text stringlengths 0 1.36k |
|---|
One of the things you're really trying to do is just basically berate each of these companies into just like giving up on that. |
**Alex Sexton:** Yeah, but I don't think Apple would do that. |
**Mikeal Rogers:** So Apple has the opportunity to disagree with a standard going out, or disagreeing with their patents going into a standard before the standard gets published. But instead, what they're doing is vetoing the work ever beginning, and that's really problematic, because then we don't get to see that impl... |
**Kyle Simpson:** \[31:53\] Well, I think we'd be better off if more of the web platform could work a little bit more, like TC39, and there's some huge caveats there, of course. I think what's good about parts of the open web is that we can see things like somebody saying "Here's an idea for a feature that should go in... |
I think that there are still ways that we participate, and I wanna just go back to -- yes, it's true that there are some mechanisms in place where people's paid memberships do give them some veto power, but having a group of four or five players that have a bit of an adversarial setup between them is still far better t... |
So when I say that the web means that we get to participate in it, I'm saying that at least it's not controlled behind some one closed door in a boardroom; there is a group that controls it, and we don't get all that we want, but we get a lot more now that we would have were flashed to be the web that we were building. |
**Mikeal Rogers:** Also, I think one of the keys though is that you really do wanna get rid of that veto power. TC39 tries to reach a consensus, but at the end of the day if they have to, they'll come down to a vote and each member just gets a vote. If you have more members, it decreases the power that any one particul... |
And I do wanna give a shoutout to TC39 who runs the -- they own the JavaScript standard. They've done a tremendous job at just improving their process over the last few years... One, in making sure that things don't get certified before we know if they can remain compatible, a lot more participation from outside people... |
**Kyle Simpson:** So now that we are all completely in 100% agreement on what the open web is and should be, let's talk about some of the things that are I think existential threats to the open web. I don't know if I agree with some of the claims that Joe Hewitt would be making, but I'm serving as a coach here this yea... |
**Mikeal Rogers:** It goes beyond just for videos. It goes way beyond just that. |
**Kyle Simpson:** Sure, but they want control because they wanna be able to sell stuff and not have piracy, and that makes sense, they need to make money, but they are biting into one of the fundamental principles of an open web, which is that a developer is completely in control of that experience, and they're saying ... |
**Mikeal Rogers:** \[36:08\] Not just developers, but users as well. The users of the open web are no longer in control of their browser. |
**Kyle Simpson:** That's true. That's why Tim Berners-Lee is so vehemently against it and why he's campaigning so vehemently against it. He created this thing and he's saying "Wait a minute, we need to be worried about the direction that that's going in." But there's no clear right answer here, because keeping it out m... |
The other big one is ads. That's the big elephant in the room, because again, there's a monetization model that we need to support for the web; people wanna build businesses on the web and not just have hobby websites. We need to support business on the web, and part of that is people giving away content "for free" but... |
**Mikeal Rogers:** I think even just beyond privacy, these ad networks have actually also become vectors for people to do different phishing attacks, and things like that... And not just phishing, but also getting you to download something and then taking over your entire computer and having that kind of extortionware ... |
Most people, when they put an ad on their site, they don't do the content deals to get that ad placed; they rely on some kind of ad network to do that placement, and those ad networks are being constantly scammed by incredibly sophisticated attacks to get these ads in. So literally, these are not verifiable; it's not p... |
Brave has been incredibly aggressive about ad-blocking and tracker-blocking, and as a result a couple sites won't work, because they're reliant on those features. But if you look at some of the announcements that came out from Safari, and I know that even some of the Chrome people are looking at it - some of these feat... |
**Kyle Simpson:** And maybe they should be, right? Maybe we did put features into the web that sounded great, and then we kind of came back afterwards and we're like "Ugh, I don't know about that." |
I remember -- I think one of the classic cases was when you have a link that gets changed to a different color when it's been visited, and then some enterprising hacker has figured out "Oh, I can track somebody's browsing history by making links in the background and checking the colors to see if they visited it." |
\[40:06\] Well, rightly so, we had to rein back in some of that functionality to say "Whow, whow, whow... Your ability to check on the color of a link is not more important than a users' need to have privacy." That goes back to -- the web platform is guided (or should be guided) by this thing called "the principle of c... |
**Mikeal Rogers:** I think one of the things that we really need to point out too is that the reason why this is an existential threat to the web is that there are constantly competitors to the web, and they are predominantly proprietary competitors. But the reason that the web has always won, and why the web won when ... |
If you go to a bunch of different websites, your computer doesn't get slower the more websites that you visit in a single tab. And the security model is really important. You're literally running arbitrary code that random developers wrote, on this computer, and somehow these browsers aren't constantly being taken over... |
So the reason why an app store like iOS has might beat the web isn't because people just love their iPhones and there's some kind of experience thing, it's the performance and the security, and we need to be better, we need to be leading the way in terms of performance and security. If we're not, we're gonna lose out. |
**Kyle Simpson:** There's obviously a huge push in the last year or two, and Google spearheaded the effort largely towards progressive web apps, to try to bring web apps to parity with native apps. Some people say "Well, maybe there shouldn't even be parity. Maybe they really should stay distinct", but there's a big pu... |
I would argue that end users understand stuff like "This zaps my battery really quick and it makes me have to recharge my device a lot." I'd argue that they understand stuff like "Wow, that cost me a lot of bandwidth." That's for people in parts of the world that don't get unlimited bandwidth, like we might get here in... |
Those are currencies that users really care about, and if we want to get to the point where end users care so much about the web versus app platforms, that they vote with their dollars that they buy the web rather than buying apps. We're gonna have to speak to those currencies and not just to the "Oh, I feel really goo... |
\[43:54\] I totally get why service workers are awesome, but an end user is never gonna care whatsoever about that. They will maybe care about offline, but they're several steps removed from what we focus on as developers and what a real end user is gonna vote with. |
**Alex Sexton:** Sure. It's a weird discussion to me, because we had a time when the web was the leading thing, and it was not a good time. I don't think the web is set up to be a good leader, because things are so difficult to change in a good way. The process for adding or removing or changing something in the web is... |
I actually find them to be this really good, symbiotic relationship; it's the gut bacteria inside of a human, or whatever. The explosion of good, fast mobile applications was an absolute 100% driver in quality, good improvements in the web. It is the single most important event that occurred that caused the web to get ... |
**Mikeal Rogers:** Are you telling me that I shouldn't build my brand new artificial intelligence virtual reality startup on the web? |
**Alex Sexton:** I'm telling you you shouldn't build that at all, because it sounds dumb. |
**Kyle Simpson:** \[laughs\] Alex, I think you have a fantastic point. It does really kind of ask the question for all of us (developers) to ponder - is the fact that the web (and especially JavaScript) values backwards-compatibility so deeply in its DNA (something written 22 years ago is supposed to still keep working... |
**Alex Sexton:** Yeah, I think it just causes us to need to create different solutions. Rather than making a backwards-compatible change, we make a sister standard that can work alongside the old one. There are plenty of ways to store data in a browser that aren't cookies now, and we didn't have to kill cookies in orde... |
**Mikeal Rogers:** I think the only wrench in your argument is that the Apple watch is the number one device that's sitting in somebody's cupboard somewhere. \[laughs\] |
**Alex Sexton:** Sorry, I'm confused about what that has to do... |
**Mikeal Rogers:** You were saying that you can only be one device back, or you constantly have to be getting the new device, and I think that there's actually like a fair amount of fails that happen as well. I'm still not convinced with all the VR stuff at all. |
**Alex Sexton:** It's riskier on their end, especially when they have to make hardware. That's a business I don't wanna be in. But I guess the openness of the web and the hopeful inclusion of the most amount of people via the web -- like, the non-prioritization of companies, or Americans, or whatever, hopefully... Of c... |
I think it's pretty fundamental to the inclusion, the neutrality, whatever. Even though we fall flat in that way and in so many other ways, we should always be getting better. I think that's why it's a fundamental part of that, and I don't think we should get rid of the backwards-compatibility stuff. Rarely is that the... |
**Mikeal Rogers:** Don't get me onto this... You're gonna make me go off on a random tangent now. |
**Alex Sexton:** Yeah, it's stupid and sucks and like there those people who use the bad thing are the ones who should change. But at the end of the day, add a freakin' letter to your thing and it'll keep working. It's an uglier platform because of that history, but it doesn't fundamentally break anything, in my opinio... |
**Mikeal Rogers:** We're gonna take a short break, and when we come back we're gonna get into our project of the week, which is the first WebAssembly based project that we've done on the show. Stick around. |
**Break:** \[50:59\] |
**Mikeal Rogers:** Alright, so the project of the week is Blake2b-WASM. I'm not gonna spell this out, I think that you're just gonna need to check the show notes if you want a link. This is Mathias Buus, one of my favorite programmers in the Node and JavaScript community. BLAKE2 is a fast secure hashing algorithm, and ... |
When they first started talking about WebAssembly, one of the things that they kept talking about was "We'll be able to do these incredibly fast and efficient algorithms." This is actually a pretty discrete algorithm that we need for a bunch of different crypto. Mathias has actually written this in C++'s native module ... |
**Alex Sexton:** That seems like the whole idea behind WASM, like being able to still write in regular web languages - you build your interface in JavaScript and React, whatever you wanna do, and then you have a worker that hits WASM stuff for really complex things, like the BLAKE algorithm, or any kind of hashing algo... |
**Kyle Simpson:** Do we have any idea -- I was just taking a quick look at the repo... Do we have any idea what the source language of this is? When I look at that repo for Blake2b-WASM, it just says "implemented in WASM." Did he literally write the s-expressions, or was that transpiled from some other source language?... |
**Alex Sexton:** Yeah, I imagine he took the standard implementation of BLAKE. I'm trying to find the thing... I think it's C++. |
**Mikeal Rogers:** Yeah, based on the comments, it's compiled. |
**Alex Sexton:** Oh, actually... There's a Go implementation as well; it could be Go. I think it does matter, but in this case it seems like he \[mumbles\] |
**Mikeal Rogers:** I think he may have compiled, but I think he may have also tweaked it by hand. The wat file is only 25,000 lines. |
**Kyle Simpson:** Only 25k? Wow... \[laughter\] What I'm getting at is that I'm excited about what WebAssembly is gonna do for performance. I'm waiting around until we start to see, because I don't think it will be too long - maybe a year at most - before we start to see frameworks saying "You know what? We went ahead ... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.