1. Twitter's Sara Haider

WEBVTT
1
00:00:00.140 --> 00:00:12.300
Geek Out with Matt Navarra is sponsored by Pinterest. Inspire your audience and grow your company on Pinterest with a free business account. Learn more at business.pinterest.com. It's time to Geek Out.
2
00:00:12.720 --> 00:00:26.340
[upbeat music] So it feels like ages ago since we first teased the idea of working on a podcast, but here we are.
3
00:00:26.540 --> 00:00:35.100
Finally, we're on air, uh, so to speak, and, uh, hello, and welcome to Geek Out with Matt Navarra, sponsored by Pinterest. I am Matt Navarra. And I'm Martin SFP Bryant.
4
00:00:35.260 --> 00:00:42.269
Uh, this is a new podcast for tech and social media professionals, geeks, and anyone else interested in what's happening in the worlds of tech and social media.
5
00:00:42.300 --> 00:00:50.040
Every episode, we'll be discussing the week's most important news and chatting to a very special guest. First, though, I think we should probably explain why we're doing this. Yeah.
6
00:00:50.260 --> 00:00:59.600
Well, basically, Martin and I used to work at The Next Web, um, where I was director of social media, and Martin was editor-in-chief. And we thought, hey, you know, this would be a fun idea to do a podcast.
7
00:00:59.680 --> 00:01:10.660
So I started asking around, see if anyone was interested, to find out what special guests might come on the show, and amazingly, we, um, had yeses from everybody we asked, and we had a sponsor interested really early on.
8
00:01:10.720 --> 00:01:20.979
So, uh, here we are. Um, yeah, that's basically it, isn't it? Uh, we've lined up some excellent guests for upcoming episodes, including key names at big tech companies, journalists, even a politician.
9
00:01:21.020 --> 00:01:26.100
So do stick with us. Subscribe, and you'll get a new episode every week. But more on them as we continue through the series.
10
00:01:26.220 --> 00:01:34.960
But if you do wanna get in touch with us on Twitter, uh, use the hashtag #GeekOutQuestions, and you can find me there as MattNavarra. And on Twitter, I'm MartinSFP.
11
00:01:35.360 --> 00:01:41.640
Our guest today is Sara Haider, Director of Product Management at Twitter. So Sara, welcome. Thank you for having me, Matt. Great.
12
00:01:41.680 --> 00:01:46.950
Could you just tell everybody a little bit about, um, what you do and your background before Twitter and now at Twitter?
13
00:01:46.980 --> 00:01:51.930
'Cause there's a lot of people like me who are complete nerds and geeks who are complete fans of yours. But at the same time- [laughs]...
14
00:01:51.940 --> 00:02:01.460
there's a lot of people that have probably no idea, would love to know a bit more about you. Sure. Uh, so I'm in product management at Twitter. I've been in product management for a few years.
15
00:02:02.080 --> 00:02:14.560
Um, come next month, uh, it will have been 10 years since I started at Twitter. Uh, and I actually started as an engineer, um, uh, about a year out of college. I went to the University of Waterloo.
16
00:02:14.660 --> 00:02:30.720
I'm Canadian and, uh, got some strong roots there and, uh, joined as a very junior engineer a year out of college, um, working on internal tools. And I've worked on, uh, several, several teams at Twitter over the years.
17
00:02:30.920 --> 00:02:40.980
Um, and, uh, at some point, I moved into product, and that's what I'm doing now. And, uh, I- and you're now kind of in a position of transition, I guess, from m- from one role to the next.
18
00:02:41.000 --> 00:02:50.080
So that's something that's, uh, exciting for you as well, I would imagine. Uh, yeah. Uh, about to, uh, about to change teams and, and, and tackle some new adventures, so I'm excited about that. Well, great.
19
00:02:50.100 --> 00:02:57.060
Well, we're gonna, um, talk to you a bit more in detail in a few moments. But what we first wanna do is talk a little bit about some of the things that are going on in social media in the last week.
20
00:02:57.160 --> 00:03:02.920
There's been a ton of news that have been plowing through the, uh, Twitter sphere and all the other platforms I gather news from.
21
00:03:03.010 --> 00:03:08.400
So I'll just hand over to Martin to talk a little bit about what's been going on this week. Absolutely, yeah.
22
00:03:08.540 --> 00:03:13.900
So, uh, just to explain the kinds of news we're gonna be covering in this show, uh, since it's a brand new podcast.
23
00:03:14.220 --> 00:03:19.500
Uh, it's basically the kinds of things you'll find in Matt's Facebook group, uh, the Social Media Geek Out.
24
00:03:19.530 --> 00:03:26.980
So if you're a member of that, then it's the kind of things that get discussed in there we'll be discussing in this show. So largely social media, but lots of other tech stuff as well.
25
00:03:27.060 --> 00:03:37.640
Uh, and, uh, the first thing this week is, uh, around, um, Facebook saying it may remove Like counts, um, and at the same time, YouTube phasing out public-facing subscriber counts.
26
00:03:37.760 --> 00:03:50.300
And, uh, so this is, this is an interesting trend we're seeing now where, uh, um, we've, we've had Instagram, they're testing, uh, the removal of, uh, Like counts, uh, publicly, uh, in seven countries at the moment.
27
00:03:50.400 --> 00:03:55.260
And, uh, now their parent company, Facebook, saying they're gonna do, uh, the same thing and, and, and try that out.
28
00:03:55.620 --> 00:04:04.500
Uh, all around the idea of mental health and not being stuck in a, a f- feeling that you're not popular enough because all your friends are getting more likes than you.
29
00:04:04.940 --> 00:04:15.100
And, uh, uh, with this idea, you'll still see how many likes you're getting, but no one else will. And, uh, then the only, uh, popularity contest you're gonna get into is with yourself.
30
00:04:15.440 --> 00:04:24.000
So, uh, if you want to compete with yourself, that's what you can do. But, uh, there'll be less of a feeling of, "Oh, they're more popular than them." Um, Matt, what are your thoughts about this?
31
00:04:24.210 --> 00:04:38.920
Well, you know, this has, uh, been a long time coming, I think, because part of the reason now we're seeing things to do with regulation from different governments around the world to do with social media is this concern over digital wellbeing and, and, uh, and what people's motivations are for, for posting certain types of content.
32
00:04:38.940 --> 00:04:47.460
And, you know, we've had issues around, you know, polarizing content and misinformation and, and some of these, uh, Like counts are a part of the, you know, problem, some think, I think.
33
00:04:47.880 --> 00:04:49.740
Uh, so I think it's a good move that it's being tested.
34
00:04:49.980 --> 00:05:02.120
Um, and we've-- as you were just discussing there, there's a co- couple of, um, places that they're trying to remove these kind of, um, counters of, of vanity metrics, I think is what people tend to call them, to see if that has a positive or negative impact.
35
00:05:02.170 --> 00:05:08.040
And I guess it's not just around positive and negative for the user, but as a business, there's a, there's a question mark over that as well.
36
00:05:08.540 --> 00:05:16.860
Um, and I think from, from, uh, the most uproar I've seen is typically from marketers and influencers who often don't actually understand what the changes are.
37
00:05:16.880 --> 00:05:25.420
That, you know, this isn't a foolhardy removal of all of the, um, metrics, so you can't see them on your post. This is more on the kind of public-facing side.
38
00:05:25.480 --> 00:05:32.700
But, um, um, I don't know if, um, Sara, if you kind of, um, had, um, had thoughts on this at Twitter at all. Is this something that you've kind of been following with any interest at all?
39
00:05:34.280 --> 00:05:44.500
Yeah, that's a really good question. Um, in our prototype app, um, TWttr, we've been testing new layouts for, uh, conversations on Twitter.
40
00:05:44.990 --> 00:05:56.200
And one of the things that we have been trying to learn about is what if we put the engagement actions, so the Like and the Retweet button, behind a tab, uh, along with the counts kind of behind a tab.
41
00:05:56.720 --> 00:06:06.360
Um, and, and, and testing out, uh, what we can learn from an experience like that. And the motivation for us behind that is really around can we let the content stand first?
42
00:06:06.760 --> 00:06:12.880
Can we clean up the UI to make it easier to read? And can we, can we encourage people...
43
00:06:13.280 --> 00:06:22.040
Even though it's a subtle change, can we encourage people to read the content first before looking at the engagement counts to make a judgment about the quality of that content?
44
00:06:22.120 --> 00:06:34.824
And I think it's been really interesting because- It is a change, and a lot of people, uh, you know, we've heard from a lot of people that, uh, they really miss those counts and that it is, it is an extra tap to get to them and that, that can be challenging.
45
00:06:34.844 --> 00:06:41.943
But we've also heard from people that they're finding that they're actually reading the content more. And this is something we wanna dig into.
46
00:06:41.974 --> 00:06:54.834
And I don't know if this is the exact right solution to get to what we're looking for, and we are, you know, testing out, uh, various, uh, um, options here. But I think something around
47
00:06:55.824 --> 00:07:08.784
l- encouraging people to read the content, not, um, not having immediate reactions based on, uh, based on those counts is something that we're really, really interested in. Um, and, you know, I think you've mentioned,
48
00:07:09.684 --> 00:07:22.584
you know, what do, uh, you know, how, how people are thinking about this, what do those numbers do for, uh, the competition and, and your psyche and, and, and vanity metrics and whatever, and that's something that we're thinking about as well.
49
00:07:23.144 --> 00:07:36.043
Uh, but we're kind of taking a, a, a comprehension angle to it, um, in our prototype app at the moment and, and trying to figure out, uh, if there is something that we can do to make the content really stand first.
50
00:07:37.344 --> 00:07:45.004
If Twitter totally got rid of, um, uh, any kind of retweet counts and, uh, like counts publicly, uh, it would end ratio shaming.
51
00:07:45.364 --> 00:07:59.144
Uh, the, uh, the idea of, uh, where people, um, uh, talk about the ratio of someone who's shared a particularly, uh, controversial or maybe, uh, bad opinion and they've got 1,000 replies and four retweets and five likes and things.
52
00:07:59.504 --> 00:08:11.524
Um, maybe a mixed blessing, uh, for some people who like shaming people on their, uh, on their ratios. But, uh, uh, it might make, might, might make things a bit more, uh, calm and pleasant and less confrontational.
53
00:08:11.644 --> 00:08:20.614
I don't know. But, uh, yeah, interesting stuff. I think this is why our prototype app, which we call internally, we call it Little T. It starts with a, a little T, T-W-T-T-R.
54
00:08:21.104 --> 00:08:31.904
Um, I think what's one of the great things is that we can test some of these things and take people's feedback into account and, and understand how, uh, the changes to these features feel before we bring them into the main app.
55
00:08:32.484 --> 00:08:44.604
Um, I think, you know, there are a lot of, uh, lot of different use cases and scenarios in which you might get a lot more replies than likes. One is kind of the ratio that you were talking about.
56
00:08:44.613 --> 00:08:55.204
Another one is if you ask a question. If you ask a question, maybe it's what's your favorite sushi place or, you know, um, should I watch this TV show or whatever, uh, you'll see a lot more replies than likes.
57
00:08:55.224 --> 00:09:05.784
And we're really trying to understand, um, what are the patterns that we see for certain kinds of content, um, and especially for, um, really engaging and interesting content.
58
00:09:05.884 --> 00:09:12.113
Uh, you know, what are the patterns that we can see, uh, around engagement for those conversations?
59
00:09:12.154 --> 00:09:19.544
And so it's not always as simple as like, oh, this is, you know, it's got more replies than likes and therefore it's a, it's a ratio tweet. Um, so some- just something to think about.
60
00:09:20.304 --> 00:09:27.244
We'd have to make this podcast about two hours long because I could literally spend about two hours just discussing that topic on its own. So, uh, this stuff fascinates me.
61
00:09:27.273 --> 00:09:34.544
[laughs] I hope it is interesting to everybody else. But, um, but what, what else is happening on this week as well, Martin? There's a couple other stories we've been picking up on. Yeah.
62
00:09:34.624 --> 00:09:45.444
So, um, been interesting, um, viral deep fake app out of China. Um, uh, now we Brits, Matt, would call it ZAO. Uh, Americans would probably call it Z-A-O.
63
00:09:45.884 --> 00:09:55.164
Uh, but, um, uh, it's, uh, it's, uh, been, uh, very popular apparently in, uh, in China. Um, although, um, but it-- there's been a bit of controversy around it as well.
64
00:09:55.244 --> 00:10:10.084
Basically, the idea is it, uh, supposedly deep fakes you into, uh, popular films. So you can put your face on you in, uh, Titanic. You know, suddenly you're in, in a big scene out of Titanic, for example.
65
00:10:10.184 --> 00:10:11.164
So, uh, this is the idea.
66
00:10:11.184 --> 00:10:28.884
You know, deep fakes, uh, people have probably read quite a bit about them, uh, and a panic around them, around the idea that, uh, you can, uh, make, uh, I don't know, you could make Donald Trump, uh, say something that he's never said and, uh, put that up as a, a, a video and, uh, cause a massive storm and of controversy.
67
00:10:29.264 --> 00:10:39.864
Uh, that kind of thing. And obviously the idea of, uh, just normal people being deep faked into, uh, being put in situations they were never in, that kind of thing. Uh, there are concerns around it.
68
00:10:39.884 --> 00:10:47.504
But, uh, but this app, um, mainly people have been, uh, from, from the report, uh, from, uh, a site called, um, uh, Radii China.
69
00:10:47.924 --> 00:11:00.884
Uh, basically people have been, uh, worried about, um, this company behind, uh, the app, uh, Momo, uh, them cr-- basically collecting lots of face data and what they might do with it, and might it end up with the Chinese government, that kind of thing.
70
00:11:00.964 --> 00:11:10.444
So perhaps some legitimate concern there, but, uh, also, uh, concern around terms and conditions and terms of service, uh, on, uh, uh, on that particular app.
71
00:11:10.584 --> 00:11:14.944
But to me, that seems exactly the same concerns we see on all these apps.
72
00:11:14.984 --> 00:11:23.284
People complaining about, "Oh, it says, uh, they, they can make copies, uh, as many copies as they like of my, uh, photos and put them in any country they like."
73
00:11:23.564 --> 00:11:30.784
You know, it doesn't mean that the CEO is going to be pinning them up in, uh, his holiday home or something. It's [laughs] so they can move them around to servers and things.
74
00:11:30.844 --> 00:11:36.364
Uh, so for me, um, I mean, the company has clarified a lot of, uh, stuff about how they'll use, uh, the data they collect.
75
00:11:36.764 --> 00:11:43.964
For me, I think it's more about, uh, making sure that companies are really clear about how they use data. Um, what, what were your thoughts about this story, Matt?
76
00:11:44.224 --> 00:12:02.163
Uh, well, th-this reminds me a lot of, uh, all the stories that keep-- seem to be coming out at the moment regarding social networks in terms of there's a kind of an instant expectation or assumption that whatever any of these-- some of these apps are doing, I know this isn't a social network per se for this particular app, but whatever they do, it, it's always with a bad intention.
77
00:12:02.204 --> 00:12:09.304
It's always with bad motives. It's something that's kind of, uh, they've got kind of a sneaky, crafty thing that they're trying to do that's not in your interest.
78
00:12:09.624 --> 00:12:17.614
And so I kind of try and always see the positive in, in most things, and that's probably, uh, not always to my benefit or to everyone's benefit to, to see things in that way.
79
00:12:18.184 --> 00:12:25.004
Um, so I'm, I'm-- I don't like the fact that we're so cynical now and we're, we're almost expecting it to be a negative thing and it's not a, a positive thing.
80
00:12:25.064 --> 00:12:34.684
You know, I, I like the idea of some of these AI and deep fake stuff, the technology behind it and how that could be used in, in more useful and, and, um, helpful ways.
81
00:12:34.984 --> 00:12:46.424
But at the same time, you know, we know that people will inevitably abuse that technology. So I, I'm hopeful that this isn't an example of, of a, of a, a new technology that's go- being abused for, for, for bad things.
82
00:12:46.564 --> 00:12:47.654
But, uh, we, we don't know.
83
00:12:47.744 --> 00:12:59.764
But I, I guess this is an- just yet another thing for someone like Twitter to, to be thinking about and worrying about in terms of content that goes out there that looks even more plausible and could lead to additional concerns about fake news and things.
84
00:12:59.804 --> 00:13:02.824
Is that, is that something that gives you any sleepless nights at, in Twitter, Sarah?
85
00:13:04.388 --> 00:13:13.908
Yeah, the, the app that you're talking about specifically is definitely very interesting and a, a new, a new take on, um, on what's possible, uh, in this space.
86
00:13:13.948 --> 00:13:25.188
And, um, I think this is something that, uh, you know, a lot of, a lot of folks are, are thinking about, and we're certainly keeping an eye on, uh, to see how it progresses and how these things are used.
87
00:13:25.248 --> 00:13:44.628
And look, if we see the use of these, these deepfakes to spread misinformation that are-- that is in violation of our policies, whether that's governing election integrity, integrity or targeted attempts to harass or abuse anyone or any of the other Twitter rules, we'll take enforcement action against that account.
88
00:13:44.668 --> 00:14:09.008
And what we've seen is that, in general, attempts to execute misinformation campaigns that could include these kinds of, um, synthetic media or other tactics, uh, they tend to use other, uh, tactics that are against our rules today, such as account manip-account manipulation, automation, spam, et cetera.
89
00:14:09.168 --> 00:14:18.278
So definitely something we're keeping an eye on. There's been another story this week as well about Twitter. Martin, what has it been all about? Yeah. Um, yeah, Sarah, we won't ask you about this one.
90
00:14:18.288 --> 00:14:23.248
We'll let you sit this one out 'cause I'm sure you don't want to comment on your, uh, CEO security arrangements, but, uh, we should mention it.
91
00:14:23.388 --> 00:14:29.188
Jack Dorsey's Twitter account was hacked and, uh, it was spewing some, uh, rather distasteful things.
92
00:14:29.588 --> 00:14:52.568
Um, yeah, it, it, it-- for me, it was less a story about Twitter and more a story about personal security and, uh, how we, uh, look after our data and how, um, our phone numbers, particularly because this was a, um, uh, uh, this was, uh, related to the security of, uh, Jack Dorsey's, uh, phone number as opposed to, uh, his Twitter account.
93
00:14:52.608 --> 00:15:05.167
And so, um, it really makes you think about, um, h-how many weak points there might be in your security arrangements and that you might not be the CEO of two public companies, but you still wouldn't want something like that happen to your account.
94
00:15:05.268 --> 00:15:24.528
Uh, so thinking about things like having a, um, an authenticator app or a physical security key for, for, for logging into things like Twitter, um, i-it's probably preferable to, uh, connecting up through, uh, SMS and not having, um, your phone connected to too many services, all that kind of thing.
95
00:15:24.768 --> 00:15:29.148
Matt, what are your thoughts? Yeah, I think I agree with much of what you were saying there and, and a lot of the reporting.
96
00:15:29.188 --> 00:15:38.048
You know, a lot of the reporting was focusing around, you know, the ways that your accounts can, um, can be attacked by, um, those bad actors that, that want to do it.
97
00:15:38.118 --> 00:15:50.368
And so, uh, you know, I always forget that, that, you know, the SMS, the two-factor auth stuff there isn't as, uh, you know, uh, safe as sometimes maybe you assume it might be, and that it's better to use these authentication apps.
98
00:15:50.788 --> 00:16:00.328
Um, I think obviously from a, from a Twitter point of view that, you know, the, the obvious headline-grabbing attention thing was this is the CEO of Twitter, and, and if their account isn't safe, whose account is safe?
99
00:16:00.708 --> 00:16:09.848
And, you know, I, I think that, that, that's an interesting make for a headline, but I, I don't think that, that for me, that was the most exciting talking point. Neither was the, the content that was put out.
100
00:16:09.928 --> 00:16:20.428
You know, the second I was on TweetDeck, and I saw a tweet go out, and it was saying some of the stuff that these tweets said, you just knew, or anyone with any kind of interest in this stuff would know this isn't him tweeting.
101
00:16:20.488 --> 00:16:27.728
This is somebody that's taken over the account. And of course, there's questions over account security and things, but, um, it was an interesting story.
102
00:16:27.768 --> 00:16:35.868
And I think today there was actually a story about Jack saying that he hasn't actually got a phone number right now after, after what's gone on, and I guess he's re-reviewing how, how he tackles that.
103
00:16:35.938 --> 00:16:40.888
So, um, I think it's, uh, it was an interesting story, let's put it that way. Yeah, absolutely.
104
00:16:40.928 --> 00:16:49.938
And, uh, we, we should just say that, um, uh, in response to this, uh, uh, Twitter, uh, posted that, uh, the phone number associated with the account was compromised, uh, due to a security oversight with the mobile provider.
105
00:16:50.228 --> 00:16:56.468
This allowed an unauthorized person to compose and send tweets via text message from the phone number. That issue is now resolved.
106
00:16:56.748 --> 00:17:03.088
So obviously, if you have ever, um, tweeted via text message, um, and you're not doing that anymore... Um, I used to do that.
107
00:17:03.168 --> 00:17:17.368
Uh, actually, I used to use that service years ago when I was in Japan, um, in the middle of nowhere, and I could tweet, and I couldn't read any-anything that, um, anyone might, anyone might reply to me, but it meant I could tweet from like the, the remote Japanese countryside.
108
00:17:17.428 --> 00:17:30.568
Uh, so it can be useful. But, uh, but yeah, um, uh, having a full security audit of, uh, uh, how different ways your accounts can be compromised and thinking about that, uh, sensibly is, is always important.
109
00:17:31.058 --> 00:17:41.408
And just some breaking news coming through as we record relating to that story. Twitter support has said, "We're temporarily turning off the ability to tweet via SMS or text message to protect people's accounts.
110
00:17:41.748 --> 00:17:49.808
We're taking this step because of vulnerabilities that need to be addressed by mobile carriers and our reliance on having a linked phone number for two-factor authentication.
111
00:17:50.308 --> 00:17:56.108
We're working on improving this," say Twitter. So there you go. Good stuff. Well, that's-- Yeah, I think it, it's some valid points in there.
112
00:17:56.168 --> 00:18:03.048
Hopefully, uh, uh, there won't be another incident with, uh, any other executives from social media companies 'cause that's, that's not good for, for news headlines for them.
113
00:18:03.388 --> 00:18:09.408
We're gonna speak in more depth with Sarah in just a moment, but first, Geek Out with Matt Navarra is sponsored by Pinterest.
114
00:18:09.728 --> 00:18:20.188
Now, you might already have your own impressions of what Pinterest is, but it's worth understanding the platform properly. Basically, Pinterest is a visual discovery engine. It's where you find and do what you love.
115
00:18:20.628 --> 00:18:28.988
It's not really a social media network. It's more like personal media. And with Pinterest, you can discover useful and relevant things that inspire you to do stuff.
116
00:18:29.248 --> 00:18:36.088
So people use it to discover ideas and inspiration in daily life. Things like, where do I want to go on holiday, and what should I cook for dinner?
117
00:18:36.568 --> 00:18:44.448
Pinterest is all about discovering things, even if you don't know what you're looking for. It provides an online experience that makes your offline life more interesting and rewarding.
118
00:18:44.978 --> 00:18:56.428
The mindset for users and businesses on Pinterest is much different than your usual typical social media. Three hundred million users now come to Pinterest each month to find the inspiration to try or buy new things.
119
00:18:56.828 --> 00:19:07.356
That could be anything, a recipe, a haircut, how to style a leather jacket for autumn, or finding inspiration to plan a new city break. And timing is everything. People come to Pinterest with a planning mindset.
120
00:19:07.416 --> 00:19:16.355
They are looking for the future and not focusing on what's just happened or what they've just eaten. Because of that, visual content from businesses is inspiring, it's actionable, and it's timely.
121
00:19:17.036 --> 00:19:24.206
And brands on Pinterest aren't interrupting the user experience, they add to the experience by providing inspirational and useful content.
122
00:19:24.596 --> 00:19:35.456
In fact, most of the content on Pinterest comes from creators such as brands, publishers, and content creators like influencers. So inspire your audience and grow your company on Pinterest with a free business account.
123
00:19:35.476 --> 00:19:37.876
Learn more at business.pinterest.com.
124
00:19:38.816 --> 00:19:50.796
So Sarah, we've got a couple of questions for you, and one of the first things that everybody who spoke to me this week who said, "Oh, you're speaking to someone at Twitter that's senior, you've got to ask them this question", and I bet you can a- guess what question that is.
125
00:19:50.836 --> 00:20:00.016
Any, any ideas what they wanted me to talk to you about? Is it about, I don't know, maybe editing tweets? [laughs] There you go. You know exactly what the score is.
126
00:20:00.056 --> 00:20:09.676
So I, this is a question that you have been asked, and others have been asked at Twitter so many times, and it'd be really boring for me to now just ask you the same question again about is there gonna be an edit button.
127
00:20:09.756 --> 00:20:18.116
So I guess, um, what I wanted to know was there's clearly mixed opinions externally. There's also mixed opinions, I'm guessing, internally at Twitter.
128
00:20:18.156 --> 00:20:21.716
There must be people there that think it's a good idea, people that think it's a bad idea.
129
00:20:21.976 --> 00:20:32.336
Uh, could you just tell them, the audience that are listening to this, a little bit about what, what the dilemma is for creating an edit button and what, you know, issues it could cause as well as the benefits it could give?
130
00:20:32.376 --> 00:20:38.236
And then, and I'd just love to know a bit more about what your opinion is personally about what you think about an edit button. Sure.
131
00:20:38.496 --> 00:21:00.836
So I'll start by saying that I think everyone internally agrees that if, if we were to implement an edit feature, we would have to do it in a responsible way because the potential for manipulation, you know, having someone retweet something that you edit to say something else, uh, at some later time is, uh, you know, a vector that we would want to avoid.
132
00:21:00.896 --> 00:21:12.976
So there's definitely some tricky things here that we would want to consider and be very thoughtful about, and it's, uh, as you can probably imagine, not as simple as, uh, you know, as, as, as may meets the eye.
133
00:21:13.356 --> 00:21:17.136
One of the powerful things about Twitter is that stuff can travel very, very quickly.
134
00:21:17.716 --> 00:21:26.216
Um, and so we wanna be, um, careful that, you know, the people that are helping amplify certain messages know what they're, what they're amplifying. Um,
135
00:21:27.376 --> 00:21:45.576
you know, having said that, we want it to be easy for people to fix typos because I think the majority use case here is for people to, uh, you know, add in that extra word that they forgot or, or that apostrophe or, you know, change the spelling of something, um, that they spelled incorrectly, and, and we totally get that.
136
00:21:46.076 --> 00:21:59.436
Um, and there are other things that we could do to maybe help you, uh, compose your tweets, um, in a way that, uh, uh, reduces the chance that you've made a mistake. Um, so that's something we're thinking about as well.
137
00:21:59.516 --> 00:22:12.436
But just something to keep in mind is that, uh, tweets have always been, uh, you know, immutable, which is a, maybe a technical term, but, uh, uneditable, and, uh, we syndicate tweets to a lot of places.
138
00:22:12.496 --> 00:22:25.416
It's not just to the apps. You know, we syndicate tweets, uh, all over the web, we syndicate them over SMS, and we'd have to be very thoughtful about how we would issue corrections or changes, uh, should they happen.
139
00:22:25.756 --> 00:22:35.916
That's all to say that it's, it's, it's a, it's a significant amount of effort, and it's not that we can't do it. Um, but it's just not a top priority right now given, uh, all of the other things on our plate.
140
00:22:36.396 --> 00:22:49.875
Uh, uh, when you, when you take a, a, um, when you have a think about the effort versus the value, um, and, you know, from a product perspective, those are the kinds of trade-offs that we think about every day.
141
00:22:50.016 --> 00:23:01.996
Um, personally, I am not anti-edit. Um, I think it would be a useful feature, but I definitely agree with the, uh, prioritization at this precise moment, which is, uh, it's not at the top of our list.
142
00:23:02.036 --> 00:23:05.536
And has, has there been a, a version of an edit button that has existed internally?
143
00:23:05.636 --> 00:23:15.796
Uh, 'cause my understanding from, from speaking to a few people over the last few years is that it has been dogfooded, it's been internally tested in some capacity, and you've played around with it, and it's never seen the light of day.
144
00:23:16.116 --> 00:23:24.356
Is that a, a, a, something you can say has happened and you've played around with a few, few different options of how it might look, or is that something that has not got to that stage of prototyping it or?
145
00:23:25.236 --> 00:23:33.316
I don't know exactly what you might be referring to, but, um, we've definitely thought about what the user experience might look like, um, and what might make sense there.
146
00:23:33.376 --> 00:23:40.516
But that doesn't take into account the significant effort on changing the system to allow editable text.
147
00:23:41.136 --> 00:23:52.056
Uh, so we-- You know, for all of our kind of consumer-facing and end user customer-facing features, we often think about the, uh, consumer experience first, so that's not unusual.
148
00:23:52.156 --> 00:24:02.716
Um, but it, uh, doesn't, uh, necessarily consider the, um, uh, the, some of the edge cases or the effort needed to kind of build everything out in, in the back end.
149
00:24:03.176 --> 00:24:11.176
If nothing else, if you did build an edit button, you'd at least stop people asking you the question, "When are you gonna get an edit button?" So that would be one big pro. [laughs] You know, if [laughs]
150
00:24:11.236 --> 00:24:17.416
if, if that's the biggest complaint I get every day, I can live with that. [laughs] You can live with that. Martin, what were you, you were gonna ask something about Twitter?
151
00:24:17.776 --> 00:24:36.876
Well, um, well, yeah, I was just thinking, uh, actually, um, AI spell checking, you know, like, like in Google Docs, that kind of thing, some kind of, uh, spell checking as you type would probably, um, stop me ever having to use an edit button anyway, 'cause most of the things I have to delete tweets for, um, are little typos.
152
00:24:37.276 --> 00:24:49.204
Um, so that, that's, uh, if, if I'm going to give a, a product request that's not an edit button, it's, uh- In line on, you know, on-the-go spell-checking. Great. I think that's, that's an awesome idea. Um,
153
00:24:50.564 --> 00:25:00.324
I imagine that even if we were to implement that, which would be very useful for you and for me, people will still ask for the edit button. [laughs] So just [laughs] something to be aware of. Yeah, true. Yeah.
154
00:25:00.364 --> 00:25:14.744
Now, um, uh, I've been using the prototype app you mentioned, mentioned earlier, um, for a few months now. Um, I use it as my main, uh, my main, uh, Twitter app. Um, so TWTTR, um, which, uh, I've been calling Twitter.
155
00:25:15.164 --> 00:25:25.404
[laughs] Um, which I... But I, I assumed that wasn't how you pronounced it, uh, yourself, so, uh, it was nice to hear how, how to actually pronounce it. Um- Yeah. We pronounce it...
156
00:25:25.484 --> 00:25:35.204
I mean, TWTTR is pronounced Twitter, but internally, because that can get very confusing, we call it Little T. [laughs] Yeah. Um, so how's it going then?
157
00:25:35.244 --> 00:25:42.044
Obviously, uh, you- you've talked a bit about this, uh, in the past, but, uh, it must be quite, uh, an interesting experience to develop in public.
158
00:25:42.064 --> 00:25:48.434
It's not the usual way of doing things, to really have a conversation with your users, especially, uh, a massive app like Twitter.
159
00:25:48.844 --> 00:25:56.484
Um, to have a conversation with your users about a new interface approach, a new way of doing c- conversations. So what are you learning from, from the app?
160
00:25:56.904 --> 00:26:00.904
Yeah, the experience has been, uh, really interesting and really awesome.
161
00:26:00.964 --> 00:26:15.664
And, uh, as you mentioned, it is a new way, um, f- to kind of develop, uh, in the public and get real, really real-time feedback as well as, you know, the feedback, uh, from folks after they've, uh, sat with the feature for a while.
162
00:26:15.764 --> 00:26:25.934
Um, and w- what we were, you know, what we were thinking behind this is that we want to make some changes to the way conversations fundamentally, uh, work and are laid out on Twitter.
163
00:26:26.464 --> 00:26:38.254
Um, and that's not something that we can just unveil, uh, uh, w- with a, with a big bang, um, on, on, you know, on, on a specific day. In addition to that, uh,
164
00:26:39.244 --> 00:26:47.974
the, uh, effort involved, um, to build out some of these changes, um, can be significant, uh, from an engineering perspective as well.
165
00:26:48.124 --> 00:27:01.583
So what we wanted to do was test out a lot of these changes and iterate on them as quickly as we can so that we can gain some confidence in, um, knowing that we're going in the right direction before we kind of bring everything back to the main app.
166
00:27:01.664 --> 00:27:12.144
What has been awesome and different is, you know, it's, it's scary. It's new, it's scary. Um, we've, we've made, um, we've made a lot of iteration. We've made a lot of change.
167
00:27:12.244 --> 00:27:19.464
Um, you know, not everyone is happy with, uh, everything we've put out, especially at first. And, um, it's, it's...
168
00:27:19.644 --> 00:27:31.244
As a, as a developer and as a, as a product person, it's, it's really being vulnerable, um, in the public. And it's n- you know, there, there are testing programs, there are beta programs, all of that stuff exists.
169
00:27:31.384 --> 00:27:41.164
But what I think has been really unique about this one is that, uh, one, there's no NDA. Anyone who's in the program can tweet their screenshots. They can tweet at us. They can talk to reporters.
170
00:27:41.204 --> 00:27:53.604
They can, you know, do whatever they like. Um, and two, this isn't, um... We're not testing a version of the app that's right before launch just to get the bugs out. That's not the stage that we're at here.
171
00:27:53.644 --> 00:28:04.704
We're testing in the very early versions of these new layouts and, and this new way of thinking about conversations, um, well before we get to the, "Okay, this is what we call beta."
172
00:28:05.184 --> 00:28:15.164
Which is why actually we've been, um, kind of, uh, trying to avoid saying that it's a beta because it's not. A beta implies that you have alpha, beta, production, and this is way before that.
173
00:28:15.344 --> 00:28:28.904
We are still in the kind of figuring out phase of what this is and are we onto the right thing and, you know, doing a lot of user research and a lot of both quant and qual studies to figure out, um, how this is being used.
174
00:28:29.364 --> 00:28:42.904
Um, really doing deep dives in customer feedback and really looking at the conversations that unfold, um, from people that are, um, using Little T, uh, to get the insights that we're, that we're looking for.
175
00:28:43.014 --> 00:28:46.984
So this has been new, fun, exciting, scary.
176
00:28:47.124 --> 00:29:01.104
I think it's been a mindset shift for the people that are working on it, um, to be kind of vulnerable in this way to the public, uh, because the stuff that we put out is not, uh, you know, in a, in a super polished, uh, state that you might expect from the production app.
177
00:29:01.784 --> 00:29:16.504
Um, so it's been, it's been cool. But what this allows us, what it has allowed us to do is be, um, you know, iterate very quickly, um, and learn, uh, faster than we could, uh, doing this in our, in our production app.
178
00:29:16.544 --> 00:29:25.114
So that's been pretty cool. Great. So and then one of the things we wanted to ask you, Phil, as well, is, uh, a bit about the, the people criticizing and, and, and saying they don't like things.
179
00:29:25.144 --> 00:29:36.684
Like, I put out a tweet, uh, a couple of weeks ago when it was one of the latest updates to the, uh, the, the app, and, uh, I showed some de- uh, examples of the screenshots of the changes to the conversation th- uh, view in Threads.
180
00:29:37.104 --> 00:29:43.164
And there was, uh, a lot of people that thought it was great, but there was also a lot of people who were saying, "This is rubbish. It looks awful. And what are they doing?
181
00:29:43.184 --> 00:29:53.604
They're stupid at Twitter, what are they doing?" How does that feel to you as working on, on a project and you're seeing that kind of conversation going on? Yeah, I mean, people, um...
182
00:29:53.664 --> 00:30:05.824
I mean, we listen to all of the feedback. And what I try to do personally is to see what the actual feedback is beyond any of the, the snark or the name-calling or whatever, to try to see if there's anything beyond that.
183
00:30:06.184 --> 00:30:14.604
I also find that if you just respond to people online and just say, "Cool, thanks for your feedback. Like, what did you mean by this?" They tend to be a lot more, a lot more civil. Uh.
184
00:30:15.004 --> 00:30:31.044
But, um, you know, we, uh, we started this venture, um, almost a year ago. I think it was last November, I tweeted a very, very early screenshot that was kind of deliberately bad.
185
00:30:31.324 --> 00:30:43.720
[laughs] Um, and, uh, ripped the Band-Aid off on all of the negative, and all of the nega- negativity that was gonna come with it. And, you know- Change is difficult.
186
00:30:43.840 --> 00:30:50.260
Um, people don't like change, and people can often react very badly or, you know, very passionately to change.
187
00:30:50.300 --> 00:31:08.080
But that's okay because in those comments and in those reactions, there are-- there is actually good information to glean out of that, to actually understand what it is about the change that people don't like, and to also understand that with any change, even if it's the best possible change you can make, there is still gonna be some, some negative reaction.
188
00:31:08.800 --> 00:31:21.400
So last November, I made this tweet. It got, uh, a lot of, uh, positive responses. It got a ton of negative responses. And, you know, from a personal perspective, you just kind of have to, to harden and, and expect that.
189
00:31:21.840 --> 00:31:31.580
And that was new for our team, and that was kind of one of the first things that we did that really kind of pulled the Band-Aid off on, uh, being vulnerable in public this way.
190
00:31:31.660 --> 00:31:56.620
And now looking at it almost a year later, uh, you know, I'm-- I think both myself and the team are way more comfortable when we do actually put out things that we, we are more proud of and we are more polished about, um, that we have the learnings of getting all that negativity out of the way, uh, that we know how to react and how to get the best out of the, the responses that we get.
191
00:31:57.100 --> 00:32:08.500
Um, so it's all been a learning experience. I think it's been, uh, on the whole, very, very positive. It really helps the team and myself understand how to even interpret, um, reactions online.
192
00:32:08.560 --> 00:32:10.319
So it's, it's been pretty cool.
193
00:32:10.360 --> 00:32:22.600
And co-conversations is the first part of the, um, um, of what you're testing 'cause you-- I think it was discussed at the beginning, uh, we're looking at various things and using this app to understand new features or test new features out of, um, um, or new UI, new UX.
194
00:32:22.940 --> 00:32:28.600
Are you able to tell us, uh, what the next thing is if it-- i-if conversations and threads is, is been done and when it gets done?
195
00:32:28.840 --> 00:32:35.240
Is there, is there a, um, roadmap in terms of what th-those things are, or is that yet to be decided or? Uh, yeah.
196
00:32:35.380 --> 00:32:43.420
So to reiterate, uh, Little T and conversations right now are, uh, synonymous together because that's the thing that we've been testing.
197
00:32:43.860 --> 00:32:54.820
Uh, but we do intend to test more things in, in Little T, um, outside of the new conversations layout. Nothing to share on that right now- Oh, nearly... but I am excited for where [chuckles]
198
00:32:55.520 --> 00:33:00.480
where prototype, where prototyping can, um, can lead us in the future. Good stuff. Okay.
199
00:33:00.560 --> 00:33:12.880
Well, um, something else, just changing pace slightly, uh, to talk about, um, the, um, mental health and conditions, uh, um, that people are trying to link and associate with the, the use or excessive use or the continued use of social media.
200
00:33:13.340 --> 00:33:23.440
Part of the reason that regulation is being thought to be brought in for, um, these social platforms is a discussion about the fact that, uh, young and vulnerable people are affected by use of social media.
201
00:33:23.960 --> 00:33:46.860
Um, and there's things that now we've seen with platforms like Facebook and, and Twitter and, and others that are trying to promote, uh, a healthier use of, of social media, to have a healthier relationship digitally, um, and to give them tools and features that, uh, enable people to control their, their online, uh, um, life, I guess, is what you would call it, um, such as removing like counts or such as giving them counters for how much time they've spent online.
202
00:33:47.320 --> 00:33:51.500
Um, I just wondered if, you know, do you think that Twitter is, i-is doing a lot in that space?
203
00:33:51.580 --> 00:33:59.960
Do you feel that sometimes Twitter's accused of maybe moving slowly to tackle some of these product challenges or to not be as reactive to these things as maybe others?
204
00:34:00.100 --> 00:34:08.740
Do you think that's a, a fair criticism, and, and what, what is Twitter doing i-in this space, uh, to address some of these concerns? Um, yeah.
205
00:34:08.780 --> 00:34:18.180
This is a really interesting topic, and I think, um, as, as you probably agree, um, it's kind of new territory for this decade maybe.
206
00:34:18.280 --> 00:34:27.860
Um, you know, if you, if you told me fifteen years ago, twenty years ago that this is something that we'd have to think about, I would've, uh, maybe said it was something out of science fiction.
207
00:34:28.000 --> 00:34:30.760
Um, but this is something we definitely think about.
208
00:34:30.840 --> 00:34:49.060
And, and one of the ways that we think about this at Twitter is that we want Twitter to be a daily utility for you, in which you, uh, learn something, engage with something, and, and kind of try-- We want you to leave the experience, um, being smarter or more knowledgeable than you were before.
209
00:34:49.100 --> 00:35:03.579
And we, you know, in that, in that light, want this to be a daily utility, so we don't actually have, um, um, specific goals or me-- or specific goals around, uh, the time that you spend in the app.
210
00:35:03.920 --> 00:35:13.960
We want you to get value out of the app every day, but we don't need you to spend hours in there, and we wanna be mindful of, um, you know, things that we do that would encourage that.
211
00:35:14.080 --> 00:35:19.880
Um, we want you to get use of it every day, but also not be, um...
212
00:35:20.600 --> 00:35:32.260
We, we also aren't as concerned of, uh, you being a, a monthly user because we don't think that if you're using it once a month that you're getting the, the value that you might, um, that, that you could be getting out of it.
213
00:35:32.610 --> 00:35:44.250
So maybe that sort of, kind of sets the framework of how we're thinking about this. It's not about time spent, um, but it is about being a utility day to day. And then we also are thinking about
214
00:35:45.520 --> 00:36:00.940
controls and, uh, different features, um, with respect to the way the app works to give you more control and to, um, make, uh, your experience on the app better.
215
00:36:01.320 --> 00:36:07.600
For example, one of the things that we're testing in Canada right now is a hide replies feature.
216
00:36:07.800 --> 00:36:22.960
So if you are the original author of a conversation, you're the f- you, you're the person who started the conversation, um, you have the ability to hide replies within that conversation if you don't feel like they were-- are productive or you don't like them or whatever.
217
00:36:23.060 --> 00:36:38.480
Um, and we're, we're trying to balance that with some transparency to make sure that people aren't just outright censoring, um, folks that are participating in the conversation by making those, um, hidden replies, uh, transparently viewable on a different screen.
218
00:36:39.020 --> 00:36:48.820
But it can help if someone's written something kind of nasty, um, underneath your conversation. You don't have to continuously see it every time you go back to that conversation.
219
00:36:48.880 --> 00:37:02.030
So doing things like that to kind of help your experience. Um, another thing is, um, algorithmically Um, we've been working on, um, hiding and down-ranking really, uh,
220
00:37:03.780 --> 00:37:15.480
replies that are less, um, productive or can be considered toxic so that you don't even see them. Um, and we think that those are important things for general mental health and safety as well.
221
00:37:16.060 --> 00:37:28.700
Martin's got itching to ask about something that he's really fed up with or wants to change about Twitter. Martin? [laughs] Yeah. DMs. I, um, use DMs loads, and I've always said...
222
00:37:28.800 --> 00:37:38.500
You know, there's been talk over the years about a standalone DMs app from Twitter, and there's been, uh, reports in the past about perhaps Twitter working on one and then maybe abandoning it or whatever.
223
00:37:38.580 --> 00:37:50.580
Um, I- I've always said that Twitter should have a DM app that's, uh, a standalone messaging app, uh, for private messaging on T- uh, using Twitter accounts. Call it Slide. [laughs] Um, uh, and, uh, and yeah.
224
00:37:50.640 --> 00:37:59.620
It'd be a, it'd be a cool little messaging app, um, that I would u- use loads. Um, so, uh, do you think we'll ever see a DMs app? Is it something that's been worked on in the past? What, what, what's happening there?
225
00:37:59.739 --> 00:38:08.460
This is a good question. Um, we are reinvesting in DMs. Um, and, uh, y- we've, we're, we've been building out a team.
226
00:38:08.980 --> 00:38:17.040
Uh, a lot of them are based in, uh, Boulder, which, you know, I'd love to talk about kind of, uh, the way we've been thinking about distributed, distributed teams and stuff too.
227
00:38:17.140 --> 00:38:27.780
Um, and we've been working on awesome features and also kind of some tableta- st- table stakes things in DMs that we haven't been able to get to in more recent years.
228
00:38:28.140 --> 00:38:41.380
So everything from performance, search, um, uh, a better secondary inbox, and there, there's, there's more to come, uh, uh, on DMs and, and, you know, like I said, we're building out the team, so definitely watch this space.
229
00:38:41.480 --> 00:38:52.960
Um, I, you know, I, I'd be curious to hear from you why you would want a separate app versus just using the main app, and I would be curious to hear from you
230
00:38:54.020 --> 00:39:05.720
what is it about Twitter DMs that makes it, you know, that makes it something that you wanna use versus, um, another messaging app, um, just so I can understand where you're coming from with the question. Go on, Martin.
231
00:39:05.780 --> 00:39:14.549
Sales pitch. Uh, well, for me- Sales pitch. Do your best. Go on. [laughs] Well, well, well, part of it is I just like the idea of a DM app called Slide. [laughs] Uh, I don't think it's a cool name.
232
00:39:14.549 --> 00:39:18.900
But, uh- That's never gonna get through finance... but beyond- They're never gonna go with that. Well, true, yes.
233
00:39:19.180 --> 00:39:34.620
Uh, but, uh, but other than that, um, uh, when I look at the amount of messaging I do, um, uh, there's a lot of it through Facebook Messenger, there's a lot of it through, um, uh, uh, WhatsApp and things, but, um, other than that, a lot of the messaging I do is through Twitter DMs.
234
00:39:34.740 --> 00:39:47.989
And, um, a lot of the time I'm on, uh, you know, TweetDeck, and it's a, it's a column, and it's a bit fiddly or, uh, um, uh, it, it's, it's, uh, I just feel like there'd, there'd be more space to breathe with features, uh, in a, in a, in a standalone app.
235
00:39:48.300 --> 00:39:58.430
So let me just kind of re- uh, re-ask that. Are you using DMs because you're already on Twitter? Or- Uh-...
236
00:39:58.460 --> 00:40:02.950
is there something specific about the people that you're connected to on Twitter, and those are the people you're messaging?
237
00:40:03.340 --> 00:40:14.820
I guess I ask that question because I wonder if that changes if it, it's in a separate app and now, you know. Hmm. Yeah, yeah. So for me, I suppose it's, it's around the idea of, uh...
238
00:40:14.860 --> 00:40:18.760
The, the people I'm connected to, with on Twitter are generally a broad professional network.
239
00:40:19.120 --> 00:40:31.820
So people I work with or, um, or talk to a lot professionally, but also the kind of people who are maybe people I messaged five years ago, and suddenly I want to just check up on, uh, about something with them.
240
00:40:32.040 --> 00:40:41.860
That kind of thing. A lot of, you know, tech investors and people like that who I might be speaking to occasionally. It's just a great way of, uh, of pinging them 'cause they've, they're obviously all in Twitter.
241
00:40:41.940 --> 00:40:51.240
And so for me, it's just like a, the best messaging app out there in terms of the network I have available at my fingertips. I'm glad to hear you say that.
242
00:40:51.840 --> 00:41:03.640
Um, I'll definitely let the DM team know that we've got a fan here. Um- [laughs] Yeah, I mean, you know, uh, no plans right now for a separate app, but we'll... I, I'll definitely take that feedback back to the team.
243
00:41:04.200 --> 00:41:04.720
Um, sorry.
244
00:41:04.880 --> 00:41:15.689
We should just say as well, you mentioned, um, wanting to talk about distributed teams because this is something that's absolutely huge at the moment, uh, with tech companies, um, wanting to expand beyond the, the, the increasingly expensive tech hubs.
245
00:41:16.179 --> 00:41:28.760
Um, and so you've got companies like Automatic that are in, in, i- entirely distributed. So, uh, so yeah, how, how is Twitter looking at, um, expanding its, uh, distributed network of stuff?
246
00:41:29.440 --> 00:41:40.420
So I think it- we're at an exciting time at Twitter in that we are, um, more and more distributed, uh, than before, and I think there's still a long way to go.
247
00:41:40.980 --> 00:41:53.200
Um, and there's still a lot of challenges that we're working through. But we have kind of reinvested in, um, uh, encouraging, um, d- distribution and, and, and remote work.
248
00:41:53.660 --> 00:42:01.400
Uh, like I mentioned, you know, we've, we've been building out this, this DMs team in Boulder. We also have teams in, uh, New York, Boston, Seattle, London.
249
00:42:01.640 --> 00:42:13.980
Um, and, uh, we are getting better every, every day, every, every week, every quarter. We are getting better at how we work together remotely. Uh, so one example of the,
250
00:42:15.080 --> 00:42:28.420
um, some of the processes and practices that we've, um, adopted is, uh, what we call the, the silent meeting, and what this involves is often, um, often at the beginning of the meeting, there's a, there's a, there's a product of work of some kind.
251
00:42:28.480 --> 00:42:38.028
Maybe it's a document. Maybe it's a deck. And we all together, um, we read the, uh, material- Silently and leave comments in the doc.
252
00:42:38.328 --> 00:42:47.848
Um, and then after that period of silent reading, we talk about the, um, uh, the important comments or the ones that, uh, are still kind of outstanding, um, uh, out loud.
253
00:42:48.338 --> 00:42:57.228
And one of the things that is really cool about this is because some of the, some of the meatier parts of the conversation, they happen synchronously but not out loud.
254
00:42:57.288 --> 00:43:07.738
So you're not dealing with, um, you know, AV issues or like, "I'm sorry, you're not close enough to the mic," or whatever the other things that can happen in a, you know, meeting with remote people.
255
00:43:08.568 --> 00:43:15.208
And it also helps people who maybe, you know, remotely don't know the right time to jump in. Um, and not just remotely.
256
00:43:15.308 --> 00:43:29.648
Sometimes people sitting in the same room with you don't know the right time to jump in, and this gives them an opportunity to have their opinions heard or, or, or make-- ask their question, um, in a way that's kind of a combination of synchronous and asynchronous that is the best of both worlds.
257
00:43:30.248 --> 00:43:45.908
Uh, and we've taken this concept in a lot of the meetings that we have at Twitter now, and w- I find that it's very effective, uh, especially getting to the, the root of, or the kind of the meat of, uh, what we need to discuss in a meeting.
258
00:43:46.148 --> 00:43:46.158
Yeah.
259
00:43:46.158 --> 00:43:56.208
And would you get to the point where you're hiring, say, a, a really talented developer who happens to live in a remote part of Scotland or something to work on this team you're talking about in Boulder, for example?
260
00:43:56.608 --> 00:44:06.728
Um, because a, a lot of companies now are, uh, are hiring around the world wherever the talent is, rather than just saying, "Oh, we're opening an office in London" or, or wherever. Yes. And so we're doing both.
261
00:44:06.848 --> 00:44:18.618
And, um, I work with a lot of people who work out of their homes or out of other, uh, kind of working spaces in areas that we don't have offices. Um, and it, it's, it's working. It's working really well.
262
00:44:19.968 --> 00:44:26.808
One of the questions that I de-desperately need to get in here, otherwise, if I don't get this in by someone else who's asking me to ask, is gonna-- I'd go nuts.
263
00:44:26.948 --> 00:44:35.228
But I am asked by people on Twitter, on Facebook, in any, any messaging app they can get ahold of me on is, "Matt, can you get me verified?"
264
00:44:35.478 --> 00:44:40.948
And I'm sure it's-- And I've spoken to a lot of Twitter staff, and they've said to me over the years, "Yeah, that's what my friends and family say as well."
265
00:44:41.288 --> 00:44:53.208
And one of the things that we heard about verification was, um, a, a little while back now, that there was a, a, a big process of review for the whole verification setup in terms of how it would be, how people would apply, what, what the meaning of it was behind it.
266
00:44:53.568 --> 00:45:01.788
Uh, I think there was a job, um, that was posted from Twitter, uh, I don't know if it was earlier this year or end of last year, for somebody to kind of head up that team.
267
00:45:01.888 --> 00:45:12.088
Um, as far as I'm aware, verification as a process is mostly closed except for exceptional circumstances, which, you know, you understand that that happened with, you know, big politicians or whatever else.
268
00:45:12.448 --> 00:45:24.148
I just wondered if you could give us any insight into where things are with Twitter and verification and, and what the current state of play is. Sure. So as you mentioned, um, verification is on pause.
269
00:45:24.268 --> 00:45:30.198
Um, let me just kind of revisit the history and intention behind verification.
270
00:45:30.808 --> 00:45:47.988
So originally back in, I don't know what year, more than 10 years ago, before, before I joined Twitter, um, the intention of verification was to help people figure out, uh, you know, official accounts versus, uh, impersonators or parodies or, or what have you.
271
00:45:48.628 --> 00:46:05.298
And what happened over the years is that people started to take verification not only as authenticity of the, um, the account but also as a, um, as a form of endorsement, uh, from Twitter.
272
00:46:05.808 --> 00:46:17.408
And that, uh, is-- was not the intention. But also, uh, just saying that, saying that, "Oh, like verification is not endorsement," is also not enough if that's the, you know, the perception that people are having.
273
00:46:17.818 --> 00:46:32.728
So we had to kind of really, uh, take, take a step back and think about, um, you know, what verification is supposed to mean, um, and what it would mean regardless of what we want it to mean. Uh, what is the perception?
274
00:46:32.828 --> 00:46:48.608
Because the perception is, is the reality. So, you know, it's, it's a, it's a complex, um... It, it, it's a complex thing. And, uh, we want to be thoughtful about how we go about, um, thinking about verification.
275
00:46:49.208 --> 00:46:59.148
So it is on pause, and, uh, I have nothing else to share about that at this moment. No worries. Well, um, we-- I'm conscious that we're rattling fast approaching the end of the time we've got with you.
276
00:46:59.208 --> 00:47:02.448
So there's a couple of questions that we wanted to get in there with you to talk about.
277
00:47:02.528 --> 00:47:18.318
Um, one of the things that, um, I want to ask was th-that I've always seen Twitter as having quite a narrow set of features and functionality for users compared to, say, something like Facebook, which is, you know, you go to post a new status update, and you've got a raft of options and formats that you can use to, to post things.
278
00:47:18.698 --> 00:47:28.828
And with Twitter you have, you know, you can put, uh, attach a GIF or an image. Um, and, and there's limited else that you can do with, like, editing of, uh, of photos that you attach to your tweet and things.
279
00:47:29.118 --> 00:47:36.428
And I just wondered, is that something that, uh, Twitter is, you know, is conscious of and intent-- you know, is an intentional part of the, of how it is, its brevity?
280
00:47:36.808 --> 00:47:47.648
Or is that something that you see as a, a possible weakness that it needs development for in that maybe being able to attach a PDF or being able to attach an audio clip in, in a different way and, and other broaden out the functionality?
281
00:47:47.688 --> 00:47:52.468
Is that something that has ever been discussed, or what, what's the background with that? Sure.
282
00:47:52.508 --> 00:47:58.508
I'll, I'll maybe just start with maybe just a, a framework for how, uh, things have been thought about in the past and how we're thinking about it now.
283
00:47:58.908 --> 00:48:13.078
I think there was a time many years ago where we, we still weren't quite sure what Twitter was, and that meant that, you know, things like, "Hey, let's attach PDFs to tweets," would be, you know, something that we could, we could have a, a serious conversation about.
284
00:48:13.168 --> 00:48:22.588
Um, but as the years went on and we, we realized that what Twitter is for is to see what's happening and discuss what's happening and all of the conversations around what's happening.
285
00:48:22.608 --> 00:48:36.868
And that is super important from a product perspective to, for us to kind of guide the features that we think about building and, and prioritize the features that we, we think about building, and that, that guidance has been very clarifying for us.
286
00:48:37.358 --> 00:48:44.388
And that's not to say that, you know, PDFs to tweets are a bad idea. Um, and if we found that there was a
287
00:48:45.452 --> 00:49:06.052
Um, there was a need, a real user need around seeing what's happening and sharing what happening that involves, you know, attaching PDFs to Twitter and, or to tweets and, um, that that was l- an opportunity, um, in our space that felt like, um, um, you know, people really need, then that's something we would consider.
288
00:49:06.172 --> 00:49:09.412
At this moment, I don't think PDFs to tweets are at the top of that list.
289
00:49:09.732 --> 00:49:18.432
[laughs] Um, and I think, I think what you've said around the, the, you know, the, the brevity and the, the simplicity of Twitter is definitely part of its superpower. Um, and that just...
290
00:49:18.472 --> 00:49:25.812
those are just kinds of the trade-offs that we have to think about when, uh, we're looking at the gamut of ideas of, of what to do next.
291
00:49:25.932 --> 00:49:39.402
Um, you know, is it, is it guided by our very clear, uh, purpose around serving the public conversation and making sure that people can see what's happening and, and, and participate in, uh, the conversations around what's happening?
292
00:49:39.672 --> 00:49:47.452
So. Well, we've got two quick questions left here. I think Martin's got one a bit about your career that, uh, we wanted to ask, and then I've got one sneaky one at the end, and then we'll, we'll wrap things up.
293
00:49:47.572 --> 00:49:59.552
Martin, ta- talk a bit about that. Yeah. Yeah. You, you've got some really impressive apps on your, uh, resume. Um, you've, uh, worked at Vine, Secret, Periscope, uh, three hot apps of their time.
294
00:49:59.852 --> 00:50:07.072
[laughs] And, uh, certainly when I was, uh, um, working as a full-time tech journalist, they were apps that I was following, uh, uh, very ve- very, uh, carefully indeed.
295
00:50:07.432 --> 00:50:16.312
Um, and now obviously you're in a very senior position at, at Twitter. Um, uh, what would you recommend someone wanting to follow a career path like that, um, what would you recommend they do?
296
00:50:17.252 --> 00:50:32.072
Um, yeah, I've had definitely a fun time working on a lot of different, uh, social apps. Um, if... You know, I think it was also during a time where it was kind of the, the glory age of, of consumer social apps as well.
297
00:50:32.662 --> 00:50:46.252
Um, so, uh, the timing there was, uh, particularly, um, uh, it amplified I think how, how much fun, um, all of that was. Um, for people who are interested in working on apps in general, I, you know,
298
00:50:47.532 --> 00:50:55.312
I was always very customer-focused, um, in my, in, in the way I, I went about engineering and, and, and development.
299
00:50:55.892 --> 00:51:09.652
And that really lent itself to, um, working on consumer-facing apps and, uh, uh, you know, trying to create delightful user experiences. I worked in Android for most of my engineering career.
300
00:51:09.752 --> 00:51:24.132
Um, and at a time when the Android user experience was notably inferior to the iOS experience, and I kind of really, um, took it to heart that I wanted to create delightful experiences on Android.
301
00:51:24.252 --> 00:51:28.492
Uh, and really thinking about it from the user perspective and from the usability perspective.
302
00:51:29.012 --> 00:51:42.972
Um, so my advice is, you know, to folks that are interested in that, is to, is to be super focused on the user, um, or to find what that sweet spot might be for you.
303
00:51:43.072 --> 00:51:55.732
Um, and, uh, to kind of carve out, um, carve out roles for yourself that can utilize, uh, what you think that sweet spot might be. Um, as I had mentioned earlier in our chat, um,
304
00:51:56.912 --> 00:52:12.112
um, I'm often, I'm often deployed in the zero to one phase of a project, and that's kind of been my sweet spot. Uh, I have taken things from nothing to something, uh, repeatedly, and it's a special skill.
305
00:52:12.312 --> 00:52:21.632
And it's a different skill to be able to take something from one to two or two to three. Um, and, uh, that's, you know, the, that's where I shine less.
306
00:52:21.972 --> 00:52:34.992
And I think for folks that are thinking about their next career move, um, I would, I would encourage people to look at where in the development cycle they think they shine the most and to try to get experiences, uh, in those places.
307
00:52:35.292 --> 00:52:42.352
Oh, well, and we really appreciate that you've come on today's podcast with us. It's been a, a real pleasure to have you on. I just wanted to get one last quick thing.
308
00:52:42.452 --> 00:52:46.892
I did ask you before you came on about, uh, a question I was gonna ask, and I wondered if, what you're gonna come up with.
309
00:52:46.912 --> 00:53:00.272
And it was just to ask you if there's any little unknown features or hidden Easter eggs or things that people are like, "Oh my God, I didn't even realize that was there," or, or quirky things that the developers have snuck into the code that maybe some people haven't noticed.
310
00:53:00.312 --> 00:53:12.702
Was there anything you could quickly share with us that, uh, people could go and find that they didn't maybe realize was there? Sure. So I've got three slash four really quick ones. Um- Ooh.
311
00:53:12.792 --> 00:53:16.802
So w- I say three sl- three slash four 'cause two of them are really related.
312
00:53:17.242 --> 00:53:31.532
Uh, so on the Twitter app, um, if you use, uh, multiple accounts, maybe you have one personal account and one for your company or your brand or something, uh, you can press and hold the home button to quickly switch between accounts, which I feel like a lot of people don't know.
313
00:53:31.592 --> 00:53:47.742
Nice. Um, so that's very useful. And then v- very related to that, you can also press and hold on the, uh, second button to, uh, quickly launch into search. So that's, uh, another kind of, uh, pro tip. Nice. Good.
314
00:53:47.912 --> 00:54:06.492
Um, another thing, uh, that I recently learned that people don't always know about is if you swipe into the dash on the left where you see your profile and your bookmarks and your lists and all that stuff, at the very bottom left there's a light bulb, and you can use that to quickly switch between, uh, light mode and dark mode.
315
00:54:06.572 --> 00:54:18.792
So, um, if you were going into settings to change, uh, the mode that you're in, uh, you don't have to do that. There's a shortcut. Nice. That's good as well. And [laughs] Finally, one for Little T.
316
00:54:19.372 --> 00:54:24.652
It's something that we've tweeted about a bunch, but I still get messages sometimes that people don't know, is, um...
317
00:54:24.672 --> 00:54:31.252
and I, I'm not sure if we're gonna keep this forever, but at the moment in Little T, you can swipe left to like a tweet.
318
00:54:31.952 --> 00:54:42.712
So if you are struggling to, uh, tap and like, and you feel like that's too many taps away, um, we've implemented slip- swipe to like. I love that. I use it all the time. It's brilliant.
319
00:54:43.052 --> 00:54:50.032
Um, are you still accepting people into, um, into the, uh, prototype, TWTL? We are, and they tend to, they tend to go in waves.
320
00:54:50.132 --> 00:55:01.672
So if you, um, apply at the link, uh, it was tweeted out from Twitter Support, and, uh, I can also send you the link to send out to, uh, your audience later. Oh, we, we'll include it in the show notes. Yeah. Yeah. Yeah.
321
00:55:01.712 --> 00:55:12.212
Well, I'm, I'm, I'm surprised or disappointed, I don't know which one it is, that you can't do something on the Twitter app where you swipe up twice, left three times, and down once and the picture of Jack comes up singing a song by Rick Astley or something like that.
322
00:55:12.292 --> 00:55:21.724
Is that... [laughs] That was what I was thinking the Easter egg was gonna be. But those are- You're good enough for me. So, um, I wanna say a real big thank you to Twitter because, um, you are our first guest on here.
323
00:55:21.984 --> 00:55:26.484
Um, we really appreciate that you, um, were willing to talk so openly about some of the things that you're, you're getting up to.
324
00:55:26.594 --> 00:55:36.224
And, uh, we really hope that you, uh, come back in, in season two and chat to us a bit more. But, um, for now, Sarah, it's been a pleasure, and I hope we get to chat some more sometime soon. Awesome.
325
00:55:36.264 --> 00:55:39.244
Thank you guys so much for having me on. Thank you. Thanks a lot.
326
00:55:40.064 --> 00:55:50.744
Now it's time for Quick Hits, which is a section of the show where we'll look at all the fun little things that have been spotted being tested in apps, uh, or new features that, uh, maybe you're not aware of.
327
00:55:50.824 --> 00:56:00.434
And Matt is one of those people who seems to spot a lot of these things and share them. So, uh, Matt, I'll hand over to you. Thanks. Yeah, I am the, the ultimate geek when it comes to this stuff, I guess.
328
00:56:00.504 --> 00:56:09.244
So I should, uh, figure out what we've been seeing in the last few weeks. So a couple of quick ones, and they're mostly Facebook this week. Uh, Facebook Messenger screen sharing feature.
329
00:56:09.364 --> 00:56:20.504
So this is something that Jane Wong f- um, spotted and, uh, it's basically an extension of, of the video c- um, conferencing facilities that you see, uh, video messaging and video calls within, uh, Messenger.
330
00:56:21.064 --> 00:56:29.024
Uh, so we've seen a couple of apps now that have come out in the, in the past year where you're able to share your screen since, uh, iOS enabled that f- um, feature for, for other apps.
331
00:56:29.064 --> 00:56:41.724
And so you can imagine a scenario now where you will be able to be having a, a call with, uh, one or more friends on Messenger, and then you'll be able to share the screen and show them what you're doing and what, what you're flicking through and what you're watching or what you're reading and things.
332
00:56:42.104 --> 00:56:49.004
So, uh, whether that will see the light of day like all of these small things, who knows? But that is something that was spotted, uh, in the last week.
333
00:56:49.644 --> 00:57:03.684
Um, another thing is a minor thing, but I think it's probably more exciting than it really is, which is, um, Facebook on its homepage where you sign up to be on Facebook and you fill in all your details about your name and, and register for an account which many of us probably haven't done for a long time and might not do again.
334
00:57:04.184 --> 00:57:12.764
Uh, it does say the words, "It's free and always will be." Um, but-- And someone spotted this week that it's changed to, "It's quick and easy."
335
00:57:12.864 --> 00:57:23.284
Um, so there was, uh, instant speculation around the fact is that a suggestion that Facebook is gonna start to charge or is there something more interesting going on there? Uh, my suspicion is probably not.
336
00:57:23.354 --> 00:57:30.744
It's more likely that this is just one of many AB tests that seem to perform particularly well at onboarding people, uh, and, and nothing more.
337
00:57:30.824 --> 00:57:36.084
But, uh, maybe we'll see something else on that in the coming, uh, months, but we don't know.
338
00:57:36.184 --> 00:57:49.214
Uh, also Facebook, um, showed off a design, a new, simple, bold, fresh design, uh, F8 for its, uh, desktop version of facebook.com for the news feed. You know, brighter colors, uh, simpler, um, icons, um, bolder text.
339
00:57:49.284 --> 00:57:56.644
Everything was a little bit kind of more minimalist. Uh, and finally, some people have started to see that roll out to them, uh, in different countries around the world.
340
00:57:56.744 --> 00:58:01.684
And then finally, also there's a tool called socialmediatoolcomparison.com.
341
00:58:01.884 --> 00:58:21.984
It's a simple website that was on Product Hunt actually, uh, where you can go in and as a social media manager, one of the tricky things to do sometimes is to be able to figure out which, uh, third-party management monitoring or analytics tool you want to use based around the criteria that you're looking for for certain features or certain functions or a price that you can afford per month.
342
00:58:22.464 --> 00:58:32.084
Uh, and although this is, uh, built by a company that is itself a third-party vendor that does social media management tools itself, uh, I still think it's a really useful tool.
343
00:58:32.144 --> 00:58:40.764
You can filter all, all sorts of criteria and figure out which platform, whether it be Buffer or Hootsuite or Socialbakers might fit the, fit the bill for what you're trying to do.
344
00:58:40.804 --> 00:58:54.544
So socialmediatoolcomparison.com is the site to have a look at, and it's made by a company called Sociality.io. Nice. Well, we've reached the end of the show. Uh, I think that went pretty well for an episode one, Matt.
345
00:58:55.234 --> 00:58:59.344
I was quite impressed myself actually, and I'm quite proud of the fact that we got through it.
346
00:58:59.444 --> 00:59:07.044
I, I'm sure people are gonna pick holes in some of the sound quality or the production values, but we're faking it till we make it, and, and that's all that matters to me.
347
00:59:07.104 --> 00:59:16.564
This was, uh, throw up as an idea many, many weeks back, and we finally made it to, to, to this stage. So I'm, I'm proud nonetheless. But, uh, next week is looking pretty good as well. Absolutely.
348
00:59:16.684 --> 00:59:27.364
So, uh, this, uh, Geek Out, uh, title for the podcast, uh, came from the Facebook group. Uh, so if you want to discuss the podcast, you might want to head to the Facebook group. Yeah.
349
00:59:27.424 --> 00:59:38.044
So if you wanna join the Facebook group, search for Social Media Geek Out on Facebook, and you will find the group. And, uh, tell us next week who is on the show, Martin. Yes. Next, uh, episode will be out next Thursday.
350
00:59:38.104 --> 00:59:48.224
That's the twelfth of September. And our guest will be Kerry Flynn, who is a tech journalist who has worked at, uh, several different places. Uh, recently landed at CNN. So she's a, a real big deal now.
351
00:59:48.424 --> 00:59:55.124
And, um, we'll be, uh, discussing all sorts of exciting, uh, things around the future of social with her. So I'm really looking forward to having her on.
352
00:59:55.164 --> 01:00:03.144
She's a, a good friend of mine, and, uh, I'm really, um, sure she'll tell us some fascinating stuff around social with her strong opinions as always, which is exactly why we, we love her.
353
01:00:03.204 --> 01:00:09.164
So, uh, I've been Matt Navarra, and I'm on Twitter if you wanna find me. So send me some messages. Um, I'd love to hear a bit more from you.
354
01:00:09.304 --> 01:00:18.524
And you can also, um, use the hashtag geekoutquestions, and I will have a look and see if there's anything in there that's, uh, really exciting and interesting from people that have listened to the show.
355
01:00:18.624 --> 01:00:30.664
Uh, also, you can record, uh, a message to us through, um, Anchor. So if you go to anchor.fm/geekoutmattnavarra, you'll find an option to record a message, and maybe we'll play some of those back in future shows.
356
01:00:31.104 --> 01:00:42.664
Absolutely. I'm on Twitter as martinsfp, so, uh, do find me there. And, uh, Geek Out with Matt Navarra is a Big Revolution production. You can find out more about Big Revolution at bigrevolution.net.
357
01:00:43.244 --> 01:00:56.804
We'll see you next week. Thanks for joining us. Goodbye. Goodbye, geeks. [upbeat music]