Lesser known things browsers can do in 2020

You know the sound of those BMC beats means it’s JS Party time, y’all. My name is Jerod, I am happy to be here, and I’m happy to be joined by regular panelists, Kball and Divya. What’s up, y’all?

Yo, yo!

So we’re kicking it live in 2020… We’re no longer predicting 2020, we’re living it! We’re living it for reals. And we’ve found a very cool repo this week; shout-out to Luigi DeRosa, @luruke on Twitter; you can find him in the show notes and send him a tweet. He’s putting together this really cool repo called Browser 2020, the pitch of which is “Things that you can do with a browser in 2020”, and here we are, in 2020… So we thought “This would be a cool conversation piece.

His description is “This repo contains a non-exhaustive list of less known features implemented in browsers today”, or in some cases just one browser today. We’ll get to some of that. But it also says “This list isn’t intended for a technical audience; instead it wants to be a ‘I don’t know we could that in a browser’ list.”

That being said, we’re a technical audience, and yet there are things in here that I didn’t know you could do in a browser, and so we assume listeners as well could learn a few things, as I did by going through this repo.

Let’s go through it a little bit, and I wanted to start with this question of surprise, and things that we weren’t aware of. We are technical; that being said, what items in this list were surprises to you all? What were new to JS people like us?

I was actually surprised by the Web Share API… Though partly I might be surprised by it because I think the UI is pretty terrible, from what I can tell, on non-mobile devices…

[laughs] That surprised you, how bad the UI is?

Well, I hadn’t seen it, and then I realized the reason I hadn’t seen it was for anything except a mobile device, the UI looks terrible. So if you do the web share test I think on Safari or Chrome or something like that, it just looks pretty ugly. But if you do it on a phone, you get access to your native system sharing thing, so… People are probably using this when they’re doing mobile web, and then falling back to Share This, or something else that actually gives you a nice UI on desktop and mobile.

So the description of this, at least in the repo, is “Delegate the share of URLs or files to the OS, providing share options based on the installed apps and user preference.” So there isn’t a UI though, right? Because that’s the whole point, you’re passing the URL off. Are you saying that in the Web Share test, as far as on macOS, that little Share UI is subpar?

[04:05] Yes – well, actually no. Now I’m testing it and it doesn’t even exist on my Chrome. It’s like “Oh, that’s an unsupported feature.” [laughter]

Oh, yes. It does work on Safari. It’s ugly as sin.

Now, is that OS’s fault, or is that the API’s fault?

I think it’s actually happening in the browser… But I don’t know.

That doesn’t make sense to me, because the whole point is to pass it off to the native OS.

That’s true. Yes, so maybe that’s macOS being like [unintelligible 00:04:40.27]

This was also surprising to me - I didn’t realize this was a thing, and maybe because nobody’s really using it, maybe because it doesn’t work in all the browsers and whatnot, and maybe because the… Definitely the share sheets on mobile tend to be more refined than the ones on at least macOS. I’m not sure how Windows or Linux’s share sheets work, if they have them at all.

But on mobile it’s nice, because it will sub-select the apps you have, like if you have the Buffer app installed, or a Slack app installed - all of that stuff just works, whereas on your own webpage you are stuck implementing “Share to Twitter, share to Facebook, share to Reddit.”

Right.

I’d almost rather if it only worked on mobile… Because if you do proper feature detection, and you say “Okay, does this thing exist?”, that doesn’t tell you “Does this thing exist, but it’s super-ugly, because you’re actually on desktop Safari.”

That’s true. The example in the web.dev website shows the feature detection, if (navigator.share). So this exists inside the navigator object, and it’s the .share function… So if that function exists or is defined, then call it, basically is all they’re doing. And it’s promise-based.

Question about Share buttons - I’m so anti-Share buttons. Maybe I’m a curmudgeon, but I like this because it’s like “Well, if your OS will do it, cool.” But if not, I don’t know; copy and paste the URL, or something. I hate having Twitter’s JavaScript on my website, and Facebook’s JavaScript… I end up hand-rolling a lot of the Share things just to construct the URLs myself, because I’m just not a fan of the Share This, or like the “Here’s the gigantic list of social buttons”, that are slowing down my web page. What are you guys’ thoughts on that? Am I a curmudgeon, or are you with me?

With my developer hat on, I’m 100% with you.

Put on your other hat.

Well, my other hat was as a small business owner, which I guess I still am, even though I’m working for a company now, at the recording’s time, but…

[laughs] It’s complicated.

It’s complicated, yeah. My business relationship status is “It’s complicated.” But as a small business owner and doing marketing, it actually makes a huge difference.

It does.

Even without tracking. It makes a difference in the – I mean, yeah, you can hand-roll it, and actually I do have a site where I hand-rolled them, because I didn’t want third-party JavaScript on my site… And that’s another question. But having the prompt and the easy “Click to share” that prefills things, I have observed that it makes a dramatic difference in the viral echo that you’ll get on a piece of material.

That fascinates me, because as a user, I never click those buttons.

You’re not the typical user, Jerod.

Well, I know that, but that’s why I’m asking - do you guys click the Share button? If you just see a button that says “Pinterest”, and it’s all non-retina, it’s in a list of a gazillion button - does that urge you to share it on Pinterest? Or would those people have shared it anyways, and they’re just hitting the easy button, or not?

Yeah.

Yeah. I think if you’re not – I’ve totally lost what I was gonna say. [laughter]

I’m just gonna sit here and let you swim in it. [laughter]

You know what I’m gonna do… On my ZenDevs blog posts I use – I think it’s “Share This”, and I have four default buttons, plus their Share thing. Let me log in - if I can remember the login - and see how many people actually click on those things… Because I think it’ll tell me. If I can remember which account that’s under; I guess it’s this one. Social analytics… Oh, I don’t have the analytics enabled.

This is compelling radio right here…

Hang tight, Kball’s doing password reset flow… [laughter] Will he get the email, will he not? We’ll never know…! Okay.

This will be fun, actually… Social activity over the course of 2019. Let’s see if we can do that. This is by ShareThis’s report. Sweet. Oh, interesting. So it tells me not only what did people click on to share, but it also tells me what things referred to me. These numbers don’t look…

Very good?

Well, they’re not very good–

You’ve got a bunch of Jerods reading your website?

Yeah… Oh, it’s not actually giving me all of the – that’s the problem. I was like “These numbers look really small.” I told it to do a year, but it’s only showing me the last 30 days.

Well, here’s something to Divya’s point, on mobile at least… What I’m a bit confused about this is - the Share UI on mobile Safari and on other mobile things is right there in the browser’s UI. You don’t have to trigger it from a web page. So why wouldn’t you just hit the little Share icon on your browser? Because you’re on the URL, you assume you’re gonna share this URL, right? Why would you want it in your page? Like as a strong call-to-action, like “Share this!” and it just pops their thing? I could see where that would be cool…

Yeah, that would be helpful. Yeah, people apparently share my stuff on WhatsApp, on SMS, on Blogger, on Evernote… All of these things.

So you’re using one of those big “Share This” things that just includes all the social networks…

[12:04] What I have is I have four individual buttons: Facebook, Twitter, LinkedIn and Email. And then I have a generic Share that pops up one of those things. I set this up forever ago… And it’s slow as sin, I totally get not wanting it. But yeah, it looks like people are going through those deep dives; they’re digging down and finding the one they wanna do.

Alright, so there you have it - some people click on it, some people don’t. Let’s move to the next API, because Web Share is surprising, but probably not worth as much attention as we’re probably giving it right now. Divya, what surprised you on this list of lesser-known APIs?

Right.

Yeah.

Can’t you do that now?

So does it work with different kind of media streams? Could you start a podcast and have it go picture-in-picture, and then navigate away from the website? Because I would definitely implement that on Changelog.com.

Cool.

Right.

I’m with you. I thought that was just a YouTube thing, and I’m realizing that’s actually an available API for anybody who’s doing video, or I guess audio as well, to implement into their website… And it’s a user-agreed-upon thing.

I think you might have to fake some video…

Oh, yeah?

Well, I’m just scanning through the doc here, and it says “Your promise may reject for any of the following reasons”, and one of them was “Video file is audio only.”

We could just do an animated GIF of Kball doing the salsa during JS Party… [laughter]

That could be why.

Right.

Well, I’ll say the most surprising one to me was the Web Coffee API. And I’ll say it’s Nick Nisi’s favorite new API. I won’t say anything else about it besides - link in the show notes. You should definitely click on the link in the show notes and check out the Web Coffee API.

[16:12] Oh, dear…

Let’s talk about the useful ones… Because surprising is one thing. And everybody likes Web Coffee APIs. Check it out if you haven’t checked it out yet.

I just checked it out, and I was not sad.

[laughs] Not sad.

You know I love coffee…

There you go.

What about useful? What are things that are on this list that you either are using today, or would love to integrate now into some sort of project, or waiting for an excuse to use? Things that people actually likely are interested in on this list here.

I think payments is huge. I think we’re already starting to see the impacts of payments, but mostly through apps, and stuff like that… But particularly when you look at mobile payment things like Apple Pay, or Google Pay, where you can authenticate with your thumb or your finger, or I guess in newer version of the phone with your face, what have you… But the frictions for buying something on the web, particularly in a mobile device, where typing and credit card numbers and things like that is a pain in the butt - it just takes that away.

We’ve had that, to some extent, with native apps for a while. I buy a lot of things with the Amazon mobile app, but… Opening that up to any e-commerce site out there really lowers the barrier to entry. That could be something that, for example, Shopify implements. In fact, I would bet that they do implement.

They do.

And suddenly, anyone who has an e-commerce store using Shopify - their products are available for somebody to buy in a seamless way, using their phone. That’s a phenomenal increase in usability for e-commerce on mobile. I would bet that alone dramatically increased the conversation rates of mobile commerce.

Yeah, and it’s reasonably well-supported; it’s on 15% green on browsers, but it’s got the big ones in terms of mobile - which is mobile Safari - and there’s partial support in something like 74% according to Can I Use, so the Payment Request API… So it’s pretty usable today. It’s not gonna hit all the browsers, but it’s gonna save a lot of people a lot of time, and really reduce that friction, like you said, Kball, and make it seamless for those who are ready to buy.

There’s that moment when you’re ready to buy, between the point that you want to and the point that you actually get it done, and lots of people drop off it that’s long and tedious. I know I sure have. You make that as seamless and as easy as possible and you’re making more money.

[laughs]

Prove it, prove it. What is it?

Oh, snap! [laughter]

[20:18] Absolutely. And I would say – as somebody who likes to support small, indie businesses and small e-commerce sites, and yet doesn’t like to just sprinkle his credit card around the web, like fairy dust, for anybody just to fall into their lap, I have this weird dichotomy where I’m like “Well, I would love to just shop at random mom-and-pop shops on the internet”, but I also don’t want to do that a lot of times… Because I’m like “Well, are they gonna really–”

So just the proliferation of platforms like Shopify, Stripe… Like, when I see the Stripe dialogue, their built-in checkout, as somebody who understands all about how that works - I realize we’re a subset here as well - I’m pretty excited about that. I’m like “Okay, this is something I feel like I can trust.” I have more trust because of it. Same with Shopify. I think this API style leads to that, because you’re allowing it through your OS, right?

Yeah. Be cautious on that…

Sure.

…because the naive implementation of this, your credit card number still goes to that site.

Yeah.

There are token-based versions that are more secure, and I think the Apple Pay implementation and all those use token-based ones… But the simplest, most naive way to implement web payments API as a website - you still end up with a credit card number.

The most naive implementation – it’s literally just an autofill; is that what you’re saying?

Essentially, yeah.

Gotcha.

No doubt that Web Payments API is very useful… What else looks useful on these things that you can do in 2020 and maybe you don’t know about, but can be super-useful for folks?

Hm… I don’t think I understand this API. Who gets the SMS? Is this a browser API?

I think so, yeah.

I’m looking through it as well… One of the things I’m trying to understand – so basically you just get access to SMSes that happen…

OTP, one-time password, right?

Yeah. I just don’t understand how the front-end receives an SMS. It’s the SMS Receiver API - like, who’s sending it? You have to have access to that system.

I’m looking at code examples, because there’s a repo for questions… Let me drop the repo I’m looking at here… So it looks like basically you run navigator.sms.receive() and you await the response, and then that probably asks for permission at that point…

And then it comes back (I assume) with the next SMS that comes in. So if somebody happened to send you an SMS at the same time, you would get that one instead, which could be awkward, but…

I would like it to be able to say – and have this be part of the permission flow, too… Say “We’ll get an SMS from this number”, or from numbers matching these criteria, or something like that…

Because I’d be really hesitant to say “Yeah, sure, you can read whatever SMS comes in to me.” But if it says “Okay, they’re asking for permission to read SMSes from this number.”

Yeah, it seems like a worldwide race condition.

[28:07] Like, who can send the SMS faster? This website that you just asked to, or significant other, who happens to be saying something to you, and now it’s reading that to run a regex against it.

Well, that’s how my wife and I always chat…

Well, that regex could be whatever, right?

Exactly. It doesn’t have to be a regex. It could probably just be like a slurp.

Yeah, it looks as though – I mean, once again, I’m just looking at this example code right here… But it looks as though the regex stuff is in your code, it’s in user code.

So instead of regex, you could say “Post whatever you get back to my web server.”

Exactly.

So this one fits in our next category, which is “How many of these are ripe for abuse?’

Easy to abuse…

Yeah, exactly.

This is definitely interesting… We should state this is in – what’s the status? It’s very early days… Origin trial?

I wonder if – do they ask you for permission each time, or do they only ask you for permission the first time?

That’s interesting… Because what if I ask for permission once, but then I just sit there, listening to every SMS that comes in? I have the valid approach of “Oh yeah, we’re gonna do a two-factor authentication. Great. We send you an SMS, you verify that”, but if it doesn’t ask you for permission again the next time, you could just sit there waiting for SMS.

Mm-hm. Read every SMS that comes in until they leave your web page.

There’s a nice warning on the SMS Receiver API announcement. It says “Attackers can spoof SMS and can hijack a person’s phone number. Carriers can also recycle phone numbers to new users after an account was closed. While SMS one-time passwords is useful to verify a phone number for the use cases above, we recommend using additional and stronger forms of authentication to establish new sessions for these users.”

Yeah. I don’t know about you two, but I’m gonna give this API a hard pass in my life. I’m just like, not interested… I don’t know. There’s other ways, like AuthN; that’s the other one.

And that’s out there, right? Oh, it’s on the list; it’s in our document, WebAuthn.

Gotcha. So maybe a thing you can do in 2021, not 2020.

Not sure.

When is the next version of you gonna come out? That’s the question. [laughter]

I don’t think that counts as a browser API.

Not yet.

[31:48] Not yet. WebAuthn is actually really cool, because I feel like I have that experience now because i use a password manager, and I have it hooked into my biomarker of my fingerprint on both my phone and my laptop… And the WebAuthn approach basically means you could do that without having to understand password managers, and without having to deal with that stuff. You’d just have – any user can authenticate with the biomarker, or a YubiKey, or whatever, and they don’t have to deal with passwords. And the whole password stuffing, reused password etc. just goes away.

Yeah, there is a spoofing question, though… Especially with things like iPhones and all of that, with using the face recognition.

Wasn’t there a great video of somebody going around–

Oh, that wasn’t the one… I was saying somebody – their Airbnb host had a similar face authentication, and they were able to spoof it just with pictures from social media.

Yeah, it turns out faces may be not the best authentication approach… [laughter]

It’s great unless you have an [33:42] twin.

Or you don’t have your bizarre avatar, Jerod, and your face is all over social media…

Ooh, points for me, for bizarre avatar.

That’s true. Let’s go through a couple of these features that people are either likely using right now, or they’re very approachable. I think prefers-color-scheme is a cool one, which basically allows you to detect light mode or dark mode, according to what the person has set OSwide.

Prefers-reduced-motion is another one that you can use. It’s kind of an accessibility feature for those who can get – what do you call it… Sick, I guess, with too much animation?

Motion sickeness, yeah.

I like features like these, because they just gracefully fallback, right? If the browser doesn’t support prefers-reduced-motion, then it just doesn’t reduce motion.

WebXR is pretty cool, too. Have y’all seen AFrame?

I’ve heard about it.

What is it?

Like a component library that turns it into VR?

That’s rad.

The other thing that’s really cool about WebAR and VR is that it’s like your epitome of progressive engagement… Because you can interact with it without any sort of VR equipment, and it falls back to WebGL-type 3D rendering… And then if you get excited or you have the equipment, you can put it on and go deeper and deeper and deeper.

I’ve also heard that one of the things that WebVR exposes that other VR situations don’t, that has the potential to make it more powerful that any other VR setting, is linkability. So you could link into different VR experiences, which is super-cool to think about.

Is that like the umbrella term for both AR and VR - XR?

Yeah, that was awesome.

Bring it.

Fully an adult. IKEA is the main use case for this.

This is Divya’s heuristic of knowing when you’re adult - when you find a new, interesting technology feature and you think “How can I use this to decorate my house?”

[laughs]

Thanks.

Kball, now that your childhood’s dead, did you wanna say something?

Yeah, I was actually gonna comment that I was surprised to see not that long ago a really valuable business use for AR, which was as a way to collaborate between architects and general contractors, and things like that, as they were developing the architecture for what they were gonna build, and being able to essentially take the plans and see them, and walk around, and experience what it is, and feel what it’s gonna look like, prior to actually building.

I thought that was a super-cool example of this technology, which has for a long time kind of felt like a cool tech in search of problem domains…

Yeah, demo…

…being used for a really real problem domain.

Yeah.

But now you can get the Snap glasses, right?

Right. Everyone’s taking a crack at it. They say Apple is releasing something-glasses in the next couple of years… I think ultimately somebody does need to crack that nut in order for it to go mainstream, these use cases…

Because ultimately, you don’t wanna hold your phone up, or your iPad up to do AR. It’s cool and it’s fine, but if you wanna be a regular part of consumers’ everyday life, put it in our viewport, and do it in a way that we actually wanna wear it. I mean, that’s the problem with Google Glass - you just looked like an idiot wearing it.

Yeah, creepy factor, for sure.

But this challenge is something that any really disruptive and new technology goes through…

We as a society need to figure out what are the ways that actually can use this, and help us in a way that’s productive.

I remember seeing stuff around when we first adopted electricity - it took 50 years before we were fully reaping all the benefits of electricity… Because not only did we have to spread distribution and get it equally out - which we’re still figuring out with AR, like what are the ways that it’s going to be distributed - but we had to redesign how we use things, how we set up our houses, how we set up our workspaces, and all these other things. And we’re getting faster at those things, but we haven’t figured it out for AR and XR. It’s just isolated niche cases so far. And I think the power of it is so large, that – you know, between AR, XR and AI type things, those two things are places where we’ve made incredible strides and we’re now in that figuring out stage of “How do we use these things throughout our lives and throughout our work to reap the benefits?”

So after looking deeply at the SMS Receiver API - not a huge fan. There’s lots of APIs that you could look at and you’re like “Why? Why are we doing this?”

Because that’s how it kind of is, right? As a developer, you get excited about something, you look at it, it looks very interesting, and sometimes you hop into the details and you’re like “Hm. I’m not so sure. I’ll come back in six months and see how it looks.”

[laughs] “I could use this, but I don’t want you to use it, because you’re gonna screw it up.” I get that.

And that person is usually Feross… [laughter]

So I’ll start on this one, because I have a bit of a beef with push notifications in general, because I feel like they are the robots running our lives, and we’re supposed to be using them to help our lives… And they can very quickly take over. So now many websites have been asking if they could send push notifications, and it’s like “I get enough push notifications already. I don’t need one from your website.”

Surely, this is best for web apps, versus websites… But any API that is generally available will be generally used. So the Push API allows browsers to register push notifications. You can send later, etc.

I say no every time.

[laughs] Exactly. And aren’t you kind of mad that they asked? I’m kind of offended. “Did you thought I might want this?”

“What I have to say is so important that you need to know immediately.”

[47:53] There’s a global principle here though, which is like “Don’t ask for a privilege out of context.” I land on your site; no, I don’t want a freakin’ push notification.

Right.

But if for example I’m using a calendar and I set something up and it says “Do you want a reminder? Okay, would you like that as a push?” then maybe I’ll say yes. Like, put it in context. Why the heck would I wanna give you a push notification? I can think of some reasons… Calendars, if I’m using a web version of Slack, if I’m doing some other things - yeah, okay.

Yeah. There are legitimate uses for this.

Wouldn’t you want an email on that?

I would almost always want email in these cases… Unless the case of like a Slack application, like Kball said…

Maybe I’m just a hawk on my email too much.

I stay on top of it, so that works for me for certain contexts that it doesn’t work for other people. A gift and a curse… But yeah, that makes sense. I like that. If there’s tickets going out, and like this thing is gonna sell out now, and your only context is this browser, then send me a push notification.

But what if it just goes to your laptop then, and you’re on the go? Missed it. You’re gonna miss JSConf again. It definitely has some valid use cases…

You could say both, right?

Send this to my phone?

Yeah, you could do that, too. Once we open the realm of “This is something that we’re actually in a dialogue with the user about how they want this to be used”, there’s a ton of possibilities. “I want this as soon as I can, wherever I am. Send it to my phone, send it to my email, send it to a push notification.” “Okay, sure.” There are situations I can imagine that.

Yeah. It’s “Okay” or “Leave this banner here until you leave the website.”

That’s a user experience choice though, and it happens to be the one they’re all doing.

I don’t take that choice. I take the “Just do what you’ve been doing” choice, and deal with the ramifications, I guess… Because a lot of that is based on a lawyer’s advice based on an interpretation of a law in a place, and it’s like CYA kind of a thing. And then you see a big website doing it, and you’re a big website, and you’re like “Shouldn’t we be doing this? Because their lawyer probably knows what they’re talking about.” And then you go talk to your lawyer and they’re like “Well, it’s not clear according to these laws, but yeah, you should probably just put it up”, and then all of a sudden you see it on every single website on the internet. Isn’t that kind of how it goes?

Yeah, I mean… You could also choose to allow – like, if you put that banner, give them an option saying “No, I don’t wanna be tracked.” “Okay.”

Window.close.

[52:02] [laughs] They say no and it just autocloses the tab, like “Well, you didn’t want it…” There you go.

Well, a lot of the stupid interaction or UI there is based on a very quick reaction of “We have tracking everywhere. It’s gonna be hard for us to remove it, so we’re gonna say ‘Say yes, or don’t use the website.” But that’s based on coming from a domain where it’s free and easy to track everyone all the time. If you’re building a new website, you don’t have to make that choice anymore. You could say “I’m actually going to listen to what my users say and choose not to track them if they say no”, for example.

I agree 100%. What other APIs are bad, in our humble opinions? Well, we talked about the File System API on our 2020 show, where I was worried that there might be potentially abuse… But it has a lot of opportunity. As Chris mentioned, there are cool things you can do with it, so it’s not like it’s bad… I guess it’s just open for abuse.

But in terms of things where you look at this list and you wonder “How many developer hours are going into this feature that – I don’t know if browsers should really do this anyways”, are there any of those?

I’m kind of wondering about that Vibration API…

Oh really? [laughter] What’s that one? I didn’t see that.

Basically, it’s what it sounds like. [laughter] It’s what it says in the description.

I was hoping for a surprise…

It provides access to the vibration mechanism of the hosting device. I mean, it kind of makes sense; you have a phone, you’re doing stuff on there… There are situations, especially if you’re doing games, where you want the device to vibrate. Okay, fine.

Yeah. Like a Rumble Pak on the old N64.

Sure. Okay. And I play games on my phone that use that to good effect. However, it’s a very minor bonus. It doesn’t help that much. Maybe it helps a little bit… And it just seems like it’s so ripe for abuse.

[singing] Talkin’ ’bout those good vibrations…

[laughs]

Yeah. Like, if you don’t see it, it’s like “Vibrate. Vibrate.” “Hey, we’re over here! Why aren’t you looking at me?” [laughter]

I was thinking about mashups - there’s also a page visibility API, so you get a callback or something, whenever your page is no longer visible… So I could just imagine, as soon as you tab away, your device vibrates…

Ooh…

That’s an abusive relationship right there.

Right.

You could also imagine – well, I wondered, do laptops have a vibration mechanism, or is it just phones?

Because you can imagine - you have your laptop, and if you hover over something in particular, instead of it shaking visibly, your laptop buzzes…

That would be pretty funny.

Not yet…

Here’s one that was probably a waste of time - the Force Touch API, which is an Apple-specific hardware feature, which they appear to be moving away from, because they’re now going long-press on everything, and force touch is not longer en vogue… So all the effort into that – I mean, it’s still gonna work on devices that do that, but it’s such a small niche, and it’s gonna be a vanishing one over the years… So maybe efforts could have been put elsewhere. But you can detect force pressure.

Yeah, it sounds ripe for abuse.

Yeah, it does. That’s another one that’s in origin trial.

Do you remember those old apps (was it like Plexo?) where they would ask you and just pull and scrape all of the people in your email contacts, and things like that? Yeah, this sounds like viral abuse waiting to happen.

When would you like that feature? If you’re gonna email somebody from inside an app, and it can get your contacts and they’ll autofill their email address? Like, if it’s a Gmail-style web app?

I wonder if that allowed you to run a sync between an OS and a Gmail-style app. So you could just sync my contacts between Gmail and my native iCloud, or whatever… I guess Android already has that, if you’re on Google. But that would be cool, because I definitely have a disparity between multiple Gmail accounts, plus my iCloud contacts. People exist in some places and not in others, and it’d be cool to have a sync.

[laughs]

It’s relationship management, yeah.

It’s a formalization that not everybody is worthy of.

I’d actually really like something that made it easy to clean up my contacts…

My contacts have gotten copied from phone, to phone, to phone, and at some point along the way they got duplicated… So I have like five versions of a bunch of people, but –

But there’s hundreds of them. So I’m not gonna manage those and remove those by hand, in my phone. If I had a spreadsheet API or something, like “Pull these into a spreadsheet, clean them up and dump them back out…”

Exactly, remove dupes. That would be awesome. So yeah, I could see a really powerful, wonderful use case for that. In fact, I wonder – I could maybe even write that. Like, what is this supported in? This gives me an idea for a hack project that would help my phone.

There you go. So we go from like “This is lame, ripe for abuse” to like “I’m actually coding for this right now.” I like that.

Yeah. Only supported for origin trial and Chrome for Android, so… Cannot work for right now.

It looks like it’s only read.

Well, you could read them all out into a sheet, and then manage them, and then you could do an import some other way.

And then I’d get another set of duplicates, right?

You’d clear out your Android – you could do this without a web browser.

According to Chris in the chat, Google does remove dupes for Contacts. I know there’s a Merge button inside of Apple Contacts as well.

Is there?

[01:00:01.27] Kball, there’s solutions out there for you.

So I’m in Apple…

It sounds like the kind of domain that has probably 101 open source contact mergers out there; it just requires a little bit of effort to go find a good one and use it. I think we can get you there, Kball. We can get you fixed up.

Okay. Yeah, there’s one person in my phone ten times… I don’t know how that happened.

Ten times? You must really like them.

I mean, it’s not even somebody I interact with very often.

Well, do it Divya - wipe that person from your life.

They’re gone. They’re not worthy of Contacts.

From what I can tell, I have to – to delete them, I have to click into them, then I have to go into Edit, I have to do another… It’s three clicks per duplicate to delete it. Oh, and then I have to confirm that I want to delete it, and now I’m down to nine instead of ten.

Is this an Android phone?

No, it’s an iOS phone.

Oh, it’s an iOS one. So pop it open on your Mac, where it’s gonna be less clicks. Or iCloud… Oh yeah, you’re on like a six-versions-old Mac; not gonna work. Upgrade, baby. Catalina, here comes Kball.

Uh-oh…

Alright, any other APIs we would like to discuss before we call it a day?

I just wanted to bring up the potential mash-up of the native file system access and WebVR, so that you can just start walking through a 3D exploration of your computer…

Oh, like on Jurassic Park, when she’s navigating the UNIX system.

I think that would be Sci-Fi come to life, right?

I like it.

Yeah…!

Well, you know, Google has this stuff where it’ll take photos and it’ll automatically automate them…

Yeah. The next step is do that into AR, so take these people and make them 3D, and automatically–

Yeah. The future is coming. It’s like winter - winter is coming.

The future is coming indeed. Any other mash-ups before we stop?

[01:02:37.20] I mean, I mentioned my vibration plus page visibility mash-up, right?

Yeah.

Anything else we can add into that to make it even more ridiculous…

Well, you could have the Web Speech API… So when you page away, it vibrates and says “Hey! What are you doing?! Don’t forget about me!”

[laughs]

Wouldn’t that drain your battery faster?

I’d be happy with just an accurate battery reading. Right now it’s like “I’m at 12%” and then it just turns off. If you could just get the percentages to work, I would be a happy camper.

How about the combo of the native file access and the WebUSB, so you can automatically steal stuff from your computer? [unintelligible 01:03:33.06] USBs can already run executable code, so that might not be useful, but I was just thinking… Like, “Plug in your device, go to this website” and it proceeds to dig through and copy everything.

Some of these features and ideas remind me of the stoner character that Ben Stiller played on – what’s the stoner movie? Half Baked. He would say “Have you ever looked at a $20 bill? Have you ever looked at a $20 on weed?” That was his whole shtick. And a lot of this just seems like “Hey, have you ever used USB? Have you ever used it in your browser?” [laughter] It’s almost just like “Well, I have a computer. I can just use it in my computer. Why do I have to send it through a browser?”

I feel like maybe there’s a little bit of a browser obsession that we have.

I do think the WebUSB thing is super-cool for updating firmware. We talked about that on some other episode.

Yeah, and I know Suz was very excited about this.

There’s some really cool things that it enables… But there’s also some potential for abuse.

Alright, that’s JS Party for this week. If you’d like us to discuss a specific topic, or to talk to a specific guest, or if you like certain panelists more than others, and you’re like “Hey, bring us more Divya”, you can hit us up at Changelog.com/request, select JS Party in the dropdown and let us know what you’d like to hear on the podcast. We’d be happy to oblige.

That’s all we’ve got for you this time. Hey, get out there and use some APIs. Use them wisely. Use them in your browsers.

If you’re ever looking for inspiration, go to Feross’ website. [laughter] TheMostAnnoyingWebsite.com, I think.

If you learn one thing from today’s show, check out the Web Coffee API.

Ooh! Divya wins the pod! I love that.