that they have bled essentially over the years as you can see. Ordinary members, which is the highest tier of paying members, and to some extent, even total members. So yes, I think it's a relevant point. I think in some cases, the organizations are specialized, but I mean, in some cases, just less involvement. So would be interesting to see how much less money they get now versus in the heyday.
Eric Meyer: Yeah. I mean, even if they had kept exactly the same members, number of members at each level, over a generation that... Didn't you run the numbers, Brian, on what percentage?
Brian Kardell: I did, yeah. I did. So ECMA's taking in, for each dollar that it took in, it's taking in about 46% less value.
Eric Meyer: Or sorry, what 52 cents would have bought when they started, you would now cost a dollar?
Brian Kardell: Yeah, something like that.
Eric Meyer: Somewhere around there.
Brian Kardell: Yeah. Yeah.
Eric Meyer: And so the W3Cs is in it.
Brian Kardell: It's about 40% still about. So I guess, yeah.
Eric Meyer: COVID did a number on inflation.
Brian Kardell: Some of these things are really hard to compare apples to apples too, because they're not equal. So W3C has this concept, has always had this concept of wide review. And so it has the ARIA working group and PING and SING, the security and privacy interest groups, and WHATWG doesn't currently have those things. Again, I don't like to use the word free rides, but you get what I'm saying, right? It's not a criticism. It's just it's borrowing these things from W3C.
Eric Meyer: Yeah, some things. And a lot of the core work gets done by people who are paid by companies to, in many cases, do that work or devote significant portions of their time with that work. W3C also has people. There are people who work at companies, but their job is basically W3C rep or something like that. So they both benefit from that, but I think WHATWG probably benefits from it a lot more.
Brian Kardell: It's also worth noting that W3C does employ people in areas of internationalization and accessibility to have experts that don't work for one of those companies and is more involved than in CSS. I guess they used to have that with Chris Lilley at CSS, but for quite a long time, he's stepped back to just the team contact.
Eric Meyer: Yeah. I mean, all that taken, you're still, both of these organizations, W3C and ECMA, are trying to do a lot more with less. I mean, they may have more literal dollars, but those dollars do not go as far and they may have fewer literal dollars, I don't know, but however many dollars they have, those dollars do not buy what they used to. And so I find it fascinating that both of these organizations have just never even tried to raise... Well, I don't know if they haven't tried.
Brian Kardell: Well, W3C has raised in the past.
Eric Meyer: In the past. Right, but not for 13 years. And in the case of ECMA, not for 25 years. And it's interesting that they don't even have a structure that says, okay, each year, membership fees will go up by whatever the global average inflation was or something.
Brian Kardell: Or regionalized even, right? Even regionalized.
Eric Meyer: It could be regionalized, but if you wanted to make it sort of simple, just be...
Brian Kardell: Well, they are regional prices already.
Eric Meyer: You could do it that way too. But whatever thing it is you pick to say, 'Hey, the cost of living goes up. The cost of operation therefore goes up. Therefore, we're going to make increases pegged to this thing.' As opposed to, jeez, it's been 13 years, everyone's gotten real used to it, or 25 years, and everyone's gotten real used to it. How do we even broach the subject of, 'Hey, maybe we should just raise them a little, just a bit.' Because if you wanted to recapture all of that all at once, in ECMA's case, you would need to effectively double all of the membership fees and that would probably be difficult for representatives to sell to the companies that fund them, some of them anyway.
Brian Kardell: ECMA is having difficulty maintaining the members that already has.
Ujjwal Sharma: Exactly.
Brian Kardell: That's the challenge. I'm curious, I did look at sort of membership in the past versus now. Is there some way you could see who in 1998 was a big deal that's just no longer participating? When I say a big deal, I mean, big company that would send multiple people would be involved in multiple standards, took positions, was engaged, and there are a number of them that I think we lost through capitalism more than anything else. So companies that were acquired by other companies, and then they don't both pay a fee anymore and maybe they're still involved, maybe they're not, but now, they only pay them one fee or the rules prohibit you from doing that and so they just don't participate, only one of them participates. Do you see that kind of stuff too in the history? Do you know?
Ujjwal Sharma: Yes. And I have seen a handful, not too many thankfully, but a handful of members also leave or drastically reduce their involvement. In front of me, I think one I can mention is Meta, I think Meta in the last couple of years sort of refocused away from web technologies or at least investing so much in web technologies because by the time that I got involved, I mean, they had React and they were like, in one way or another, quite involved in the JavaScript ecosystem and the direction it would take through different technologies. I think that there is very little incentive or willingness within the organization to really invest a lot on web technologies at this moment. And so yeah, I think this is the counterpoint to Eric's proposal, which is like, I guess it's very difficult when you're losing members or to one thing or another, I think it can feel pretty difficult to sort of justify raising the dues with the idea that that might cause you to hemorrhage even more. So yeah, I think it's a difficult problem. There's also, I think, Oracle for example, used to be more involved, for example. So I think there's different things. I think there's corporate strategy and then there's also just capitalism, mergers, all of these things, which means that there's, in some cases, less diversity in the committee.
Brian Kardell: Oh, competition even. So in 1998, Netscape was a member, AOL was a member. There are these companies that they're just not even, basically don't exist anymore. I don't know. I get the sense that people are really questioning the value of paying big money to standards organizations. And part of the reason they're doing that is they're looking at things like WHATWG and open source, but I don't know that they're exact equivalents and that if you try to move everything from the W3C into WHATWG, that you don't inevitably recreate some of the same problems. I think we've talked about this a bunch of times with like, whenever there's a revolution, people are like, 'I hate the old thing.' Inevitably, you end up recreating parts of the old thing because there's probably some reason that they were the old thing.
Eric Meyer: Yeah, usually. We're back to Chesterton's Fence, as we've I think discussed on this podcast before. I will consider allowing you to remove the fence if you can tell me why it was put there in the first place.
Brian Kardell: Yeah. At the same time, I do think that W3C needs to change. Its processes, I mean, I know very good people who work on the process document and stuff like that. I think that there were some really positive changes in the past year where they really sort of asked, came around and tried to force discussions because that's a thing that's difficult at W3C, is actually getting people to discuss things among the membership because the membership is, in theory, it's a membership organization so we could change whatever we wanted. And the trouble is that the process is very cumbersome compared to WHATWG. So there even have been recently requests, surprising ones, along these lines in a way. Like Mozilla recently suggested bringing some W3C work over to WHATWG for, what is it Web Bluetooth and I mean, it's these things that were part of Project Fugu that Mozilla was before against and they still probably want significant changes, but there are reasons we would prefer to bring this over to WHATWG. Again, some of them have to do with the process and bottlenecks that occur in W3C and I think some of those criticisms are valid. There are theoretically good things, very good things in W3C like wide review, but in practice, they frequently happen too late. They're missed because there's too much noise. And I think in all of these, that's the thing we got to figure out because at the end of the day, everybody's most precious resource is time, right? It's like we got to get everybody to work together. That means we have to find common ground about what to spend our time on. And to me, this is where things like Interop are super, super valuable because that's exactly what it does. It's like, how do we find a way to agree on some things, maybe not everything, but in really important, practical ways, let's focus on these things this year. I don't know, those are my thoughts.
Eric Meyer: Yeah, it makes sense.
Brian Kardell: I think it's also worth thinking about how much do we want membership in this way? Again, I don't know what the lookout for ECMA is, but I can tell you that it's very much like a progressive tax system in that it's like the people who pay the most carry the most of the burden, right? So because the prices are progressive and they're almost an order of magnitude difference, you have to add a really lot of small organizations to equal one big one. And so you could be like, 'Well, there's 52 organizations, but there's really three major organizations worth of pay in there.' So I think it's worth thinking about how we restructure those because the way this works in WHATWG is also web platform tests and stuff like that where there are chairs, there are people who spend extra time and resources on this and organizing it and those are just paid for by Google. Just back to this point about the evolution, the constant evolution, there is also a proposal from somebody at Google for WHATWG to change the policies regarding stage advancements to include not just implementers. So they want to include, if you have somebody who's an accessibility expert that is invited but doesn't work for an organization, then that should... We want them to be able to vote. But it's interesting because the thing that W3C, that WHATWG is different about them is that it is more grounded in the practical. And so I don't know, I see them evolving kind of more toward one another in a weird way. It'll be interesting to see how this all plays out. I would like to hear what people write in about and just write to us, ping us on social media, email us, find us on Discord, whatever, and tell us how do you think it should work and what do you think that these orgs should do with money collected? What are things that are not being done that should be done? What are ways that we should be spending our collective power? Well, one thing that I really like just as a last final note before we go is Microsoft used to do this exercise where Microsoft would go to their internal partners and say, 'Here's 500 fake dollars, that's what you have to spend and here's the costs of all these things.' So you can't have all of them, you have to pick. You can only pick one because they all cost $500 or maybe this one costs $1,000 so you can't buy anything for two years if you buy that, right? Somehow grounding in a reality, I would like to see pools of money where you could use that to say, 'Okay, us 12 whatever, publishers, math people, gaming companies, whatever, who don't make a browser and all want our own things, clearly, we're not getting them. Here's a pool of money that we can use to advance one of those things in implementation sense.' And the fact that there is money and they have to choose somehow would ground that. Maybe it doesn't have to be even real money. Maybe there are ways to just change that in the stages process or something, to ground it in something concrete to force people to say, 'I want you all to get your thing, but we can't do everybody's thing. So which thing are we going to do?' Anyway, I think this was a really fun topic. I think it would be really interesting to see where all this goes in the next several years and maybe when we see another SDO spin out of all these, who knows? And these are not the only ones. There are more and they work even more differently. And so yeah, thanks.
Brian Kardell
Eric Meyer
Ujjwal Sharma