So. Many. Times.
Quoting a tweet by @michaelajawks ↗
♥ 2
September 2018
306 tweets
Replying to @davidbaron
Exactly!
Hey @JustinTrudeau, come join the EU; we can build a post-geographic world together! 🇨🇦🇪🇺
nyti.ms/2MDBjKR
♥ 4↺ 2
Replying to @sandhawke and @JustinTrudeau
Yes! It wouldn't be properly post-geographic without you!
Replying to @lukOlejnik and @benadida
I know, that's why I started looking initially. Having regulation riding on a failed standard would have been a really stupid idea, so seeing it very much alive is encouraging.
↺ 1
The thought crossed my mind earlier and now I can't shake the feeling that la Marseillaise would make an outstanding anthem for a House of the Klingon Empire. (Paging @trekonomics.)
♥ 1
This is the spokesperson for adtech once again telling journalism and its readers "it's our way or the highway."
Media will do much better when browsers stop enabling wiretapping.
Quoting a tweet by @dfgrimaldi ↗
♥ 36↺ 16
Solution: use YouTube Music's mainstreaming recommender algorithm for videos and the radicalising algorithm from Video for Music. Easy!
Quoting a tweet by @baekdal ↗
♥ 2↺ 1
Replying to @barefootliam and @RyanGermann
I know, but this really isn’t mostly trees, there’s a lot of overlap.
Replying to @ArtirKel and @TheAnnaGat
Related (French) classic: "S'il n'y a pas de solution, c'est qu'il n'y a pas de problème."
♥ 1↺ 1
A programme I can get behind.
♥ 2
Replying to @cpeterso
Indeed. There are ways to mitigate fingerprinting, and there's a strong case that it's illegal under the GDPR to boot. First party sharing also has some limits when it can't be backed with third party.
Replying to @JasonWilliamsNY
I know you're not all that far from where I live, I just need to figure out exactly where 😏
♥ 1
Replying to @simevidas and @cpeterso
It's hard to have subscriptions or metered free content without.
The Exponential View newsletter is usually good, but this week’s edited by @JeniT was outstanding. Thanks Jeni! exponentialview.co
♥ 5↺ 1
Replying to @barefootliam and @RyanGermann
Ha! But DAGs aren’t inherently messy :)
“A Crisis of Permissions”: Good stuff from @ThisIsJoFrank.
medium.com/samsung-intern…
♥ 4↺ 1
Replying to @trekonomics
Panspermia is the problem with a discovery of life "just" on Europa or Enceladus, though. The emergence of life could be extremely rare yet it could still have spread through these three over meteorites.
♥ 1
Replying to @trekonomics
Yeah, we wouldn’t know how to dig through that much ice here on Earth. I hope they have some bioluminescing signals on those vents!
♥ 1
“Because it's difficult for web users to keep track of who they're dealing with. Now, the Chrome team says it's time for a massive change.”
So, is that saying that Chrome is going to start protecting its users’ privacy?
Quoting a tweet by @__apf__ ↗
♥ 1↺ 1
Replying to @barefootliam and @RyanGermann
That’s news to me :)
Outstanding work showing that while some browsers and extensions do better than others, there are circumventions for all browser privacy protection mechanisms in play today: “Who left open the cookie jar?” blog.acolyer.org/2018/09/05/who…
♥ 1↺ 4
Replying to @_abialas and @femalegazebot
Anna! Don’t read Twitter during meetings!
Replying to @johnwilander
Yes, it’s unquestionably the superior option. And I’m not sure what to do with their listing of redirects. I mean, yes, they are certainly used for tracking but short of looking into the future it’s hard to defend against.
Replying to @johnwilander
Yes, I think they’re due for an update of their framework. Should be pretty straightforward though!
Replying to @IGPrestige
I don’t know, you should ask them!
The #OpEd is just a senior Republican patting themselves on the back for doing barely one tenth of what should be expected from a half-decent public servant.
And containing POTUS without authority to do so is a much graver constitutional issue than invoking the 25th.
♥ 11↺ 4
The timing is also ridiculously self-serving.
♥ 1
Accountability in action.
Quoting a tweet by @nytimes ↗
Replying to @CliffordSosis
Let the witch-hunt begin!
Replying to @AmeliasBrain and @DotProto, @_developit, @BrendanEich
Position tracking makes some sense if you run your regex in a loop (including in the test condition), which we used to do quite a lot back then.
♥ 1
Replying to @TzviyaSiegman
Phones? Where we’re going, we don’t need phones.
Replying to @TheAnnaGat
I read that! Where is it from? It's just so cruel. Note that the baby survives.
Replying to @TheAnnaGat
Hmm, that's not my source. I could swear I've read the exact same footnote.
Outstanding work: “Public By Default“, visual storytelling of just how much stuff you can guess from Venmo’s data being public by default.
publicbydefault.fyi
♥ 3↺ 1
Replying to @TheAnnaGat
Americans are reading this and not realising that they are forgetting Puerto Rico.
♥ 1
Replying to @lukOlejnik and @tobie, @slightlylate, @w3ctag, @kennethrohde, @sangwhanmoon, @torgo, @hadleybeeman, @plinss
What Alex said. What occupies you depends on your choices, you can also spend time writing (but the commitment would be the same).
♥ 4
Replying to @samccone and @slightlylate
1) Performance is hard.
2) It's a slippery commodity such that it tends to degrade in small increments that boil the frog.
Budgets help. Graphing performance of the codebase over the time can also be a great way to show the loss adding up.
♥ 7
This plays well with research showing that politically polarised Wikipedia pages produce high quality output. It doesn't work on social media because you don't collaborate there.
Quoting a tweet by @zeynep ↗
♥ 3
I'd love to see a @nytopinion experiment in which disagreeing writers have to write an article *together*.
♥ 4
Replying to @skdh
I'm a few months short of 42, but I know that feeling: I passed the same milestone fifteen years ago. Hang in there 🤗
And science needs some serious governance; it can be fixed!
♥ 1
Replying to @gchampeau
Pareil...
Do you use an outliner? Are there any good ones? I love the idea and outline constantly on paper, but the software versions I've tried have all seemed terrible.
Replying to @TheAnnaGat and @imcatnoone
Ce serait un plaisir de discuter, mais je doute que ce soit vraiment la plus belle langue. N'est-elle pas un peu trop nasale?
♥ 1
I am not much into biographies, but @platobooktour's "Incompleteness" is a great read, good writing unafraid of the technical. She makes even some of Gödel's weirder aspects make sense.
♥ 6↺ 1
Replying to @imcatnoone and @TheAnnaGat
There's a way to learn French really quickly (and Anna can confirm it's true): youtu.be/GSeaDQ6sPs0
♥ 1
Replying to @jeanqasaur and @Spotify
The GDPR places strong data protection requirements upon personal data controllers. Granting an access request with insufficient validation is indeed a breach of @Spotify's GDPR obligations. If I were them I'd be begging you not to sue.
♥ 4
Having said that it's hard, there's no standard of verification and you have to balance the risks to the person against not making the process so hard you're preventing them from exercising their rights.
This is also what most people utterly fail to understand about governance. You can't just make rules: you have to make them work.
Quoting a tweet by @Noahpinion ↗
♥ 3
Are there no kids' magazines in the US? If they exist why aren't they sold at airports?
♥ 1
Replying to @JeniT
I highly recommend muting the agressivity. That's not an approach to expanding one's bubble that works anyway...
♥ 1
Replying to @tobie
Frictionful.
Replying to @benwerd
I thought there had to be. But airports never carry any. It's like there's a gap in the very fabric of capitalism.
♥ 1
Replying to @benwerd
Nope! They systematically have three or four NG variants, but not for kids.
Replying to @fantasai and @benwerd
I can't seem to find their tracker on GitHub...
Is there a good document on the complexity of browser source code and how it compares to other large systems?
♥ 7↺ 2
Replying to @null and @accessiblestef
That sounds neat, though to be honest I'm not sure it's fair: the default calculator probably has more code than that mission!
♥ 1
Replying to @yoavweiss and @AmeliasBrain
It would be good to make SRI the expected default for third-party tracking. The GTM folks could really help push for it.
♥ 1
Replying to @podehaye
Actually, by contract they usually claim to be data controllers. In either case the processor/controller distinction is established by facts and not by contract so it wouldn’t matter: they’ll be controllers.
Replying to @JibberJim and @yoavweiss, @AmeliasBrain
If by users you mean embedding sites, I would definitely reach out to at least some of those (finding all might be painful).
Replying to @yoavweiss and @JibberJim, @AmeliasBrain
Signature-based SRI would be great; but to be honest if embedded third parties were forced to communicate with us when they update their code (and explain what the change is to motivate us to upgrade) that would be a feature. A big feature, actually.
Excellent work by @jenvalentino, @natashanyt, @Aaron_Krolik, et al. about seemingly common COPPA violations by Google, Twitter, and a host of apps.
Gotta track them right from the crib!
nytimes.com/interactive/20…
♥ 2
Replying to @skdh
Part of the problem is that package deals help to block new entrants (who can't make attractive packages), leading to monopolistic issues (high margins but also low quality of service, that software hasn't been upgraded). I wonder if a co-op model wouldn't work for packages.
♥ 1
The lack of standards also doesn't help, it means that you have to get the full pipeline of services from a single source instead of going to the best for each step. That's fixable but publishers have little incentive. Regulation maybe?
♥ 1
People also don't help writing in poorly-engineered old formats: it costs a fortune to reliably get production-quality output from Word or LaTeX.
♥ 1
I’m very encouraged to see the progress of the /e/ mobile operating system, it’s moving much faster than I thought it would. I hope to load it up on a test device soon! hackernoon.com/leaving-apple-…
♥ 1
Replying to @LeonieWatson and @ChanceyFleet, @ugly_chimp, @lukOlejnik
Are you replying to something? I can’t see it.
Replying to @null and @SimonDeDeo, @TheAnnaGat
Using villain and monster names might be a neat fix.
♥ 1
Replying to @TheAnnaGat and @SimonDeDeo
Oh yeah, or Goldfinger, Thanos, Glory, Zuckerberg… The headlines just write themselves.
Replying to @JasonWilliamsNY and @SimonDeDeo, @TheAnnaGat
I’m not entirely sure that trying to catch them all is altogether in line with what I originally had in mind, Jason.
♥ 1
Replying to @LeonieWatson and @ChanceyFleet, @ugly_chimp, @lukOlejnik
Ah, that’s a good question — and the answer is, of course, “it depends”.
If it’s detected client-side (or with very minimal processing, eg. a redirect) I don’t see that it would constitute a violation. At least, the case would be hard to make, I would think.
Replying to @LeonieWatson and @ChanceyFleet, @ugly_chimp, @lukOlejnik
If, however, that information were stored or used in any way beyond just adapting the site’s behaviour then that would be processing special-category data (here health/biometric) and doing that without the user’s express informed consent would be dodgy as hell, v. likely illegal.
Replying to @LeonieWatson and @ChanceyFleet, @ugly_chimp, @lukOlejnik
This of course does not detract from all the other reasons why it might be a bad idea anyway.
Replying to @lukOlejnik and @LeonieWatson, @ChanceyFleet, @ugly_chimp
I would have a hard time arguing that disability isn’t either health- or biometric-related information; I’m curious what your argument would be.
Replying to @lukOlejnik and @LeonieWatson, @ChanceyFleet, @ugly_chimp
I'm not sure I follow your distinction, it's Article 9 both ways, no?
Replying to @skdh and @PaulGuinnessy
That is true but in fairness they have high costs because they're not trying hard to be efficient. It's not a justification for the prices they practice.
♥ 3
Replying to @msantolini
I wish they'd picked a better name though. Not only is the "2.0" done to death, but the idea that this is just v2 is horribly anthropocentric.
This is an excellent point. I'm considering switching back to an iPhone for the privacy, even though I find the user experience poorly designed and the price excessive. It's a bit like going back to Linux, except at least the WiFi works without recompiling the kernel.
Quoting a tweet by @johnwilander ↗
♥ 7
Replying to @null and @dtemp_09f9
I know, it's much much better than ten years ago. I switched to OSX back then because I really can't be arsed to think about hardware.
♥ 1
Replying to @null and @ndw
How painful was it? I use intents to share things all the time, last I gave it a try on an iPhone it was a total train wreck.
Replying to @JibberJim and @LeonieWatson, @ChanceyFleet, @ugly_chimp, @lukOlejnik
I don't think that looking at it in too black & white terms helps. If your access logs are only processed as logs, accessible to a limited group, and have limited retention I would call it fine. The rights and freedoms are okay, as we say.
For behavioural profilers, the biggest issue is that they have access to the same information that allowed detection of a screen reader in the first place. That's scary. Browsers are a huge walking Article 25 violation, they should be taken to task over that.
♥ 1
Replying to @null and @ndw
Ok. That worries me a bit. I should probably try to borrow one for a day to see how much I want to throw it at the wall.
The dwarf planets have better names anyway.
Quoting a tweet by @elakdawalla ↗
My cat does not fear cucumbers. WHO DO I CALL? 🙀 🥒 😱
♥ 3
You can always trust @TheAnnaGat for creative solutions to unlikely problems.
Quoting a tweet by @TheAnnaGat ↗
♥ 2
Replying to @anthony_ricaud
Bondage, Discipline, and... Dinosaurs? 🦖
Replying to @karlgroves
She might just be crazy enough to fight back 😏
Replying to @karlgroves
That might be wise of it, cats are full of sharp endings.
Replying to @msantolini
But that's the thing: I don't think that current Gaia is self-aware, some various bits are. I'm also not sure that consciousness is relevant when comparing to terraforming like the Great Oxygenation Event or the 40 million years during which nothing could biodegrade trees!
I am also not convinced that self-awareness is better than self-regulation, we might want to look for ways to improve the latter seeing how poorly we're doing improving the former.
I feel the 2.0 label misses the opportunity to take a deep history view of the problem. Refreshing the Gaia concept is certainly useful, but I think the way to go is to make it deeper and to show just how many Gaias there have been - and how they survived.
Maybe @trekonomics has a view here? I'd love to hear.
♥ 1
The African Photojournalism Database is lit.
♥ 1
After a long day spent tilling the fields my family and I gather to splurge on our annual photograph.
♥ 19
Replying to @AmeliasBrain and @penumbranyc
Absolutely! All of our eyes have to be on the same plane otherwise it comes out blurry. It makes for a long and somewhat awkward pause/pose 😁
Replying to @erikphoel
First, I would refrain from considering social media as the Internet's final form. Social media is heavily rooted in an economic model that is very dependent on a set of technological assumptions that are a major ongoing battleground. Whoever carries that will change social media
To your question though, I think it helps foster multiple overlapping identities. I think the mainstream identities are more uniform (everyone makes cat jokes, worldwide), but the smaller ones are more diverse because they can have communities that were impossible before.
♥ 1
Replying to @erikphoel
Oh, it may well all be colonised but not by media in the current form.
Replying to @turboconnard and @mariejulien
Excellent!
Replying to @LeonieWatson and @JibberJim, @ChanceyFleet, @ugly_chimp, @lukOlejnik
Well yeah, but most browsers don't work for you and the ramifications are pretty scary pretty fast.
Replying to @turboconnard and @mariejulien
Ha, j'adore!
Replying to @null and @SimonDeDeo, @TheAnnaGat, @LarsESchonander, @webdevMason, @rivatez, @micsolana, @tylercowen, @nytdavidbrooks
"Mountain" sounds a bit drab, it might catch on better as "les Montagnards". You can almost see the knife held between gritted and grimly determined teeth.
♥ 1
Replying to @benwerd
I'm hesitating. On the one hand I have a novel idea that I could really outline in time, on the other there's another project I kinda want to do more...
Replying to @TheAnnaGat and @SimonDeDeo, @LarsESchonander, @webdevMason, @rivatez, @micsolana, @tylercowen, @nytdavidbrooks
You know, it's kinda tempting to write a script 😏
This might come in useful.
washingtonpost.com/news/answer-sh…
♥ 1
Replying to @TheAnnaGat
Wait. Is there anything that doesn't feel this way?
"Convenience Store Woman", by Sayaka Murata. A great short novel, I found it stomach-churningly relatable. It's a wonderful inside depiction of life amidst normal people when you're not quite the same.
♥ 3↺ 1
"Convenience Store Woman", by Sayaka Murata. A great short novel, I found it stomach-churningly relatable. It's a wonderful inside depiction of life amidst normal people when you're not quite the same.
♥ 8↺ 2
Replying to @null and @doctortovey
Let me know what you think!
"Causal Inference in Statistics", by @yudapearl, M. Glymour, @NP_Jewell. More technical than "The Book of Why", less daunting than "Causality". It covers a lot of ground clearly in little space, definitely worth reading!
♥ 12↺ 2
Replying to @benwerd
I go back and forth thinking Wikipedia should be that and thinking that it probably requires a different style.
If you see some deliberate disinformation being spread ahead of the US midterm elections, the @nytimes would like to hear from you through its dedicated tip line: nytimes.com/2018/09/17/tec…
♥ 8↺ 1
OMG THEY REMOVED THE BYLINES!!!
In… 1851.
Quoting a tweet by @NYTArchives ↗
Replying to @benwerd
People should be more Stoic, but not necessarily stoic.
Replying to @skdh
Analytics demographics are notoriously crap. They tend to infer based on online behaviour and there is little actual incentive to really improve them because marketers can only verify accuracy so well anyway. Before I killed tracking most thought I was a woman.
♥ 1
So I would not be surprised if it turned out that Google’s profiler were biased in very stupid ways, notably "interest in physics" -> man.
♥ 1
"How To Exercise The Power You Didn't Ask For", by @zittrain. A very good description of the fiduciary approach to privacy. It is worth noting that under the legitimate interest regime the #GDPR has many aspects of a fiduciary approach. hbr.org/2018/09/how-to…
♥ 6↺ 1
Companies went crazy with consent instead because they saw the chance to abdicate responsibility to the user (for processing they can't possibly understand). That won't last long. Prepare for what's coming. Set up a corporate IRB-equivalent, work your ethics muscle.
♥ 1
There are risks with this approach, notably that it be toothless. I certainly would not trust the current administration with anything involving ethics. I mean the thought itself is sadly comical. But the conversations should be started now. I know we have.
This sums it up: "Restaurants must meet minimum standards for cleanliness, or (ideally) they’ll be shut down. We don’t ask the public to research food safety before grabbing a bite and then to “opt out” of the dubious dining establishments. No one would rue being deprived of..."
♥ 1
...the choice to eat food contaminated with salmonella. Similar intervention is needed in the digital universe."
If you're interested, get in touch! The NYT data governance team loves to share!
♥ 2
Replying to @sandhawke
It does. Yes, and on the right scale: you can consent to small things deemed safe albeit imperfect.
♥ 1
I have friends who want to keep using @googlechrome but also are tired of @Google selling them downriver to advertisers. What extension do you recommend to make Chrome a modern user-first browser?
♥ 5↺ 3
Replying to @andreisambra and @googlechrome, @Google
But uBlock is an ad blocker. They don’t want to block ads, just tracking.
Replying to @dontclickonthat and @KristenLarski, @googlechrome, @Google
I use Firefox with TP enabled so I’m not too familiar with what Chrome users would do. @julien51 suggested Ghostery which looks like a good option indeed.
Replying to @thurrott and @stshank, @googlechrome, @Google
Switching browsers can be a PITA. I’m sticking to Firefox despite the loss of Tab Mix Plus making it almost unusable UI-wise, it’s still less painful than switching.
♥ 1↺ 1
Replying to @thurrott and @stshank, @googlechrome, @Google
But it’s mostly the UI that makes it harder to switch. Most users don’t even know what the rendering engine is — I don’t run a web team :)
Replying to @stshank and @thurrott, @googlechrome, @Google
Ah, @KristenLarski might be interested in this since she said she’d tried Brave but didn’t find it ready yet.
Replying to @dontclickonthat and @KristenLarski, @stshank, @thurrott, @googlechrome, @Google
Yes, a lot of Google’s “web” stuff is almost Chrome-only. They claim it’s not on purpose but it’s pretty obvious they’re not trying to fix that either.
Replying to @othermaciej and @googlechrome, @Google
Chrome is certainly very pro-tracking but I wasn’t aware that even extensions couldn’t fix that. Do you have specifics? That would be *very* interesting fodder.
Replying to @stshank and @thurrott, @googlechrome, @Google
I believe I’ve voted on most, but no one at Mozilla seems to care. Just a few days ago I saw that the tab ordering issue was classified P5 (Mozilla won’t do, community can).
Replying to @mikewest and @othermaciej, @googlechrome, @Google
Well at the very least it actively sustains third-party tracking — that much is deliberate support for tracking as a feature. But I’d be interested in hearing what Maciej had in mind regarding that not being extension-preventable.
♥ 2
Go team @nytimes!
“NYT sues FCC, says it hid evidence of Russia meddling in net neutrality repeal”
arstechnica.com/tech-policy/20…
♥ 2↺ 2
Replying to @othermaciej and @mikewest, @googlechrome, @Google
Wow, that’s so much more than what I thought there were, and I already thought it was bad. Thanks for the pointer, that will definitely prove useful.
♥ 1
Am I the only one reading this Chrome whitepaper and counting the GDPR violations? There is no world in which Google isn’t an independent data controller here with no legal basis to stand on. google.com/chrome/privacy…
♥ 12↺ 3
Replying to @othermaciej and @mikewest, @googlechrome, @Google
I love how on iOS they just throw a cookie in there with blatant disregard for just so much legislation 🤗 Has anyone from Google legal looked at this?
Michelle Alexander has started writing for @nytopinion and it is the smartest, truest thing on the Internet today:
"Donald Trump’s election represents a surge of resistance to this rapidly swelling river, an effort to build not just a wall but a dam.(...)
nyti.ms/2Dss8sq
♥ 4
Replying to @ehsanakhgari and @mikewest, @othermaciej, @googlechrome, @Google
You're entirely right, but at this point it is sadly clear that no matter how nice people on the Chrome team may be they really don't give the last damn about privacy. So as long as people want to use Chrome it's worth helping them at least a bit.
Replying to @othermaciej and @ehsanakhgari, @mikewest, @googlechrome, @Google
I know those people too, am fond of many. But every time I take a look at Chrome I'm ashamed that this is the Web to most people. I know these choices can be hard, I often wonder why no engineer whistleblew in the VW emissions scandal. No doubt some cared about the environment.
Replying to @othermaciej and @ehsanakhgari, @mikewest, @googlechrome, @Google
And even forgetting ethics it's almost certainly illegal. If they are being instructed to keep fighting privacy, or prevented from supporting it, that's a clear violation at least of GDPR Article 25. We have a tipline, I got Signal, it's all ready to help.
Replying to @null and @wiawm
That's true, but the Fabians were pretty cool. This is a great occasion for people to check them out!
Are any of you fine people attending the @w3c permissions and consent rigmarole next week? I'll be there!
♥ 7
Replying to @__apf__ and @matthew_d_green, @quinnnorton, @robocallaghan
Wait, you're saying that the Web login status and the browser login status are systematically in sync?
♥ 1
Replying to @null and @sneakdotberlin, @__apf__, @matthew_d_green, @quinnnorton, @robocallaghan
So the idea is that Chrome is treating Google Web properties preferentially, in the same way that Google Web properties are made to work just with Chrome. I guess it is a good time to milk the Trump administration's lack of interest in antitrust issues for all it's worth.
Yet more evidence that Chrome is giving preferential treatment to Google properties. Not even Microsoft was ever this brazenly cynical.
Quoting a tweet by @matthew_d_green ↗
♥ 6↺ 2
Best title for this affair so far.
"Art of the pussyble: quantum theory"
espresso.economist.com/bb4b90201e39e5…
♥ 1
Replying to @NaturePlants
Any chance that you could stop publishing articles in that way? It's a horrible experience on mobile and it's aggressively hostile to people with accessibility needs, for no obvious gain beyond the gimmick of mimicking paper.
Considering that *animal* life is over half a billion years old, I really hope that some form of the Silurian Hypothesis is true. The reverse is sobering and a touch depressing.
Quoting a tweet by @LogicalAnalysis ↗
"Asking companies whose business models revolve around exploiting data-based consumer-influence techniques to explain their privacy policies seems about as useful as asking sharks to hold forth on veganism." nytimes.com/2018/09/22/sun…
♥ 8↺ 2
came to chew gum and kick arse, and she's all out of gum.
♥ 3↺ 1
Replying to @ZachWeiner
I've been dying to find a good short story collection, the last were two (awesome) Murakami collections.
Replying to @tabatkins
That is entirely true. I mean, I reckon we could have another Bronze Age, but it would not involve mining, more finding our stuff. But that all assumes our path is the path. A conscious predecessor would likely have been aquatic: they'd be less about fire.
Replying to @PrivacyMatters and @newrepublic
I don't think that's fair. Media have a gun to their head: do it adtech's way (and serve as cover for them with users) or die. Heckling them won't change anything. The system is enabled exclusively by browsers and mobile platforms: that's the one key.
♥ 1↺ 1
Replying to @PrivacyMatters and @newrepublic
It is evidently entirely up to you to choose whether to be part of the solution or of the problem.
Replying to @lukOlejnik
Are you preparing a complaint?
Replying to @PrivacyMatters and @newrepublic, @newrepublic
took a stand for what's right despite this position presenting a direct existential threat to them; you're playing armchair purist on Twitter going after them for it. They've opened a much more useful conversation than you have.
♥ 1
Replying to @PrivacyMatters and @newrepublic
I'm not preaching, I'm pointing out that you're policing someone trying to help. And if you don't understand why they are facing the contradictions they are I don't see how you can help.
Replying to @PrivacyMatters
All I'm saying is that @newrepublic is a victim speaking up. Blaming them does not help.
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
Well Alexander, this was a polite discussion until you joined. I didn't say he had done nothing more than armchair purism, only that that's what I was seeing right there in that thread. I guess that makes for less grandstanding though.
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
I'm not new to privacy, either. And yes: I'm tired of seeing otherwise highly intelligent advocates attacking the symptoms and giving the root cause a pass.
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
As for my job, I might not have "the balls" to "stand up" and make revenue out of thin air as you seem to expect but this isn't my first rodeo and I have reason to hope that when I'm done we'll all have much better privacy.
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
And how much did that help?
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
Where did I insult him?
Replying to @podehaye and @alexanderhanff, @PrivacyMatters, @newrepublic
That is a useful question to gauge agreement on, do you mind if I respond in a quote tweet?
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
Again, read back, it was entirely within the context of this conversation, not some lifetime achievement award in armchair purism.
If you have feedback beyond testicle-related injunctions I'd love to hear it, otherwise I suggest we both have a better Sunday elsewhere.
Replying to @PrivacyMatters and @alexanderhanff, @podehaye, @newrepublic
I'm sorry if I hurt your feelings, its was not my intent. I just sincerely do not think you're helping by attacking people who are trying to help, no matter how imperfect their current situation may be.
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
Yet again, I did not mean to insult and apologise de if I did. But maybe spending five minutes understanding the situation of media would help more to find a solution? Calling it hypocrisy shows a clear lack of understanding.
Replying to @PrivacyMatters and @alexanderhanff, @podehaye, @newrepublic
The left picture is a conversation that will change the right picture. It's not the only approach, but it's one prong. I don't have magic to make that move faster.
Replying to @podehaye and @alexanderhanff, @PrivacyMatters, @newrepublic
Cool, I'll get to that after lunch.
Replying to @alexanderhanff and @podehaye, @PrivacyMatters, @newrepublic
I have 25 working on the Web, nevertheless sometimes I'm wrong and I don't take it as an insult when people think I may have missed something. Have a great Sunday!
Replying to @PrivacyMatters and @alexanderhanff, @podehaye, @newrepublic
We don't need more time, but rather focus on what will effect actual change. For consent I'm looking at ways to bring DNT back to life. It'll make people laugh for a while before it work, though 😏
Replying to @alexanderhanff and @McDapper, @PrivacyMatters, @podehaye, @newrepublic
Contextual is what we want to bring back. For that to happen we need to fix the current game theoretic structure of the ad industry. I'll expand more on that when I get home.
Replying to @alexanderhanff and @PrivacyMatters, @podehaye, @newrepublic
What are you proposing that's better? AdTech actually pushed hard to make sure that DNT would be a meaningful signal rather than a default. I cannot wait to use that to bite back.
Replying to @podehaye and @alexanderhanff, @McDapper, @PrivacyMatters, @newrepublic
I do not believe for the one second that the press needs an exception regime. I do however believe that without a free press we won't be have any kind of privacy for long. We *absolutely* can make this work and the press must help. Attacking it just solves nothing IMHO.
Replying to @alexanderhanff and @PrivacyMatters, @McDapper, @podehaye, @newrepublic
Journalism has been ad-supported for a long time. Then the ad industry changed, and I agree no one reacted fast or well enough. But now we have the means to change it back.
Replying to @podehaye and @PrivacyMatters, @alexanderhanff, @newrepublic
I think we're going helping the same person 😊 I invite what's going on, that's why I have such a strong view on what needs to change.
Replying to @alexanderhanff and @PrivacyMatters, @podehaye, @newrepublic, @PrivacySurgeon
I would be happy to generalise blocking lists — but why do that when we could just basically kill third-party cookies?
This is a good question: what is the root cause of the lack of privacy online today? Why does media track so much? My personal take is that it boils down to browsers and mobile platforms. Not only is that the master fix, but it is within reach. Follow the thread❗👇
Quoting a tweet by @podehaye ↗
♥ 23↺ 13
First, I have some assumptions that I want to ensure you know:
1) Without a free and well-funded press before long we'd have no privacy at all. This does NOT justify an exception regime for media, but it constrains the solution. Getting rid of media is not the right option.
♥ 3↺ 3
2) I do not buy into strict deontologism. It's not enough to make a rule, we need to make it work. If you force people into a choice between the law and survival, don't be surprised that they at least bend the rules. If you incentivise defection, expect defectors.
♥ 3↺ 2
Replying to @alexanderhanff and @PrivacyMatters, @podehaye, @newrepublic, @PrivacySurgeon
I just replied to Paul’s earlier question about the root cause with a broad spectrum option. But until such a time as we’ve made that work, I’ll take something imperfect over nothing at all, no?
Replying to @quinnnorton
Well sure, people had been trying to optimise paper-based targeting for a while too but the scale was nothing comparable. Also, it was mostly in the US rather than a global issue. Privacy law existed in many other places. In the US just for the public sector…
Replying to @podehaye
Allow me to answer with a question: Chrome shares your identity with third parties, Safari doesn’t. Who made that decision?
Replying to @podehaye
I’m not sure why you would point me to the source code for those. In Chrome, they would reject the change. In Firefox, the fix is already on the way (at least the early stages) because like they decided to align with Apple in following the law.
♥ 2
Replying to @podehaye
Addressing the cause of a problem is a pretty damn good fix where I come from.
♥ 1
Replying to @podehaye
So they determine the purposes and the means of the processing of personal data.
Replying to @IrishmanAboard and @FeargalHogan
That’s a question I’ve been wondering about and I’ll be honest: I don’t know. The incentive to sustain a team as expensive as, eg, the Chrome team without income is not obvious. Maybe we should pay for browsers? It’s not an insane idea!
♥ 2
Replying to @ehsanakhgari
As I said in the thread, you can (and should!) apply the same reasoning to mobile platforms. I just didn’t want to make the thread longer by doing that twice.
Thanks for the recommendation, I’ll definitely read that!
Replying to @quinnnorton
Yes, there already exist a number of options worse than cookies. I believe that browsers have the means to shut down a lot more than they do. But I don’t think that giving up is the answer to it being possibly worse, no?
Replying to @podehaye and @alexanderhanff, @PrivacyMatters, @newrepublic, @PrivacySurgeon
I’m not «admitting» to anything; I’m discussing options to help sustain better privacy across the Web.
Replying to @podehaye
I’m sorry but I’m not sure what you are describing exactly. When was that long time ago, what happened exactly, what documentation do you have about those events?
Replying to @podehaye
That solution is so radically impractical I don’t even know if I understand what you’re proposing. You want the NYT to make a fork of tracking browsers and then what happens? What’s the path to a billion installs?
Replying to @podehaye
I’m not trying to be difficult but I still don’t quite understand what you’re getting at? My fix removes tracking. Why is it not a solution?
Replying to @IrishmanAboard and @FeargalHogan
You do have a point. And there’s a risk that innovation in browsers would stop, they’re big and painful beasts to maintain, a real hard task for open source.
But the current model is unsustainable too — we need a way forward.
♥ 1
Replying to @IrishmanAboard and @FeargalHogan
At some point browser development might need to become purely foundation-driven (but actual foundations, not a mixed model à la Mozilla).
♥ 1
Replying to @quinnnorton
I absolutely agree, I’m mostly trying to stop one haemorrhage at a time.
Replying to @IrishmanAboard and @FeargalHogan
Yes, I don’t think that splitting from the OS is a viable thing anymore. It’s sort of expected functionality.
♥ 1
Replying to @othermaciej
I think we can debate an exact attribution of causality (as in @yudapearl’s ftp.cs.ucla.edu/pub/stat_ser/r…) but it is an enabling cause at the very least.
I know there’s more to do; one mole at a time!
Replying to @othermaciej
Every possible tracking power, no, but it’s possible to make it much harder, which in turn means people have to more actively break the law.
Replying to @podehaye
No offense but that’s the kind of «just» that is just missing so much detail it can’t possibly work. There are privacy-friendly browsers, why don’t they «just» get a better marketshare?
Replying to @podehaye
What software is here «on its own». It didn’t just decide «on its own» to be this way. By your reasoning if I install a photo editor that reports everything about my life it’s not subject to protection.
Replying to @podehaye
The household exemption assumes user agency. No one goes «hey, I’ll install myself a browser that just shares everything with advertisers!»
Replying to @podehaye
To whoever asks for it. Yes, there are protections, notably protections under Article 25.
Replying to @dmarti and @podehaye
I’m attending the workshop on permissions & consent next week in part to chat with browsers. I’ll see how I can help.
♥ 1
Replying to @othermaciej
The modern abuse potential is not by design; for cookies we knew it was a problem in the mid-90s and went ahead anyway. I would argue the case is different there.
I believe we can tighten up the platform quite a bit, though I worry the changes may be too drastic to be acceptable
♥ 1
Replying to @null and @dtunkelang, @MikeIsaac
It’s not so much about creepy — it’s about Google using Chrome to treat Google properties preferentially. The next step is that sites that use Google SSO can be logged into more easily… and then it’s the one profile for the Web.
For creepy there’s google.com/chrome/privacy… 🤗
♥ 1↺ 1
I wonder how many yard sales you need to send aides to to find a sufficiently unmarked 1982 calendar.
♥ 4
Replying to @MikeIsaac
Ditching him means admitting they fucked up and nominated a rapist, weeks from the midterms. How credible would they be with anyone, really? Plus evangelicals want this rapist.
♥ 3
Replying to @LAM_Barrett and @MikeIsaac
Agreed. I'm starting to wonder how many weeks' worth of revelations are left and at what point they'll just confirm anyway. The longer it goes on the more toxic it is.
♥ 2
Replying to @othermaciej and @ehsanakhgari
Wouldn't there be value in standardising the fixes into the platform?
Replying to @alexanderhanff
I have actually made proposals along those lines, yes (except more practical). You keep being aggressively hostile with your conspiracy theory "you guys" so this is my last response. Bye.
Replying to @Chronotope
Oh, I did not mean to imply that this was the only thing that needed be fixed everywhere for all things. To boot, there are problems (as you note) in the ad ecosystem that go well beyond tracking.
But the influence of this tiny entry point on the data market is huge.
♥ 1
Replying to @Chronotope
To stay within an ecosystem metaphor, blocking at the browser level has a massive incidence on the food web.
Also, changing a browser’s behaviour isn’t an engineering fix, it’s heavily political ;-)
♥ 1
Replying to @podehaye and @jason_kint
The decision from the browser to share personal data is itself processing. It involves maintaining a data store, deciding to attach the information on an outbound message, etc.
This doesn’t absolve downstream from compliance — but the decision to share is the browser’s.
♥ 1
Replying to @whvholst and @quinnnorton, @BrusselsGeek, @PrivacyMatters, @newrepublic
FWIW I’m always up for drinks and de-escalation. And I agree with your “losing faith” comment — in fact, that’s what started this thread.
♥ 1
Replying to @podehaye and @jason_kint
It certainly warrants being tested in court. But I have been looking into this for a while and I have yet to read a solid blocker.
Replying to @daithesong
I think the semantics can be debated up to a point, notably depending whether they are provided to first parties or third parties. But Safari is doing much good in this area anyway, it’s not sharing!
Replying to @natashanyt
There’s a lot that’s quite fuzzy and could easily end up toothless in there; but in fairness there are good things.
The points on territoriality and data exports at the end seem tailored to try to reduce GDPR applicability, though. I’d watch those.
♥ 4
Replying to @podehaye and @JuliaAngwin
Yes! I saw this and I’ve set myself up to be notified of anything from @team_markup, it looks like it could be very exciting.
♥ 1
Replying to @natashanyt and @jason_kint
This is a tricky part. Detailed consent is user-hostile and easily gamed, you end up with users having to fight an army of the best and nastiest UX designers. There are compliance/fiduciary alternatives but they have all the downsides of corporate responsibility.
Replying to @LydiaFdelaTorre and @DLTsays, @natashanyt
If I were trying to make the case (say, eventually, to the WTO or some such institution) that data protection law should not be extra-territorial, those are pretty much the arguments I’d be marshalling.
♥ 1
Replying to @LydiaFdelaTorre and @DLTsays, @natashanyt
«Think of the small businesses! They can’t GDPR, poor things… What if some crazy Frenchman decides to sue them out of existence?»
Replying to @LydiaFdelaTorre and @DLTsays, @natashanyt
Well yeah, that is my actual opinion — but it doesn’t read like that’s Google’s from that document.
The notion of a court in Helsinki going after a mom & pop Web shop out of Kansas is absurd, but it’s a good scare story.
♥ 2
Replying to @LydiaFdelaTorre and @DLTsays, @natashanyt
It does, but in my experience courts also don’t waste time trying to enforce what’s unenforceable.
This could have been a serious threat to the Web; thankfully no one has wrapped it in a React component yet so malicious developers aren’t sure how to use it.
Quoting a tweet by @heydonworks ↗
♥ 3
Replying to @timoreilly and @sundarpichai
I would argue this one isn’t user-hostile, it’s Web hostile: this is Chrome giving preferential treatment to Google properties. It’s not the first time (they document many other case in google.com/chrome/privacy…) but it’s increasingly brazen.
↺ 1
Replying to @LydiaFdelaTorre and @DLTsays, @natashanyt
Oh yeah don't even get me started on those people... I did have a grand time being the crazy Frenchman who was going to sue them if they kept spamming though. 😏
♥ 1
Replying to @laparisa
Do those product changes involve not using Chrome to give Google properties preferential treatment?
♥ 14↺ 2
Replying to @null and @SimonDeDeo, @LGcommaI, @sfiscience
I wish conferences made use of blackboards more often. I once spoke at a CNAM amphitheatre in Paris, deep 19th century circular pit, huge blackboards, no need for sound amplification despite it being a huge volume. It was amazing.
♥ 3
Replying to @null and @SimonDeDeo, @LGcommaI, @sfiscience, @IAIASantaFe, @DavidFeldman
Yes, it needs to be dug deep. It's funny how fast you get used to speaking 45 degrees up while prancing around on a stage.
♥ 1
“It’s not even consent fatigue anymore, it’s consent fatalism.” #permissions
♥ 1
Christina Boyce-Jacino and @SimonDeDeo used a @nytimes crosswords dataset to identify what makes a question hard, advancing applied erotetics in the process.
“Opacity, Obscurity, and the Geometry of Question-Asking“ arxiv.org/abs/1809.08291
♥ 11↺ 2
Some words just don’t get used as much as they deserve to be. t.co/tf5hlPVSFO
♥ 1
“We don’t really have good accountability for JavaScript.” — Martin Thompson, #permissions
It’s almost like the declarative and Principle of Least Power people weren’t _entirely_ wrong.
♥ 1
Replying to @dckc and @agoric
I hadn’t seen that, it looks quite promising — thanks! I doubt that we’d be able to bring that promise to the browser though…
Replying to @teleclimber and @dckc, @agoric
Yes, to clarify: this is in the context of people being accountable for Web features. I think that generally on the Web there’s just too much toothpaste to put back to get general accountability for JS.
But realms show great potential, eg. to sandbox third parties.
♥ 1
Replying to @0gust1
Yes! There’s a lot of good in there.
Nelle listing a few things interesting privacy threats from the wonderful world of XR. #permissions
Helen Nissenbaum shares her excellent take on consent and privacy over at the @HarvardBiz. She’s also proof of just how practical philosophical framework can be when done right.
“Stop Thinking About Consent: It Isn’t Possible and It Isn’t Right”
hbr.org/2018/09/stop-t…
♥ 1
I wonder if usage of the sentence "You could just do X" correlates strongly with elevated Dunning–Kruger effects. Has anyone looked at this?
♥ 2
Conversely, "Why can you not do X?" should correlate in the other direction.
Replying to @sandhawke
No, just discovered it today!
Replying to @null and @JoWolffBSG, @SimonDeDeo
Are we not already there? (On the off chance that you hadn't seen this piece of brilliance from a few years back.)
fauxphilnews.wordpress.com/2012/02/22/kri…
Replying to @sandhawke
Yeah, same with Mastodon. The experience isn’t there IMHO, and much of the stuff I’m here for isn’t there.
♥ 2
Replying to @tobie
I think it’s just an open source project; it doesn’t need infrastructure so it can stick to that model.
♥ 1
It has been too long @w3cmemes, it has been way too long. 66.media.tumblr.com/34f8917b16ab22…
♥ 1
Replying to @seanmcarroll
My experience is that making them shorter makes them more interesting. Speakers have to stick to the strong points and not go into their pet digression.
Confused about what the theft of 50 million tokens from Facebook means? Look at what @nytimes folks were able to get by using just *one* such token.
You read that right.
Quoting a tweet by @gabrieldance ↗
♥ 16↺ 26
I'm hearing quite a few people saying there can't be a fine unless there's negligence. Folks, if it can be established that they did not properly minimise data collection that's negligent.
This is Facebook. How good would you guess they are at only collecting the data they need?
Quoting a tweet by @CaseyNewton ↗
♥ 1↺ 1
Replying to @stef and @EricaJoy
There is so much that Mozilla screwed up developing identity, it's almost like they wanted it to fail... I wish someone would pick it back up, it had some nice properties.
♥ 1
Replying to @trekonomics
To show a link to the #FermiParadox it is not really relevant that it's hard to match the efficiency because evolution can optimise that over huge time spans, the tough question is whether it is hard to bootstrap. Is that the case?
♥ 1
Replying to @trekonomics
Abiogenesis may be hard, yes, but photosynthesis comes later. It's not obvious (to me) that it is hard to kick-start photosynthesis (perhaps anoxygenic) for example next to volcanoes.
♥ 1
Replying to @trekonomics
Sure, but is that not primarily a question of getting the ball rolling and waiting for a few Ga?
♥ 1
Replying to @trekonomics
Yes, but don't those precisely show that life is resilient as fuck? You can turn the Earth into Hell, wipe 96% of species out, and it still bounces back.
♥ 1
Replying to @trekonomics
BTW if you have good reads on mass extinctions I'd love to hear. I tried "The Ends of The World" but I found the writing grating.
♥ 1
Replying to @trekonomics
I promise I won't! It just, I don't know, it felt like it was constantly trying to force enthusiasm when the topic is exciting on its own and doesn't need cheerleading. I gave up a couple of chapters in 😔
♥ 1
Replying to @philipjrosen and @seanmcarroll
Paying is a plus, it means you're the customer. I'm pretty happy with Evernote, it handles thousands of notes well. I also use it to scan my paper notes, its document mode is decent. That said for just snipping stuff off the Web I also use Pocket.
"First Lady of Fabulous", the bag I didn't yet know I needed.
♥ 3
"First Lady of Fabulous", the bag I didn't yet know I needed.
