December 2018

180 tweets

There's a novel involving entrepreneurs in ancient Sumer peddling Initial Clay Offerings in there. t.co/aYXnck9A1X

Replying to @johnwilander

When can we read it?

Replying to @gvoisin and @sas_assion, @sassion_en, @IABEurope

With pleasure!

Replying to @jenvalentino

If you're doing that for work don't forget we have a team of folks who know a lot about privacy policies and can help!

Replying to @null and @SimonDeDeo

The new wheelpunk genre where people in sumerspace hack tablets (to pieces).

Replying to @johnwilander

Good luck!

Replying to @jenvalentino

You never know, I sometimes read regulatory decisions for fun 😏 But PPs are tedious affairs indeed, they are wrought with dark art; one's instincts will feel the evil magics and try to ward you away.

Replying to @jonibrennan and @okeefekat

I beg to differ. At times it's consent, but relying on consent systematically is taxing to users & that will fast erode their agency. Sometimes privacy is being responsible *for* others. I highly recommend Helen Nissenbaum on this, too. Cc @SarahJamieLewis hbr.org/2018/09/stop-t…

Replying to @mcnees

I'm afraid to look up what people buy raccoon baculums for. Very afraid.

Replying to @histoftech and @zephoria, @kdphd, @ubiquity75

I've been wondering if they were planning to merge with NIPS at some point... 😏

Replying to @SarahJamieLewis and @jonibrennan, @okeefekat

But it's consent really the best tool to reduce dataflows? To take a concrete example, as you know the dominant browsers are actively privacy-hostile. Would fixing that not already do more than asking all the users in the world for consent?
I feel that people shouldn't even have to think about whether they their browser is a monitoring device. But maybe that fits in your view of consent? It's an annoyingly overloaded word.

Thread of 2 tweets

Replying to @TheAnnaGat and @polite_netizen, @latimes

You need much more than lawyers, much of the work is operational. I'm sure they'll be back eventually.

Replying to @null and @polite_netizen, @TheAnnaGat, @latimes

No, it’s more complicated than that. Greater privacy in the ad ecosystem actually favours first-party publishers (like the LAT) over Google and Facebook as it returns the premium to context over surveillance.

Replying to @null and @polite_netizen, @TheAnnaGat, @latimes

I can’t speak for the @latimes but the push to GDPR happened at the same time as they were being sold by Tronc. It could be that starting a major project in the middle of an acquisition was ruled out, and then they needed time to set up a new strategy.
I'm looking for solid, pragmatic, non-BS book recommendations around design thinking. Any ideas dearest twitters?

Replying to @LAM_Barrett and @hartzog

Thanks Lindsey — I’m jealous, I want my own bat signal now ;-)
Breakfast meetings, or how to most effectively discriminate against working parents.
↺ 2

Replying to @MichelleRobbins

Thank you, that’s definitely a useful source!

Replying to @skejtas

Thanks! Getting that one for sure :)
I hope this means I will never hear «but Google has done so much for security!» ever again. They took your call and text log, and they just gave it to anyone. For two billion users.
↺ 7

Thread of 3 tweets

This is what moving the Overton Window looks like, folks. nytimes.com/2018/12/05/opi…

Replying to @trekonomics

That describes most of the presidents and prime ministers in my lifetime.

Replying to @trekonomics

Is it, or is it just they we give up faster because we've seen it? Sarkozy and Hollande ran the country as if it were a small local politics dominion, Chirac and at least Mitterand's second term they were in the wrong century. All in over their heads.

Replying to @trekonomics

What would the exceptions be? Maybe Jospin? Maybe Rocard?

Replying to @null and @maritza_johnson, @sedyst

I agree, I noted the same thing. Not that it makes things any less creepy!

Replying to @sedyst and @maritza_johnson

Indeed, the tide has turned!

Replying to @jaffathecake and @HenrikJoreteg

Yeah. I just built an app, it doesn't need to be crazy fast as it's internal only so I was just minimally careful to make sure it didn't blow up. I can't seem to get it to grow much larger than 30k...
“Ethics are not enough” — in which @tmcw is entirely right. (via @femalegazebot) macwright.org/2018/05/10/eth…
↺ 2

Replying to @SirajAHashmi

You good.
The worst part is that that's one of the nicest things she says in that interview. I mean, she compares having the UK in EFTA to inviting a drink-spiker to your Christmas party. Sadly deserved, but ouch.
Quoting a tweet by @Channel4News ↗
↺ 1

Replying to @Chronotope

I won't spoil it for you but the end of the season almost gave me hope for that show. I say almost because they went where I was hoping they would (you'll agree) but at the very last minute they seem to hesitate to have the actual courage to do it for real.
Yes, there are only 6,999 others like you!
Yes, there are only 6,999 others like you!
↺ 1
Uber, but for naps.
↺ 2

Replying to @TheAnnaGat

I don't know how many books I read, just that they don't fit anywhere... I rarely use YouTube, YouTube is one of the worst parts of the Web.
The Internet Dunbar Number is the number of people for whom you can remember which messaging platform to reach them on. Recent estimates place its limit around 8.

Replying to @TheAnnaGat and @SimonDeDeo

You can teach non-CS and still be in the Californian ideology.

Replying to @TheAnnaGat and @SimonDeDeo

Indeed, but properly parsing out one's underlying assumptions is a particularly neat thing to do when working on education, no?

Replying to @fvsch and @Chronotope

I might be missing something, but can't the extension hit Facebook's API to get your friends list and hide any post that isn't from someone there?
Do you feel better knowing your location is only tracked several times an hour instead of continuously? How often do you change location? The root cause here is that Apple and Google simply don't care about putting you at risk of stalking. They'll tell *anyone* where you are.
↺ 3
The classic defence is to claim this data is "anonymous". Is it okay if I know where you live and work, who you hang out with, what time you wake up at, which doctors you go to, which gatherings you attend... but not your name?
One thing I'd like to know is how much this data costs. I suspect a lot less than you'd think. If I were into organised crime, I'd see a way of making burglary a lot more effective.
If I worked in terror... several of those yellow dots must occasionally spend a couple hours at a hotel or house without their spouse, or must visit a casino, or AA/NA… Finding targets at such a facility used to be a major surveillance op; now it's off the shelf!
I don't have funny GIFs for this thread. The situation defies alarmist characterisation. Adtech has been promising self-regulation for decades now. Who out there still believes them?
↺ 3

Thread of 6 tweets

Who said ethics have to be hard? Some solid rules of thumb from @Chronotope.
Quoting a tweet by @Chronotope ↗

Replying to @null and @nhoizey, @jenvalentino, @natashanyt, @mhkeller, @Aaron_Krolik, @satariano

Agreed!

Replying to @jaffathecake

I half expected him to start crushing some skulls with it. I mean, it *is* a mace. Making that available in the middle of disagreeing people seems to be begging for an accident to happen.

Replying to @AmeliasBrain and @jaffathecake

I’m sort of tempted to apply for that job.

Replying to @santoshc123 and @__apf__

It’s a wild guess, but I would assume that mobile-specific sites shifted to HTTPS more slowly than the rest.

Replying to @santoshc123 and @__apf__

Having a mobile-specific site is already a sign of lesser technical advancement.
I’m particularly interested in item (2) below. Every family deserves dad jokes, so it should be a more inclusive term. I’m struggling to find a good gender-neutral one, though the motherly version should be a “bon mom”.
Quoting a tweet by @kzwa ↗

Replying to @kzwa

Having followed you on Twitter for a few years now I have to say that this fact does not entirely surprise me.

Replying to @kzwa

A “begeddit”. As in when one of your begetters says “geddit?”.

Replying to @scottjshapiro and @LogicalAnalysis

My high school economics teacher: "You'll have no trouble finding work later, but it will be difficult to find someone willing to pay you for it."
↺ 1

Replying to @Cennydd

Also, "They said 'bring your +1', so I figured I would really bring the +1."
I don't understand the "crisis of meaning because decline of organised religion" people. I can't even find time for all the meaning there already is...
Quoting a tweet by @trekonomics ↗
↺ 1
Les #GiletsJaunes arrivent à New York! 🗽

Replying to @martinmuzatko and @w3c, @tabatkins

I don’t know why they dropped it, if you’re interested I would reach out to the group.

Replying to @martinmuzatko and @w3c, @tabatkins

More cynically, the process is: • Propose thing • Devs love it • Browser folks don’t see the point, won’t engage • Drop it • Time passes • Browser folks rant about why JS devs keep adding all these libraries that make things slow.

Replying to @JakeAnbinder and @LogicalAnalysis

How else could you know where to order your own copy from?

Replying to @rachelandrew and @torgo

The key word there is "just". "You can just", "Why don't you just" are all admissions of ignorance. It's tempting to call it "stupidity signaling".
You're no doubt used to arguments about ads being an active and growing threat to privacy — which is true — but Julien asks a different question: are ads even good business? He makes a good case that they aren't.
To begin with, the adtech world is spectacularly bad at effective self-regulation. Because of that, it has failed to enforce effective control over the simple problem of ad resource consumption. Without even looking at privacy, adtech is basically encouraging ad blockers.

Read the whole thread: 9 tweets →

I'll skip over debatable points about accountability under the TCF (saying Vectaury violated terms of the TCF is meaningless without also explaining why the TCF failed to be enforced) and focus on just one item that I find surprising…
↺ 1
They list what TCF controllers need to do in order to comply with the @CNIL recommendation, including: • You can't pre-tick purposes. • You can't mislead users. I agree. What I find surprising is that they draw no consequences from those requirements.
↺ 1

Read the whole thread: 7 tweets →

Replying to @RickByers

Those are certainly interesting arguments, but I am not sure that I am convinced. Or perhaps I feel that they are naïve in their own way too :) First, the commoditisation of privacy isn't a progressive tax, at best it's flat, and it could become regressive.

Replying to @RickByers

By which I mean: only the rich will be able to afford privacy, the rest will have to have everything about them tracked so that they can afford anything at all. We're really only a few wrong turns away from that, it's not science fiction.

Replying to @RickByers

Second: advertising is highly inefficient. If it were a redistribution from companies with products to companies with attention, you might have at least part of a point. But most of the profit goes to the intermediaries. It's a whole lot of parasites with little value.

Replying to @RickByers

Finally, other business models (eg. even just subscription to be simple!) can do price differentiation for instance by geography. In fact they often do. I'm not saying it's perfect or easy, but I don't buy that ads are the best option there. No bandwidth anyway!

Replying to @RickByers

I don't believe they've shown that at all. When you dig into user experience and expectations it's obvious they have no idea of the scale of what's going on.

Replying to @RickByers

It varies *a lot* depending on what you’re serving, what type of targeting, etc. Report coming out soon shows that behavioural targeting brings 4% bump to publishers but 500% bump to marketers. Who’s getting the difference?

Replying to @RickByers

I am precisely *not* vilifying all advertising. On the contrary, I am of a small group who think that another way of doing ads is entirely possible — and that we should not just because it’s right, but because it’s sounds business.

Replying to @RickByers

That’s because they’re not consent dialogs, they’re dark patterns meant to give an impression of consent. And the law has pointed that out in several cases already. Stuff in this article is consented to as “Personalisation”… nytimes.com/interactive/20…

Replying to @johnwilander and @RickByers

And what is remember should pass the “Shop Test”: roughly, it’s okay to be targeted/remembered/served in the same way that a shop owner would if we went there regularly, ie. just from clear and direct interactions with them, decaying over time.

Replying to @RickByers and @johnwilander

That’s waaay too underspecified a problem! There are many ways to personalise ads, some are privacy safe and some are not. I’m okay with those that are safe. I do not, however, think that privacy should be monetised.

Replying to @RickByers and @johnwilander

That… that is the world we’re getting from the *lack* of privacy.

Replying to @RickByers

I’m super happy about ITP2. The lack of privacy helps Google and Facebook to the detriment of publishers and a free press. Publishers actually want better privacy.

Replying to @RickByers

I would like it to be, I'm still looking for someone on our end to join.

Replying to @RickByers

Ah, but it does not cost 4%. That is what it would cost if it applied to one publisher and not others (actually much more due to bundle effects). But since it applies to all equally, and ad budgets don't vanish, it does not adversely impact us.
On the contrary, the less tracking there is the more valuable our inventory as it becomes valued on content quality rather than surveillance-based targeting. We honestly cannot wait for Chrome to start protecting its users' privacy!

Thread of 2 tweets

Replying to @RickByers and @johnwilander

I suggest "The Internet Trap" by @MattHindman if you're looking for a far more sobering view grounded in solid research.
↺ 1

Replying to @RickByers

Please meet @jason_kint !

Replying to @null and @SimonDeDeo

I somehow saw it in a bookshop and read it as my first "serious" computer book after having taught myself barely enough to be dangerous. It was interesting (if bloody drab) but it didn't teach me much that ever came in handy, and I wouldn't recommend doing the same!

Thread of 2 tweets

Replying to @torgo and @AroraXD, @ThisIsJoFrank, @Lady_Ada_King, @ltulip, @jeffbarr

Niiiiiice! Way to go!

Replying to @JedBracy and @PrivacyPros

Hi Jed! Thanks for asking. I could be, assuming confirmation, but it couldn't be before 2019. It's fine if you think that's too late!
I completely agree with this, but I'd love to understand why @nytopinion supports de-indexing data brokerage but opposes de-indexing irrelevant past convictions, often as harmful if not more.
Quoting a tweet by @sarahjeong ↗

Replying to @supergovernance and @pixabay

Better than if it returned sticks.
Two things are missing from this (good) @Recode article: * Is this not an admission of GDPR noncompliance? * Facebook is actually strengthening its browsing history tracking game by switching to first-party tracking so as to circumvent ITP.
Quoting a tweet by @Recode ↗

Replying to @Chronotope and @GenePark

I think the baffling thing is that Wikipedia happened at all. I don’t believe anyone has managed to reproduce anything even vaguely like it.
I'm running low on popcorn. 🍿 "As Facebook Raised a Privacy Wall, It Carved an Opening for Tech Giants" nyti.ms/2GqnbC9
↺ 1

Replying to @LAM_Barrett

Glad I'm not the only one to go into automatic eye roll when I read that!

Replying to @LAM_Barrett

I think we need to align the public's definition of selling personal data with that of the CCPA.

Replying to @LAM_Barrett

What's even worse than providing that level of access is that I actually believe those companies when they say they didn't ask for it or even knew they had it. Someone at FB just switched that on on a whim, and never switched it off.

Replying to @LAM_Barrett

The carelessness is almost worse than the deliberate misdeeds. Their tool must have had no oversight. I envision a UI in which it's faster to "Select All" permissions than to pick specific ones.
Ooh, this is a cool new one to try! I love developing on the @_developit stack; already using preact, unistore, and microbundle, and JS is joy again.
Quoting a tweet by @_developit ↗

Replying to @brianschatz and @getyrtrouserson

I don't know, with a few fixes (*slaps top of his GDPR printout*) this bad boy can fit so much privacy.

Replying to @_developit

I mean it! I was dreading having to do some client-side stuff again but that stuff brought me back. Simple, fast, orthogonal. It feels like using JS rather than being shoehorned into a lib. Thank you, and to the others!

Replying to @nickconfessore

I would contend that's actually selling (under the CCPA it is). Providing in exchange for an advantage is selling even if the advantage is not monetary. Same as it would be for instance for bribery.
↺ 1

Replying to @LAM_Barrett

Amazing gif game there, Lindsey!

Replying to @LogicalAnalysis

Well it would have to be transitive, right? Influence one person who does something big is worth more than influencing many who do nothing. From there you're straight into the fun madness that comes with the eigenmorality framework 😁
Can we get, like, 24 whole damn hours without a Facebook apology?
Quoting a tweet by @WolfieChristl ↗
↺ 1

Replying to @fatemehx2 and @zeynep, @JustinBrookman, @AnnCavoukian

Privacy by Disregard.

Replying to @fatemehx2 and @JustinBrookman, @JulesPolonetsky, @jason_kint

I would be very curious as to whether it varies with retention (using location on the spot vs by knowing your every move), number/type of intermediaries, how much the data is reshared, etc.

Replying to @jason_kint and @fatemehx2, @JustinBrookman, @JulesPolonetsky

Yes, relevance is a problematic framing, you can't even ask the reverse question ("Would you do X to get more irrelevant ads?") so it has to skew positive. And how much more relevant? Is it still worth it if one in twenty ads is better?

Replying to @fatemehx2 and @jason_kint, @JustinBrookman, @JulesPolonetsky

Yes, I have zero concerns over methodology, only conceptually about whether relevance is the right concept to roll up to. It feels hopelessly skewed, but that might just be from me trying to extract some truth from too many adtech companies. They use it like punctuation.

Replying to @fatemehx2 and @JustinBrookman, @JulesPolonetsky, @jason_kint

Yes please! If you're up for coming to give a talk at The Times I know we would have a lot of interested people ☺️

Replying to @profcarroll and @jason_kint

Frankly I think even sheep expected more from Facebook at this point.

Replying to @binarybits and @duncanjwatts

The what and how matter. To take a basic example, until a few months ago when you logged in to a site using your Facebook account that site received your gender and age range without ever asking for it. Users certainly did not expect that.
Facebook was just throwing around user data to try to gain an edge in online identity, because lots of marketers want those demographics so it's an incentive to favour Facebook.
Having users use their data is evidently fine, on fact right now Facebook does not let users get all of their own data back. Throwing around people's data without their say-so for competitive advantage is unethical, and in some places illegal.

Thread of 4 tweets

Just because you're Advocate General to the Court of Justice of the European Union doesn't mean you can't have a sense of humour. Read Michal Bobek's opinion on embbedding Facebook: eur-lex.europa.eu/legal-content/…
Come work for the @nytimes, it’s a wonderful place to work at! If you want to hack on cool stuff *and* be part of the newsroom, you should really look these two up: • nytimes.wd5.myworkdayjobs.com/NYT/job/New-Yo… • nytimes.wd5.myworkdayjobs.com/NYT/job/New-Yo…

Replying to @karlgroves and @nytimes

Failing so hard we're having a hard time recruiting fast enough!

Replying to @null and @ZombiePaisible

J'adore les ekranoplan. Je ne sais pas si c'est possible quelque part en touriste.

Replying to @null and @ZombiePaisible

COOL
"Future Ethics", by @Cennydd. I could spend a while telling you how good this is, but frankly just go read it! If you're in tech, data, product, adtech, marketing of course, but also if you're in philosophy. This is what ethics in the trenches looks like.
↺ 2
A deontologist is just someone who's never had to wait hours for a delayed flight alone with two young, tired children amid throngs of grouching French people in an overheated airport.
Flying away from a shutdown to potential riots, questioning my life choices.

Replying to @ppk

How right that it could be worse ☺️
Ah, la joie de voyager avec @airfrance. Tu as 4 places parce qu'ils ont refusé de t'en rembourser une, mais comme le quatrième voyageur n'est pas là ils réquisitionnent son siège. Là pour prendre l'argent, partis pour le service. Un bon vol, en somme.

Replying to @null and @SimonDeDeo, @TheAtlantic

Like anyone is going to say no ☺️

Replying to @anablzk and @Blinis, @SakumiBLR, @econoclaste

Yes, please!

Replying to @WKCosmo and @skdh

I'm not a sociologist but I don't think that's controversial. What would an incentive-free cooperative structure even look like?
In computer engineering we call that "computer science".
Quoting a tweet by @Hannah_Chutzpah ↗
↺ 3

Replying to @RickByers and @johnwilander, @stratechery

I think that article significantly oversimplifies what can be put under "open" (it's not a one-dimensional value), it also misses the importance of purpose limitations under the GDPR. The differential profit point is correct but it's what we already see, it's not new!
↺ 1
"Coin Locker Babies", by Ryu Murakami. I usually find growing-up/coming-of-age stories boring, but this one is very enjoyably fucked up (and excellently written).
While we're here: please stop comparing Ryu and Haruki Murakami. Two brilliant Japanese authors have the same family name. Get over it.

Thread of 2 tweets

Replying to @TheAnnaGat

You coming to Paris?

Replying to @TheAnnaGat

Ah, c'est pas tout à fait pareil!

Replying to @TheAnnaGat

Tu m'en vois désolé! Si tu passes par Paris au retour fais moi signe!

Replying to @RickByers and @johnwilander, @stratechery

So, how do you see this translating to data protection as it applies to behaviour for instance?

Replying to @RickByers and @johnwilander, @stratechery

I think that depends on what you see a tradeoff; we haven't seen that many but you could construe some choices as tradeoffs. The exemplar is browsers have a choice between favouring first-party+user expectations or third-party+market concentration (as Chrome and Edge do).

Thread of 2 tweets

Replying to @RickByers and @johnwilander, @stratechery

Part of the central problem though is that the most important point of control - the browser, the OS - are by volume most often working for those who benefit from the sharing. That's not a balanced situation by any reckoning.

Replying to @RickByers and @johnwilander, @stratechery

Much of my day-to-day is occupied with privacy issues that aren't black-and-white, I'm well aware that it's not simple in many cases! But for this specific case I have yet to hear an argument, even tweet length. I don't see the analogy with the @stratechery piece, help me?

Replying to @RickByers and @johnwilander, @stratechery

Sorry, who is benefiting?

Replying to @v0max and @PayPal, @Target, @GiftCards_com

Are these all sharing data?

Replying to @v0max and @PayPal, @Target, @GiftCards_com

I had no idea gift cards presented such a threat ☺️

Replying to @LogicalAnalysis

Good, but it will take time and commitment.
I always hesitate to recommend this book. Much of the content is excellent, but Unger's half is painfully poorly written, it feels like a rambling set of brilliant but repetitive, unedited notes. (Smolin's is outstanding.) This great summary might help you decide!
Quoting a tweet by @LogicalAnalysis ↗

Replying to @laurettaland

If you believe in set gender roles then you believe that men are just like that. You teach your daughters that it's the natural order of things and how they should navigate rather than fight it.

Replying to @LogicalAnalysis

It might be a weird way to look at it but do you need to get rid of it? Can you turn it into fuel, cancel out the painful bit and feed on the rest? I haven't been able to make it work with everything, but with dread yes. It has fire to it.

Replying to @LogicalAnalysis

It sounds trivial but I've been known to use the Litany Against Fear at times. Whatever works!

Replying to @skdh and @seanmcarroll

I'm guessing each toggles one side?
And it's time for that group to get together and hit reset.
Quoting a tweet by @Chronotope ↗
↺ 1
"Women Invent The Future", an anthology of short sci-fi stories by and about women, published by @doteveryoneuk. Consistently strong writing, I want to read these authors in longer form now! The book is almost free, too. (discovered via @histoftech)
↺ 3
Paris.

Replying to @0gust1

I can only wish 😊

Replying to @cyberbaloo_

La FAMILLE 👪 ☺️

Replying to @versac

Pareil 😘

Replying to @__apf__ and @sundress, @w3ctag

One vote per member!
I've been out of the loop in standards and not tracking W3C elections for a while now, but you'd have to be crazy not to want @sundress on the TAG. t.co/XnNKPJ38tn

Replying to @null and @sundress, @__apf__, @w3ctag

Oh wow, hearing of TV Raman really takes me back! Good luck Alice!

Replying to @_sen_sharmila and @skdh

"Revolutionary" should only be usable by writers able to produce proof that a monarch of some sort was beheaded in the process.

Replying to @null and @SimonDeDeo

I'm not sure that transcribing it as "culture/academia" is necessarily correct: it mentions *media*. There may be a sense that controlling the media is controlling all the things in a liberal democracy.

Thread of 2 tweets

Replying to @macsym

Ça va?

Replying to @macsym

Aïe, tu es fou 🤪! Bon, remets toi bien, bisou!
"In The Miso Soup", by Ryu Murakami. Follows a strange American and his young nightlife guide on a walk through the grottier side of the Tokyo sex industry; in turn meditative, surprising, and gruesome.
I liked: "If an online space makes more money the more time you spend on it, use sparingly."
Quoting a tweet by @kottke ↗
↺ 2