July 2020

277 tweets

Replying to @null and @tylerhower, @HangoverHaze, @SellarsRespectr, @LogicalAnalysis

The only downside of sardines is they don't go that well with thick cut ham. So you have to choose pineapple + jalapeños + either of sardines or ham.
A good example of how to flatten a curve.
Quoting a tweet by @bendreyfuss ↗
A great read about how "the scientific method" misrepresents scientific practice and impairs an integral understanding of the world. nybooks.com/articles/2020/…

Replying to @Forsheit

I'm not a lawyer as you well know but the alignment to core GDPR principles is very perceptible in the CPRA. I don't think that an adequacy decision is necessarily a slam dunk but I'd think they have legs to stand on. It would crack me up if California beat the UK to it 🙂

Replying to @Forsheit

Heh, the northeast might be on that path too!
The difference in pronunciation between "infinite" and "finite" will never cease to grate. I mean, pick one, people.
↺ 1

Replying to @_narratorvoice

Well played!

Replying to @JuliaAngwin and @themarkup, @adrjeffries, @LeonYin

I guess that's one way to promote racial justice.
The hardest job in technology is coming up with a cool acronym. Everything rides on that, everything.
↺ 1

Replying to @Paul__Walsh

I don't think I need anyone "helping" my natural tendencies, here 😉

Replying to @null and @benlikestocode

I once submitted to W3C a draft of "SVG Extensions to XPath". Sadly, someone noticed.

Replying to @null and @benlikestocode

You should come hang out with the Data Governance team, we maintain PURR, have been working on standard docs known as CAT and PUP, were responsible for the REVAMP document, and are thinking about a new system called BARK. It’s a riot.

Replying to @brucel

I thought it meant “Makes Bitter Acronyms”.

Replying to @karaswisher and @joejerome

Yes but they can testify about how it's done. @joejerome is right, it would be a hell of a show. Have them cross-examined by Ballmer! I hope @RepCicilline gives it a shot 😁

Replying to @JamesGRobinson

I remember well finding out you had put a meeting labelled "DRINK" on my calendar in the middle of the day, and I was like "OK, I guess!"

Replying to @briankardell and @LeonieWatson, @ewaccess, @karlgroves

I'm not sure where the video is but I'd watch again!

Replying to @parb and @brucel

Well who was dumb enough to let a frenchman do that to them, I have to ask.

Replying to @jesslynnrose

Oh so you just enjoy walking up to people and ruining their day, do you? *breaks into uncontrollable sobbing*

Replying to @null and @iandevlin

The fun one with this one is that non-natives often think "infamous" means "very famous". I've seen hilarity ensue.

Replying to @jesslynnrose

If you keep proselytising this you'll end up either famous or infamous.

Replying to @jesslynnrose

Hahaha no I did not see that but it sounds like the perfect pairing for popcorn and wine.

Replying to @jesslynnrose

Quality burn 🔥!

Replying to @DynamicsSIAM and @CT_Bergstrom, @JenLucPiquant

*slaps roof of 2020* This bad boy can fit so much nightmare dystopia.

Replying to @alexanderdanilo

Aren't we all!

Replying to @LeonieWatson and @briankardell, @ewaccess, @karlgroves, @w3c

It's on Vimeo! Now I remember. Thanks a lot!
On second thought, driving innovation through the patronage of rich morons may have been a mistake.
↺ 6

Thread of 2 tweets

Replying to @jwajsberg

Si, mais il y a plein de crétins qui les admirent. Aussi, ils décident des startups qui seront financées.

Replying to @null and @aliceafung, @k_johnsn

I have a PUP draft on the way!

Replying to @null and @wileyprof, @LogicalAnalysis, @birchlse, @timcrane102, @smellosopher

Was a paper edition ever considered? My only criticism of the SEP is how much time I spend printing it 😁

Replying to @null and @wileyprof, @LogicalAnalysis, @birchlse, @timcrane102, @smellosopher

I'm stupid enough that I might 😬 More seriously, maybe not the whole thing, but thematic volumes, eg. all medieval, all quantum... I have *heaps* of SEP printed lying around, I'm seriously considering pasting PDFs together and getting my own binding from Lulu or some such 🙂

Replying to @vinnysgreen and @acfou

OK, I'm curious 😀
For those of us not quite so lucky it's coming out next month! bookshop.org/books/entitled…
Quoting a tweet by @Skye_Cleary ↗

Replying to @vinnysgreen

I liked cooking too!

Replying to @vinnysgreen and @acfou

Thanks for the details! I've been curious about trying Zeus (not my call, but still curious). We don't do personalised in the EU so that part at least won't be a problem.

Replying to @hartzog and @Iwillleavenow

Lol, 9!

Replying to @LogicalAnalysis and @shearted, @wickrotate, @add_hawk

Is the geometric definition not superior in all respects? cuberule.com

Replying to @tomdale and @can

Hey, that was my question!

Replying to @can and @tomdale

I hope not *too* soon seeing as I'm not in NY right now!

Replying to @null and @marcosc, @amyvdh

I was gonna say: you have to.
A bunch of good recommendations here. "Seeing Like A State" is one of the most important books I've read these past years, and I'm very curious about "Are Prisons Obsolete".
Quoting a tweet by @themarkup ↗

Replying to @LauKaya and @ka_iwanska

Le plus gros choc c'est d'apprendre qu'Oracle a une solution cloud, et qu'en plus elle est suffisamment utilisée pour apparaître ici.

Thread of 2 tweets

Replying to @DinaSrinivasan

I'm curious as to why withholding YouTube inventory is seen to be problematic but Search inventory isn't?

Replying to @DinaSrinivasan

Creating singular user IDs is a terrible idea! I'm not sure how that would ever happen, it's begging for abuse and way too dangerous to be justified. Banning cookie syncing seems like a much better idea and where things are headed.

Replying to @sarahdrinkwater and @OmidyarNetwork

Hahaha, every bit helps! Judging from early reactions, the “Slow Down — Ask Questions” sticker is going to be popular. Is the plan to make it possible to order relatively large batches, eg. so Governance teams can spread them around?

Replying to @DinaSrinivasan

That’s fair, but I don’t think _that_ much Search inventory is available through other parties. Information from Search certainly isn’t!

Replying to @DinaSrinivasan

I think that single ID are exactly the wrong solution to this problem. Assuming the privacy issues were solved (which seems impossible on first principles, but let’s assume!) IMHO it still fails to address the competition issue.
Data is non-rivalrous, multiple parties can observe it at the same time without detracting from each other. So perfect cookie syncing takes nothing away from Google, though it does add to others.
However, the value of data isn’t additive. Individual datum don’t have value, it’s the *links* between them that do. “3” is without value, but tied to me you can use it as my birth month. The value of tracking a user on sites A&B is more than the sum of behaviour on both sites.
This effectively translates to network effects in the value of data. The entity with the most data, the most links, can derive more value from the same data that another party also has, because it also has links to other data.
So a universal ID still structurally privileges the biggest party. It also furthers intermediary capture, which means less accountability. I think the trick isn’t to spread a non-rivalrous commodity around more; it’s to make it rivalrous. Prevent all sharing across contexts.

Thread of 5 tweets

Replying to @sarahdrinkwater and @OmidyarNetwork

Gotcha, that makes sense. If there were a way to order a batch by covering costs there could be interest (no promises, but there could be)! But I can imagine how it might not be viable.

Replying to @DinaSrinivasan

I don't think that cookie IDs are rivalrous (everyone gets the tracking insights), though indeed making it easier for marketers to shop around helps a bit. But structurally you end up with a market that's Google vs a bunch of smaller Googles. They get gobbled up!

Replying to @DinaSrinivasan

We're getting there!

Replying to @DinaSrinivasan

I hear you, but I don't think that works unless the structure of the market changes. Otherwise we're back here two years from now. Splintering data collection helps a lot, IMHO.

Replying to @mariafarrell

Ah! Memories... They used to be so combative in standards that people used to talk about how to spite their proposals even when they didn't care about the actual tech. Now they're the sweetest people you can work with.

Replying to @mariafarrell

Ha! Indeed. I think one of the others currently under investigation has tried to learn from that. I’ve never had someone tell me to go fuck myself so nicely (if somewhat arrogantly). In the end, I’m not sure it’ll be enough to hide the message, though.

Replying to @lounecl and @_jasterix

Oh no! Noooooo. I’m not a fan of Workday, the only thing is has going for it is that it less horrible than the alternatives…
Developing large scale open technology that works for more than one category of stakeholder takes time. It may be tough for those discovering that, but it beats returning to the adtech morass. adexchanger.com/online-adverti…
↺ 3

Replying to @Paul__Walsh

Adtech has been operating with very shoddy standards-setting for decades; it’s difficult for trade journalists who are used to covering the @IAB to suddenly shift to a whole new world. They’re usually smart cookies, though, they’ll adapt!

Replying to @lounecl and @_jasterix

Yay! Fingers crossed!

Replying to @Joygraciaph and @JoyHarjanto

"Oh, you're into machine learning? Name all the gradient descents."

Replying to @alexanderdanilo and @svgeesus, @CodepointsNet, @TiroTypeworks, @lianghai, @KhaledGhetas, @kalapi, @shellshear

I only said that out of spite because the first time I had to take minutes at my very first f2f I got to scribe a debate over supporting ligatures across bidi boundaries. I think I never fully recovered.
↺ 1

Replying to @Annaleen and @emilytheslayer

I came here to say all four; there's just so much the world needs to hear about this, I wouldn't self-censor in these times of censorious intolerant illiberalism. Loving the section headers idea!

Replying to @kashhill

This might be what the word "covidwashing" was invented for.

Replying to @PrivacyMatters

I agree that it’s a very good presentation of the situation, I just thought it was a shame that it did not put what SDKs do in the context of what the OS does. Android is a *massive* geolocation tracker, somehow I think that should be pointed out along those smaller companies.
I love this piece. Ban the cars!
Quoting a tweet by @fmanjoo ↗

Replying to @fmanjoo

The one thing that I thought was missing from those images was outdoor restaurants & cafés. The café culture in New York is almost inexistent (despite the coffee being better) and restaurants are insanely noisy. More outdoors space for these would rock.

Replying to @LogicalAnalysis

It's also the only way I can write sentences that don't have twenty seven subclauses 😂

Replying to @MonocleDebacle and @fmanjoo

Living in a city with substantial outdoor culture produces huge positive externalities, including for people who don't go. You get the kind of urbanism Jane Jacobs talks about with neighbourhoods that gain a real density of life.

Replying to @null and @eyesondesign00

This may be a closer source
Quoting a tweet by @egg_dog ↗

Replying to @null and @eyesondesign00

Credit to @kathyvsinternet for sharing this, it's beautiful.

Replying to @jilliancyork

I prefer vinaigrette over mayo for that purpose. That's probably terribly French though.

Replying to @null and @KendraWrites

Is this bad because it's mayo when it should of course be vinaigrette?

Replying to @null and @KendraWrites

When I'm lazy I just olive oil, but vinaigrette is the go-to! You've been to France, you must've seen this 😁

Replying to @null and @KendraWrites

Sometimes my kids run out of vinaigrette halfway through the avocado and I have to refill them a bit. It's not "a dressing", it's vinaigrette!

Replying to @realmaplesyrup and @floggingkatie, @KendraWrites

Hahaha 😂 In fairness, though, avocados tend to be pretty tasteless in France so you sort of have to spice them up.

Replying to @trekonomics and @KendraWrites

There's also that thing they do to sushi and rolls here, criss-crossing them with mayo. I... why?

Replying to @lilianedwards and @ShannonVallor

What would cats use money for, they own everything already.
"Leftist movements today see our crises as intersectional. Police violence, global warming and unaffordable housing are not disconnected, discrete problems; instead, they emerge from colonialism and capitalism." -@orangebegum Loved this. Let's throw tech reform into the mix! 🤓💣
Quoting a tweet by @JenParker393 ↗
I guess the people *are* eating cake?

Replying to @jjcitron_ and @daniellecitron, @JoeBiden

I can't believe I'm defending this but in fairness there is a demographic of people who read newsletters assiduously but don't have a news app, and will probably learn it faster that way. They might skew a little on the older side, though!
I can't believe there is no horror classic about snails, it just writes itself.
Quoting a tweet by @qikipedia ↗
↺ 1

Replying to @gsnedders and @Apple

Congratulations to them on a most excellent hire!
I hadn’t seen this quote when that article came out: “We are moving from a world where you can put anything on your website to one where you can’t because Google says so.”
Quoting a tweet by @WebDevLaw ↗
↺ 1

Replying to @ChloeAngyal

I'm no expert but there are plenty of lock-picking videos on YouTube. Maybe you could look into it, of course completely for journalistic purposes, not to break the law or anything... 🤭

Replying to @Yascha_Mounk

These elderly folks are also putting high-quality comfortable footwear on display.
You can't fool me @NYTStyles, I know that's cake.
Quoting a tweet by @NYTStyles ↗
Are any of you very fine people using @RoamResearch Quick Capture? I'm confused that it seems never to sync unless I switch to the full site version, at which point it asks if I do want to sync. Do you ever not just want that?

Replying to @null and @karinasaidi, @natashanyt, @AJStream, @AJEnglish

Hi Karina, I appreciate the offer but I'm afraid that is too tight a deadline and also not my area of expertise. Thank you!
Happy Bastille Day everyone, remember to overthrow your government and ransack your prisons every now and then!
↺ 2
It turns out software wasn’t eating the world, it was just eating venture capital.
Quoting a tweet by @techreview ↗
↺ 2
One interesting thing with writers leaving newspapers and posting to their blog instead is that we get to see just how crucial having an editor can sometimes be.
↺ 8

Replying to @tabatkins

Mostly I put “m” at the end of things because I went for punctuation on the keyboard and missed. I’ll see myself outm

Replying to @stshank

Editors never get anywhere near enough credit. Having worked with one on a project, I find it cruel to have to write anything without one!

Replying to @Chronotope

My team writes reports for leadership all the time. We'd save precious time for them if we had an editor. I'm considering training as one, if that training exists.
"We're defaulting to not backing up some pics in order to preserve our very scarce Internet resources (because Covid-19, you know), we picked this set at random it's totally not a list of our competitors."

Replying to @null and @funnymonkey

Thank you, sir 😁

Replying to @Chronotope

True, true, but when I get headcount I always have a million other priorities anyway, so... I'll still try to at least train myself!
Anticompetitive Behaviour: The Movie. wsj.com/articles/googl…

Replying to @__apf__

Pasting it straight into OKR reporting.

Thread of 2 tweets

That so much rides on the data economy is reason to focus *more* on improving DSRs. "Oh there's a lot of data and it's really important so let's half-arse it" isn't a very strong argument.
↺ 1
I doubt that anyone with a substantial data operation does DSRs absolutely perfectly in every instance. As a rule, tech has tended to grow faster than our ability to govern it. DSRs help ensure people stay in the driver's seat. Let's push *forward*!
↺ 1

Thread of 3 tweets

Replying to @random_walker and @s_englehardt

I think there’s a third way that is more promising and has a future: improving the Web. It’s a bit of a painful slog, but those ducks are lining up and we should seek ambitious progress here.
The legal option is difficult because regulators will hesitate to go after small, struggling publishers who, as you point out, don’t have the resources to do that right but are often the worst offenders. But law & enforcement can be tied to technical capabilities when existing.

Thread of 2 tweets

Replying to @kacie_ha

It does! I guess great minds like the same colour schemes.
I get those warnings too and this is extremely true.
Quoting a tweet by @Jesse_Hirsch ↗
So, let's be honest for a second here. How many of you checked to see if your account was important enough to have been #hacked?

Replying to @ShaneHudson

Same!
I have to admit, this idea of cancelling all the powerful accounts at once is so much better than having them all sign some asinine letter in small batches.
↺ 2
OH SHIT THEY'RE BACK ACT CASUAL
I have to admit, this idea of cancelling all the powerful accounts at once is so much better than having them all sign some asinine letter in small batches.
↺ 2
↺ 1
Filtering out verified accounts on demand would actually be a pretty damn good feature, @Twitter.

Replying to @heidilegg and @Twitter

I know, I know, as did I. But still, I think it would make for a neat mode. Listen to the least amplified first.

Replying to @mnot

Heeyyyy, buddyyyyy! What up, man?

Replying to @cwarzel and @BobbyBigWheel

How could you tell the difference?

Replying to @fmanjoo

Yes — but I can’t come up with a coherent scenario of how it would play out. Has anyone polled with Pence at the top? Mayhem and murder as GOP factions fight to pick either P or VP candidates or both is all that seems certain.
One interesting effect of Core Web Vitals (web.dev/vitals/) using Chrome performance telemetry is that optimising for Chrome means you'll rank better on Google Search. Just putting this out there.
↺ 12

Replying to @vinnysgreen

I never really got why people used the Chrome dev tools. They feel pretty clunky.

Replying to @vinnysgreen

I'm generally on Firefox. I should have been clearer on Chrome dev tools: usually when people swear by something is can see why, even if it's not my thing, eg. I get why emacs even if no thanks. But here they just seem pretty normal? And missing some bits maybe?

Replying to @cramforce

At the type of scale and evolutionary speed we're dealing with even a very small benefit will drive a structural advantage. Also, given the opacity of rankings and the cutthroat competition for them a rational dev will have to optimise for Chrome.

Replying to @null and @icopilots

That would be a great rule!

Replying to @cramforce

I haven’t researched a current issue, but there’s precedent for substantial perf differences over seemingly trivial details, eg. the famous YouTube transparent div. I once fixed a buglet in React that was a Chrome-oriented optimisation that failed in Firefox & they didn’t know.

Replying to @__apf__

Sometimes it will but sometimes it won’t! There’s plenty in these beasts that’s engine-specific, and all kinds of corner cases (as you know much better than I do!).

Replying to @__apf__

If you’re a web dev and you notice a disparity you’ll have to pick the Chrome-friendly variant, even if Chrome isn’t the #1 browser for your mobile population. Also, if in doubt the smart move in this context is to just optimise for Chrome.

Replying to @stshank and @__apf__

Chrome isn’t dominant in *all* cases just yet :) I don’t know that the issue will come up, but for us, on mobile, this certainly makes it more relevant than it would be from just marketshare.

Replying to @twifkak

I’m not implying motive; it’s possible to produce anticompetitive outcomes without explicit motives (and we probably shouldn’t care).

Replying to @BrandonWarmke

James C. Scott?
And that's just those in news! Come work with us, you'll have lots of amazing colleagues!
Quoting a tweet by @TimHerrera ↗
↺ 1

Replying to @swodinsky

Like, I think you'd fit right in?

Replying to @fmanjoo

Hiring a trainer 1h a week. It's not cheap, but arguably some of the best money I've ever spent. Sometimes I even exercise for pleasure now, and if you'd known me five years ago you'd be demanding proof.
Any chance those are the same experts, lawmakers trying to impose backdoors and forbid end-to-end encryption? nbcnews.com/news/amp/ncna1…

Replying to @ubiquity75

This haircut, so fashionable.
How come this article didn’t get called “Pervasive Surveillance Companies Too Write Open Letter About Getting Cancelled”? theregister.com/2020/07/17/agg…
↺ 2

Replying to @null and @funnymonkey

Absolutely. They have a very good case that Google is cheating. But they also have been proceeding with no governance whatsoever for twenty years. Whining about the W3C isn’t going to fix that.
A Duty of Loyalty for Privacy Law from Neil Richards & @hartzog just freshly off the presses: papers.ssrn.com/sol3/papers.cf… Pairs perfectly with any glass of wine for a sunny Friday evening’s reading.
↺ 1

Replying to @thezedwards and @funnymonkey

Do you have a specific example? Having rules is kind of a hallmark of governance, refusing to learn them doesn't mean the others are doing regulatory capture. The web is a massively distributed system with organic evolution. There are reasons why things work differently.
Every few years a bunch of companies show up and scream bloody murder that things don't work the way they do in their own lobbying group that does a bit of bad tech as a side gig. These folks seem no different?
I've heard the reg capture argument before. But as a kid n00b from a tiny startup no-one had heard of with a motley alliance I prevailed over the monsters of the day, IBM and Microsoft, by paying attention to how things work and why. So, uhm, citation needed?
I'm not saying things are perfect, implementors always have excessive power. There's no magic. But it's almost shocking that they don't have more.

Thread of 4 tweets

The @w3c members' contract should make companies pay more for every white man they send to a group above an initial quota.
↺ 2

Replying to @hober

I'm not opposed :-D

Replying to @frivoal and @w3c

People can self-identify, I'm pretty sure that's enough here.
"You can talk about journalistic objectivity all you want but I don’t want to live in a place where federal agents in unmarked vans abduct people."
Quoting a tweet by @cwarzel ↗
↺ 9

Replying to @reidmore_online and @wendy_a_reid, @w3c

I think we're on to something!

Replying to @fatemehx2 and @Iwillleavenow

I confirm, they're a fixture at The Times and after initial horror I'm pretty hooked!

Replying to @derGeistbot and @Geistb0t, @virgil_30

Just Dilles with it.
I seem to be seeing the Twitter follow bug again, this week I had to re-follow people I'm positive I followed already. Anyone else notice that? It also reverted to "Top Tweets First" again, which I can always spot because it just shows me tweets I've already engaged with...
↺ 1

Replying to @hartzog

It had stopped for me, but it seems to be regressing.

Replying to @BrendanNyhan

Clearly, testing kills.

Replying to @LogicalAnalysis and @neilsinhababu

Carthago delenda est! Oh wait, sorry, wrong battle 😅
You'd think writing code to fight a pandemic would be a moment to exercise ethics. For Google, it's just a chance to grab more of your data. "Why did you do that," asked the frog, "we'll both drown!" "Because it's in my nature," answered the scorpion. 🐸 🦂
Quoting a tweet by @natashanyt ↗
↺ 21

Thread of 2 tweets

Replying to @johnwhegeman and @kevinroose, @crowdtangle

Thanks for the details. I am curious about a few things. * Isn't any content a tiny % of what most people see? * Is reach the number of people who might have seen an item if they happened to be there? * Are impressions just scrolling past an item or something more?
* Do you have any internal studies about mental persistence or likelihood of re-engagement with the same content from various ways of engaging? Just curious, thanks!

Thread of 2 tweets

Replying to @EvanSelinger and @markhurst

I'd forgotten it was in that movie too!

Replying to @rektide

It makes sense to bundle Bluetooth and location, yes. But. They could open up location without granting themselves the right to constantly collect it too. And yet, that's exactly what happens!

Replying to @rektide

I would be sympathetic if their threat model included the fact that they will collect the location of a phone not being interacted with something like fourteen times per hour (IIRC). As always, they carve themselves out.

Replying to @jasnell and @rektide

That makes no difference. By enabling geolocation Google starts tracking it constantly. See digitalcontentnext.org/blog/2018/08/2….

Replying to @seanmmcdonald

A guy can dream, Sean.

Replying to @jasnell and @rektide

I'm not sure what you're trying to say? It's profiteering from a pandemic to data grab whether it's done with the right hand directly or shoving towards the left.

Replying to @seanmmcdonald

I sometimes dream my research is wrong. That would be nice.

Replying to @BrianT_SEA

I have reservations about Apple on a number of topics, but actually on this one they're doing better than Google. Perhaps just because it fucks with Google, but it's still much better. Android is basically a data broadcasting device.

Replying to @BrianT_SEA

Sure, that's not great, again I'm not saying that Apple is great, just that on privacy they're a lot less evil than Google. But then again, Google's surveillance is as shameless as it is breathtaking.

Replying to @jkohlmann and @TimHerrera, @rachelsklar

Haha, I was trying not to get mad 😁

Replying to @johnwhegeman and @kevinroose, @crowdtangle

Thanks for the details. I am curious about a few things. * Isn't any content a tiny % of what most people see? * Is reach the number of people who might have seen an item if they happened to be there? * Are impressions just scrolling past an item or something more?

Replying to @jkohlmann and @TimHerrera, @rachelsklar

Ever since discovering that being tactful meant I could troll people more without being called out for it I've become much more tactful. I'm all about tact.

Replying to @huitseeker and @jenvalentino, @hartzog

The problem is not the logic of setting. The problem is that when location is on, Google collects it seemingly a minimum of 14 times per hour, which it never discloses, unlike Android does for others. They never include Google in their threat model, it's bad.

Replying to @bowlinearl and @jenvalentino, @eredmil1

This isn't actually hard. They could enable location for just an app without granting themselves the right to collect it too. But they chose not to.

Replying to @bowlinearl and @jenvalentino, @eredmil1

Why would it ever be OK for your system to transmit surveillance data? We're not talking about looking up SSIDs (which can be blinded) but about sending precise geo and gait.

Replying to @bowlinearl and @jenvalentino, @eredmil1

Sure, there are always tricky bits in such systems. But it's not like they tried hard and missed by a hair's breadth!
And at least two of them aren't dates!
Quoting a tweet by @msexcel ↗
891,772 cookies & 1,121 trackers for a bit of browsing makes it clear why browsers must stop this. But an adtech coalition is trying to prevent them (lists.w3.org/Archives/Publi…) and argues that tracking is so good there should be a universal ID for people (discourse.wicg.io/t/advertising-…).
Quoting a tweet by @WolfieChristl ↗
↺ 14

Replying to @catthekin and @thezedwards, @iab

It would also be easier if the @iab were much better at building tech solutions and weren't dominated by intermediaries who profit from this no matter what.
Publishers use a "standard" called ads.txt to identify themselves on ad exchanges. Thanks to a spectacular supply of incompetence from @iab and @IABTechLab, instead of being an actual identity scheme you just post whatever ID you want.
↺ 7
Apparently, public key cryptography didn't exist (checks notes) three years ago when this was invented. Neither did .well-known apparently.
↺ 3

Read the whole thread: 10 tweets →

Replying to @tobie

The whole industry is quite something 😁

Replying to @catthekin

I think your thread is more informative, I just thought I'd bring some gifs to the party 😁

Replying to @dmarti and @tobie

I pushed that in the @wicg_ thread but no takers!

Replying to @martingund and @EJGibney

Indeed! And that good work showing they did cookie syncing by default, too.

Replying to @TimHerrera and @NYTStyles

Easy, it's just like "gifs".
People think I'm being hyperbolic when I say that adtech is spectacularly bad. Look at how insanely easy it is to *steal money* meant to go from legit advertisers to legit publishers! And the people who built this and support it are getting rich abetting this fraud.
Quoting a tweet by @nandoodles ↗
↺ 10

Replying to @DG_III

There's some stuff starting at github.com/w3c/web-advert… but it's very early days!

Replying to @mnot and @mozilla

↺ 2

Replying to @martingund and @acfou, @EJGibney, @thezedwards

I don't think anyone knows what the exact implications will be just yet!

Replying to @orenizhii

I agree, while this is already a lot of work there’s a lot more to dig into.
I wrote an overview of how we approach privacy at @nytimes and of some of the things we've done to improve our practices. Naturally, it involves wine. open.nytimes.com/how-the-new-yo…
↺ 64

Replying to @bechgc and @nytimes

Those are not the funnest hangovers. (thank you!)

Replying to @EvanSelinger and @nytimes

Thanks Evan!

Replying to @DoraCrisan and @nytimes

Merci!

Replying to @JesseDamiani

Any time! Thanks a lot for reading.

Replying to @daniellecitron and @hartzog, @nytimes

Thanks Danielle! This is introductory, it's nothing you won't be familiar with, but I would very much welcome feedback if you have any!

Replying to @crenshaw_dev and @m4c9416, @nytimes

Thanks! We know there’s still a lot left to be done, we would love to work with others on this.

Replying to @rigow

Thanks Rigo! Technically, I guess I count as a EU data protection person who thinks they’re ahead ;-)

Replying to @maxloisel

Merci!

Replying to @Paul__Walsh and @lukemulks

I'd have to switch to French.

Replying to @null and @DSepDC, @nytimes

Thank you! I'd certainly be interested in reading your own take.

Replying to @null and @DSepDC, @nytimes

Also, don't hesitate to send questions!

Replying to @JulesPolonetsky and @hartzog, @nytimes

Thanks Jules!

Replying to @MarechalPhD and @nytimes

Thanks a lot! That's very kind.

Replying to @montezumachavez and @daniellecitron, @hartzog, @nytimes

Thanks a lot Luis! I don't disagree that control is valuable, but I think it is being overemphasised (or from the wrong angle) in most of today's discussion of privacy, at least in industry. To steal from @hartzog I believe it needs to be "rejuvenated".
The way most of digital privacy works today is that you do the thing people don't want by default and then you provide a cryptic setting to change it — then you can tell them they have privacy because they have control, and if they feel victimised they should blame themselves.
↺ 1
This doesn't seem people shouldn't have control, but in the same way that confidentiality/transparency/security become (again, from @hartzog) discretion/honesty/protection I wonder if there isn't a case to be made to switch from control to autonomy. Back to Belmont! :)
Sometimes autonomy is control, but a lot of the time it is not requiring you to control anything because technology did the right thing for you, on its own, so you have time and mental resources to dedicate to flourishing — which was the point of having tech in the first place.
↺ 2
Anyway — I could ramble for a while! Whenever someone suggests control as the solution, my team tends to ask: "how do we do this so they don't even need that?" It's useful, but I see it more as a fallback.

Thread of 5 tweets

Replying to @null and @podopie, @nytimes

Hahaha! When you get back let's storm a jail!

Replying to @montezumachavez

Yes, I believe that's common in German law. IIRC it was used to make nonbinary gender data fields compulsory because without that you can't self-determine your gender. I thought that was a great decision. But it's much closer to autonomy than to control!

Replying to @MrDanZak and @OKnox

In fairness, this is quality art. I'd appropriate federal funds on this basis. For effort if nothing else.

Replying to @palmitas10 and @carissaveliz, @nytimes

Thank you!

Replying to @epc_angela and @nytimes

Thanks a lot Angela!

Replying to @null and @amyvdh

Oh my god, so much goodness!

Replying to @null and @HNissenbaum

Thanks a lot, Helen. And indeed, the work is far from done!

Replying to @montezumachavez and @hartzog, @daniellecitron, @nytimes, @airavn

Digital privacy is a big beast and we all have a lot to learn, Luis! I haven't (yet!) read this article and it's my job too 😁

Replying to @jason_kint

That's a good piece! I guess we all must cater to our cultural tropes one way or another ;-)

Replying to @colinajones and @mnot, @mozilla

Your "even handed" assessment kicks off by ascribing intent to others that just isn't there. That's hardly unbiased (in addition to being wrong). YouTube's toxicity is a matter of sound research, not "fears and bias about evil".
I... I'm not sure how that works.

Replying to @colinajones and @mnot, @mozilla

Not at all! I don't care who uses it, and it wouldn't matter if YouTube were operated by a humanitarian commune in the Amazon. What matters is outcomes. It's not "YouTube bias". YouTube is known to damage society. This makes it stronger.
The times when we made APIs without considering their impact is long gone. I'm sure that Google engineers cover this in ethical training just like everyone else.
Now, as I told @mounirlamouri, I think this could be solved with a tie-in to policy. For sites with more than n million users, only expose this if they are following a set of best practices audited by a trustworthy third party.

Thread of 3 tweets

Replying to @colinajones and @mnot, @mozilla

Oh, wow, you're writing this from the 90s?
I have a joke about short-lived Twitter memes, but you've seen it already.

Replying to @internetdaniel and @nytimes

Thanks Daniel!

Replying to @svgeesus

Thanks a lot, Chris 😁

Replying to @torgo

Thanks Dan! Yeah, we've got work to do 😅

Replying to @mariafarrell

Does democracy even drive clicks when functional?

Replying to @mariafarrell

I think there is definitely something there, but I wonder if there isn't another parameter in the structure of the discussions. Democratic debates, when they work, should lead to consensus (at some time scale). This means they have more of a narrative structure than permafear.
That means they're monetisable but more by narrative media. I'm not sure they're all that monetisable by nonlinear firehose media.

Thread of 2 tweets

Replying to @mariafarrell

Hmmm, that's a very interesting angle, I like that! Also fragments are easier for recommendation systems (because they can iterate at a very granular level, which longform can't), and recommendation/micro-targeting are... two uses of the exact same thing.
I love your idea of loss of social/collective from the fragmentation of emotions & narratives. I think the virus metaphor is the wrong one, though. Different stories prosper in different environments. Thor: Ragnarok would struggle to find an audience as an Austen novel.
The SM narratives aren't inherently bad, but that's what survives best in the SM environment. Also, Dawkins killed virus metaphors for me. So, uhm, when's the book available?

Thread of 3 tweets

Replying to @mariafarrell

One other thing this made me think of is this idea from Sewell that historical events are those that transform the structures of daily life. So what happen when we no longer all see the same events? Or if the events lose structural richness because they're just short bursts?

Thread of 2 tweets

Replying to @v0max and @thezedwards, @TwitterSafety

Indeed, this is trivial, I'm not sure what the hold up is. These scams have been going on for quite a while. Just block tweets the domain and call it a day?

Replying to @v0max and @thezedwards, @TwitterSafety

Yes, that's what I had in mind. Run a search for messages sent more than eg. ten times, with that brand, a money amount, a link, and block them until a human approves them.

Replying to @v0max and @thezedwards, @TwitterSafety

Also, at this scale I’m guessing it’s either a compromised third-party app, or reused passwords, no? I wonder what’s preventing action on those?

Replying to @v0max and @thezedwards, @TwitterSafety

Yeah. This should be negligent, but there's CDA230.
This instantly iconic tee, designed by the timeless geniuses at @TheOnion, defined as much as defied fashion from Paris to Brooklyn through Tokyo for almost fifteen years. It even made TSA agents friendly. Farewell old friend.
↺ 1
I have started reading an early copy of this and so far it looks like an amazing book!
Quoting a tweet by @CarissaVeliz ↗
↺ 2

Replying to @AndrewM_Webb and @LucasVB

I don’t think I ever got a 100% answer, IIRC it has something to do with scientists doing LSD :)

Replying to @ashleyboyd and @nytimes

Thank you 😁

Replying to @nomadtechie and @tobie, @rachelnabors

Hi! Ha! I... I still have that song in my primary playlist!
The truth is that "Google Search" should more correctly be called "Search Google".
Quoting a tweet by @adrjeffries ↗
↺ 7

Replying to @nomadtechie and @tobie, @rachelnabors

That's the ho-o-o-o-orse!

Replying to @tobie and @nomadtechie, @rachelnabors, @ljharb

This is so beautiful, I'm tearing up.

Replying to @DanRose999 and @drose_999

I beg to differ, the keyboard is still the one thing from the Kindle that I miss everywhere else!
Oh my god this is how I feel all the time.
Quoting a tweet by @Chronotope ↗
And also the apps must be destroyed.
Quoting a tweet by @Chronotope ↗
↺ 2
Why stop at 99%? The stuff I'm trying to do is *so* interesting that it fails to pan out 100% of the time.
Quoting a tweet by @wtgowers ↗

Replying to @Chronotope

Replying to @puiwingtam

So... is there a sign-up sheet somewhere?
This may be one of the more underdiscussed topics in the data economy: forget privacy and discrimination, surveillance technology is also great for corporate espionage. Do you know how much data browsers and phones are collecting about your business?
Quoting a tweet by @adrjeffries ↗
↺ 1

Replying to @dsearls and @emilybell, @themarkup

That's because the hearing questions are for show. The ones you're looking for are more likely to show up in Questions for the Record.

Replying to @TimHerrera and @cwarzel

Note that there are two parts: the hearing, which is there to put on a show and draw attention, and the QFR, which need to be answered diligently and in writing, and serve the investigations.
↺ 1

Replying to @TimHerrera and @cwarzel

I found out, like, three days ago while looking at the questions so now I can show off my expertise to my Internet friends.

Replying to @swodinsky and @TimHerrera, @cwarzel

Write the QFRs you’d like to see! They’re due tomorrow :-D

Replying to @temptoetiam

Pourquoi.

Replying to @null and @funnymonkey

Also, "Bill" is a pretty crap unique identifier (no offence!). PII can often be less identifying than "anonymous" identifiers.

Replying to @tabatkins

Any chance SciHub covers IEEE?

Replying to @histoftech and @logic_magazine

Do you know what the diagram is on the board? Is it a network from before they were drawn as clouds?

Replying to @histoftech and @logic_magazine

I love this, thank you!
Strongly agree with this. The only thing you get from a focus on transparency and control is learned helplessness. User empowerment starts with privacy by default.
Quoting a tweet by @random_walker ↗
↺ 3

Replying to @publictorsten

Exactly.

Replying to @ShiraOvide

Infrastructure week keeps getting bolder.
Anyone else getting voicemail from lawyers "specialised in W3C and also JavaScript engines"? I wonder if it's the same kind of people who derailed DNT trying to do the same for WebAdv?
↺ 1

Thread of 2 tweets