Replying to @jesslynnrose
I'm going right now, you should too!
November 2021
245 tweets
Bowker & Star's use of the *phone book* to demonstrate the pervasive nature of political choices, particularly where categories are created, is both simple and highly effective.
♥ 10
Replying to @gotdairyya
Thank you — I saw it was longer than the alt and didn't think to thread it.
I wish Twitter allowed embedding text snippets…
Replying to @gotdairyya
I think they're trying to avoid people using it to tweet longer tweets, but that seems silly tbh.
The polarising effects of exposure to others are established, but the conclusion drawn from it here is not: social media can be designed differently and managed better.
None of this is "baked into social media itself."
Quoting a tweet by @michelleinbklyn ↗
♥ 12↺ 4
Replying to @DoraCrisan
That's not the problem, the problem is that not all interactions are the same and some structures are bad.
Don't stay stuck in the search/social algorithmic media intellectual goop — go rewild your attention!
Quoting a tweet by @kottke ↗
♥ 2
So much hard-hitting truth in this brilliant talk from @ariezrawaldman about how companies define privacy in ways that lead privacy-friendly employees to eventually support extractive practices.
♥ 26↺ 8
• "So long as people have transparency, I'm comfortable."
• "Our users have a lot of choice!"
• "You should see how many controls we have over this data, it's impressive!"
• "Our security team is the best in the world!"
♥ 10↺ 1
Anyway — I bought his book within the first five minutes. Be careful with @PrincetonCITP events, they can be expensive! bookshop.org/books/industry…
♥ 11↺ 1
Replying to @jesslynnrose
We keep two printers around, it's the only way I know to make sure that there'll be one working when it's needed.
♥ 2
A very true thread. Most of the JS big libs and the advice around them reminds me of J2EE with nicer graphics.
I would add that it's not just libs, but also standards: regular size companies using typical business tactics are woefully underrepresented there.
Quoting a tweet by @jeanqasaur ↗
♥ 13↺ 6
Replying to @PrivacyMatters
In the talk (and I presume in the book too) Ari made the case that we can't rely on people beating the system at every turn, even if supported by stronger privacy regimes when these still expect traditional approaches to compliance. I think he's right.
♥ 2↺ 1
I am lucky to benefit from real support and resources where I am, but it's still uphill every way. Everything from software defaults to how the internet works is part of the problem. We need to find radical change.
♥ 3↺ 1
Replying to @DoraCrisan and @PrivacyMatters
No?
Replying to @null and @Ben35846532, @PrivacyMatters
Right, that's one reason why I feel that DPO requirements are often ineffective. The way to flip the incentive is to make it part of the culture so that you are seen unfavourably when not supporting privacy even though the company is paying you, but that is hard.
♥ 3↺ 1
Replying to @divclassbutton
There's definitely a mix of jQuery, Web Components, and a controlled data store that Just Works™ for this.
Replying to @null and @MadronaCitizen
I don't think it's even that deliberate, the incentives don't stack well. But yeah, being in a discussion about fixing the constant issues of the Web with a group of people who couldn't run a small digital marketing campaign to save their lives can be a problem!
Replying to @PrivacyMatters and @Ben35846532
Frankly, what surprised me the most when I switched to this from other governance areas is just how hostile to privacy so many "privacy pros" are. I got a sense that I was weird trying to make it work. DPO seems supposed to be "Don't Privacy Officer".
♥ 3↺ 1
The topic: 😍😍😍😍😍
The price: 😱😭😱😭😱
bookshop.org/books/the-camb…
♥ 6
Replying to @catthekin
I'm not sure that I can borrow from the uni libs. I might be justified in expensing it, though, for a project. Failing that, it might be a Christmas present!
At that price level, it's like they're trying to encourage SciHub/Libgen.
♥ 2
That feeling when you're asked to consent to legitimate interest…
"Yo dawg I heard you like legal bases so I put your legal basis on a legal basis so you can legal basis while you're legal basising."
♥ 36↺ 8
Replying to @calebwatney
Wait — you're saying that all those poor shower designs in the US are actually made that way on purpose?
♥ 1
Replying to @consttype
Right — but supporting complex behaviour from a non-complex system *is* costly.
Loving this new mural my wife made above the bed.
♥ 24
Replying to @null and @blakereid
I'm usually too-thirsty for milk.
Replying to @farbandish and @blakereid
None of that will address algorithmilk bias, though.
Replying to @farbandish and @blakereid
Well, you could start by asking which horn of this dilemma content mooderation sits on.
♥ 1
Replying to @btsavage
Thank you! Sort of, it's not figurative overall.
Replying to @ariezrawaldman and @PrincetonCITP, @WashULaw, @PrivacyPros, @daniellecitron, @neilmrichards
I'd be happy to ask if there's interest at @nytimes if you're up for it!
♥ 5
Replying to @kottke
I usually have no problem sleeping, but I had a particularly stressful summer that kept me up. There's an app (on Android, I suspect same on iOS) called Prana Breath that has 4-7-8 and other patterns. I confirm that it worked great for me.
♥ 2
Replying to @ariezrawaldman and @PrincetonCITP, @WashULaw, @PrivacyPros, @daniellecitron, @neilmrichards, @nytimes
I'm thinking of a book talk though, so more speaking at The Times than with. I think our data stewards group would be interested. Happy to hash out details, my DMs are open.
♥ 3
Replying to @bigblackjacobin
Isn't that flat out unhealthy? I mean, it's basically fat and sugar, its purpose is to make calves grow.
♥ 3
I just looked it up: a calf will drink about double that, but it's its only source of nutrition and it's growing like fuck. There's no way it's good for humans, especially adults, to get near that.
♥ 3
"We're doing it for the SMBs" is the new "think of the children."
Quoting a tweet by @IABEurope ↗
♥ 3↺ 2
In a Caturday state of mind.
♥ 11
Very Caturday.
♥ 5
Replying to @null and @borisschapira
Ça me tue toujours ce commentaire. De la chance d'avoir des enfants, en bonne santé, etc. certes, mais bon. Je me retiens régulièrement de répondre "J'ai en truc! En fait il ne suffit pas de les nourrir, il faut aussi les élever!"
A clear and dispassionate overview of the damage wreaked by AMP and the reason why it's become very difficult to treat Google-led initiatives with anything other than intense suspicion. wptavern.com/amp-has-irrepa…
♥ 283↺ 104
Replying to @lehors
And they are still forcing publishers to use it. It has become less prominent in Search, but it's still required in other products.
Also, they're doubling-down on the strategy with Showcase.
Replying to @peligrietzer
And if you cook a can of it in a pressure cooker you get dulce de leche. I'm leaning "great."
♥ 1
Replying to @JamesHercher
Same. I used to assume that there were good intentions. I still think it's true in some cases, but it would be irrational, based on experience, to look at a Google proposal as anything other than guilty until proven innocent. It didn't have to turn out this way.
♥ 4↺ 2
Replying to @WolfieChristl
In that vein, given existing indications that the broad sharing of data leads to anticompetitive outcomes, should agreements to support greater sharing be investigated for collusion?
♥ 3
Replying to @sandeepofmovva
We have to, to some degree, though. The current system isn't working well for us. FLoC is full of issues, FLEDGE has some good ideas, though it has architectural flaws. PARAKEET is promising 😁
Replying to @TheHonorableAT and @doriantaylor
Knowing and finding good writeups are different things. We also have increasing evidence, which matters.
♥ 1
Replying to @ukla_ukla and @histoftech
That is true for AMP but not initially for every project. There was a time when trust was there.
♥ 1
Replying to @TedMielczarek and @SteverRobbins, @jason_kint
Except fragile and full of hacks, it breaks all the time. But you have no choice to use it.
♥ 2
Replying to @TheHonorableAT and @doriantaylor
Into solving this mess? We're not there. Right now there's a lot of push-back but most of it is stop energy, not alternatives. Heading that way, there won't be a tipping point, just an entrenchment of slightly less power. We need more alignment on other organisations for the Web.
♥ 1
Replying to @TheHonorableAT and @doriantaylor
That's possible, it has certainly crossed my mind that there may be no other option, especially given their forays into military and "antiterrorist" work. I still see a path to good from here that's more gradual, but it requires getting a lot right!
Replying to @jathansadowski
Legibility and visibility are two different concepts. The dataification from beacons won't remain limited to vehicle detection.
♥ 3
"Dad, you look like Einstein, but only the hair, and then also a lot less good."
Ow.
♥ 18↺ 1
Replying to @RDBinns
You can keep going! I took a similar trip, starting from Grenoble but continuing beyond DK to Göteborg, Oslo, Bodø, and eventually Å. I mean, what's a few more weeks on a PhD?
♥ 2
Replying to @Tupp_Ed
Don't go about giving them new ideas!
It's about time this entered broader consciousness. And it's particularly true of infrastructure.
Quoting a tweet by @rcalo ↗
♥ 4
Does anyone know what the plan is to make Twitter Blue Ad-Free *actually* work given that it seems to be built on tech that's already obsolete? 😬
♥ 19↺ 1
Replying to @RickByers
That's sort of what used to be the case with Firefox, but I can't seem to dig up any info about that!
♥ 2
Ah, the 90s with its infrared badges that can signal your in-building location so calls can be routed to the phone nearest you. 💕
♥ 8↺ 2
Replying to @RickByers
The neatest solution I've seen so far in this space is @UnlockProtocol. I'd love to see people take a closer look and build iterations on it if and when issues are found.
♥ 3
Replying to @MaxGendler and @Chronotope
Proof of work is such an American capitalist burnout concept, that's how you end up blocked and chained. What I want to see is proof of long weekend, proof of chilling with your family, proof of reading a good book…
♥ 11↺ 5
Replying to @bayesianboy
Thank you for asking the questions that matter. I've been wondering why people ride bikes for pleasure my entire life.
Replying to @MaxGendler and @Chronotope
Look who's grumpy from failing at their proof of having had a glass wine.
♥ 2↺ 1
Exactly the interoperability cut that I've been (cough) recommending! We should make sure that the standards are ready, cc @mnot.
Quoting a tweet by @CT_Bergstrom ↗
♥ 11↺ 3
Replying to @CT_Bergstrom and @mnot
Next time, come to the social media as a crisis discipline workshop so that you can steal my takes to my face, Carl!
♥ 1
Replying to @temptoetiam
I wish it were only five 😁
Replying to @krave
Ah! Thanks, that makes sense, the docs made it sound really weird.
The problem with trust is that, in this case, it's circular. For sites to be trusted, they need to be trustworthy. But opening broad holes for data sharing also enables untrustworthy behaviour...
♥ 2
...and so none of us can have nice things. I'm not sure that I've seen apps do this any better, though?
Replying to @krave
Reputation and regulation aren't catching either in apps, though. Stuff like geolocation sets are being stopped the same way as in browsers: by adding restrictions. No one's figured out trust of arbitrary parties at scale without limiting capabilities (that I know of).
♥ 1
Has anyone looked into whether Lloyd vs Google might legitimate similar schemes to circumvent user protections, like SWAN, in England & Wales? Or is it moot because it's the DPA98?
♥ 3
Replying to @LourdesTurrecha and @44
They explained how it works, the docs are actually misleading!
Agre's "Surveillance and Capture" is worth re-evaluating in today's context. Much of what we're dealing with looks more like capture than surveillance, and thinking about it as such is useful. He also helps define legibility better. blogs.commons.georgetown.edu/cctp-506-sprin…
♥ 5↺ 1
Replying to @Chronotope and @swodinsky
Yeah, why would that make you look like a shill? Interoperability, done right, is an important part of the answer. Haugen is a great whistleblower but it's unclear to me why she's being asked about solutions.
♥ 3
Replying to @null and @podopie
I'm a man of the people, Ed.
Replying to @null and @dansereduick, @montezumachavez
What's the other kind of scholarship?
Luis: for PRA, you want something strong and hearty, I could see a Malbec.
Daniel: injury in fact?
♥ 1
Replying to @null and @dansereduick, @montezumachavez
I agree that it's not enough, but it can be one contributing step. I *think* that one of the state laws was considering at least some breach or violation types as automatically qualifying as injury-in-fact precisely to eliminate that debate. Fuzzy memory, though.
♥ 1
Replying to @null and @paulduplantis
I haven't read "Deschooling" but "Tools for Conviviality" was great!
Replying to @PrivaCat and @dansereduick, @montezumachavez
Definitely in the US!
Replying to @null and @paulduplantis
Credit where due: this came to me from @dauwhe.
Replying to @stuartathompson and @GregBensinger, @ShiraOvide, @jtoonkel
LOL Stuart, everyone knows it's immdeeb!
♥ 1
Replying to @ShannonVallor
Naive that I am, I thought that the perks of joining the Illuminati would include controlling email providers so you wouldn't run into such issues.
Replying to @swodinsky
♥ 1
WaPo gets one of the best people possible to work on privacy there! Absolutely love to see this.
Quoting a tweet by @Chronotope ↗
♥ 16↺ 1
Replying to @temptoetiam and @Bouletcorp
Les voitures c'est pas vraiment mon truc, mais celle-là j'avoue que je lui trouve de la gueule 😬😁
♥ 3
Replying to @Katelinsaurus and @FQXi
♥ 1
Question: is there any evidence, empirical or at least from principled arguments, that sharing data broadly across contexts would ever improve competition?
I've now read a few thousand pages on this, and I've only found indications of the contrary. 🧵
♥ 44↺ 10
I ask because of the prevalence of the Big Knob Theory (BKT). BKT posits that there exists a Big Knob that can be turned towards "privacy" (no data shared across contexts) or towards "competition" — but it's very much a zero-sum game.
It's popular — but where's the evidence?
♥ 8
I've seen arguments that if one party gets all the data (because they can work around protections or they can mix from multiple contexts they own) and the others get none, then you'll get bad competition.
I agree, but then that party is clearly violating privacy so it's not BKT.
♥ 5
Replying to @MarechalPhD
I realise that, but they don't seem to be aware that it's a fiction. Or even to have bothered producing a fake study in support. It's just accepted on faith.
♥ 2
Replying to @alextcone
I think that most of the findings from non-data markets are hard to port over, though, they're just so different.
♥ 1
Replying to @willseth
I think you touch on the right point but indirectly. You can generate more value with more data, so sharing more creates greater value. That's true internally as you say and also externally. But the competition question isn't total value maximisation, it's capture!
There might indeed be a case for saying that ad targeting (and comparable products, like insights) is a natural monopoly and should just use Ramsey pricing. Maybe! But then we should build that, not argue against privacy.
That can't be right.
♥ 16
Replying to @catthekin and @ANderagakura
I don't know, my trip finished 😁 I'm not much of a games person, it's only when I'm tired and I don't persist much.
♥ 1
Replying to @alextcone
Sure, but if you enable cross-context sharing you will get the place that tracks users in the most places being that place where there are the most customers, and it's going to be winner-takes-all anyway, no?
♥ 4
They said "Tim Burton themed" so...
♥ 13
Replying to @alextcone
My hope is that we can fix that, but we have to move beyond this narrative.
Replying to @Caffar3Cristina
Thank you!
Replying to @vasudhapassi
Thanks!
Replying to @kvnweb and @akapczynski
Thank you so much! I will definitely read that.
Replying to @kvnweb and @akapczynski
Well, if you know how to prevent them, you also know how to build them 😁
Replying to @kvnweb
I don't know that I'm well read, the pages add up *fast* in this space, but I definitely agree that the data dividend is nonsense. And these issues are opaque enough that it's easy to sneak in some populism.
Replying to @kvnweb and @akapczynski
Indeed, though I don't know what his actual take is. When I met him, he seemed sincere but at the same time he didn't hesitate to push a position on stage that had been discredited in direct discussion the previous day.
Replying to @kvnweb
It works for me, but the site is extremely slow for some reason. Maybe it timed out on you? It's aimed at marketers, so it might not be your usual fare!
Replying to @KateKayeReports and @nytimes, @linakhanFTC
The problem with "selective" is that if there are power asymmetries in the market, you might not be the one making the selection! That's what happened with the GDPR. In theory, you can choose not to work with a third-party data controller, in practice the terms were dictated.
♥ 3
Replying to @JamesHercher
Common argument, but no. That just shows what wins with buyers but not that it increases competition in the market. It also mixes up the values of all the things that cookies do irrespective of whether there exist private alternatives.
♥ 1
Replying to @KateKayeReports and @nytimes, @linakhanFTC
It's hard to compete against default installation on Android and the marketing power from badgering people into installing your competitor.
♥ 1
Replying to @ruchowdh and @KateKayeReports, @nytimes, @linakhanFTC
I'd be curious to hear your take, is the talk (or other) available? I tend to think that we don't have a pacing problem. Figuring out the right set of regulations for the digital world is hard, but we only need to do it once.
♥ 2
Replying to @kkomaitis and @javierpallero
The way I have seen it presented is that the underlying systems *must* support the sharing of data, otherwise there will be adverse competition effects. It's a strongly held belief but... I can find no evidence!
Replying to @ruchowdh and @KateKayeReports, @nytimes, @linakhanFTC
Thank you!
Replying to @MelMitchell1 and @hardmaru
Many experts at the time thought that it was underwhelming. You could have broken links easily, which a proper hypertext system would prevent. It also had an unimpressive markut language and initially no way to convey content types.
♥ 4
Replying to @BoojumSpeaks and @MelMitchell1, @hardmaru
We actually managed to fix the error handling and future proofing in 5, so not that bad, it only took twenty years 😁 I'd put unresolved, entrenched problems on my list, like how search or ads work!
Replying to @KateKayeReports and @ruchowdh, @nytimes, @linakhanFTC
That sounds a lot like my job so I'm all the more interested :-D
Replying to @kvnweb and @akapczynski
Thanks, that's very much up my alley (it sounds related to berjon.com/stewardship/), I look forward to reading it!
Replying to @julien51
Controlling access is key to a number of institutional forms. I see no reason to limit that part of the toolbox.
Replying to @gpxl_
They're the best.
Replying to @julien51
There can be, but historically there hasn't. I'm pretty sure that ruling this out excludes useful institutions.
♥ 1
Replying to @julien51
Sure, I'm not saying you must always require approval, just that sometimes you do. For instance, you might want to check that a new member has high quality standards before accepting them on a shared wire service.
♥ 1
Replying to @doctorow
In this vein, I'm curious as to whether you've found Agre's capture vs surveillance distinction useful? I think it helps escape metaphors that might not be pointing to the right solutions.
♥ 1
Replying to @Chronotope
Wow. I knew they were biggish, but not this big. They say they want to be competitive in ads, but I don't see them much in the groups? Might be worth a chat.
♥ 3
Replying to @neilmrichards and @hartzog, @BillMcGev, @ariezrawaldman, @salome_viljoen_, @julie17usc, @rcalo, @MargotKaminski, @mhintze, @ColinRushing
Whenever I think of my browser dutifully sending a GPC signal of my privacy preferences with every request it makes, I imagine it quietly singing to itself:
🎼 That's the way — uh-huh uh-huh — I liiiiiike it — uh-huh uh-huh! 🎶
♥ 5↺ 1
Replying to @donohoe
Not a great fix, but maybe just remove the shortcut?
Replying to @seldo
You… didn't include <sarcam> 😭
Replying to @montezumachavez and @TrialByTruth
People are complaining about the VP's ability to speak French?? That's just... Why?
♥ 1
I tried reading Giddens's "Constitution of Society" because I'm curious about structuration and structure/agency relations, but it's not working for me (too much about language, references whatever debates sociologists had in the 80s waaaay too much).
Any better options?
♥ 1
Replying to @null and @loganfranken
I picked this one up because I had actually liked The Consequences of Modernity. Not always the lightest writing but at least I could tell where he was going. Thanks! I'll give that a shot.
Replying to @tim_libert
Thanks, it does indeed seem cool.
Replying to @kevinmarks
But that only works because it's being pooled in public.
Replying to @judyshapiro
I don't think that's accurate, FLoC doesn't actually share the data. It's a failed model, but for different reasons. It relies on wrong assumptions about privacy-preserving tech, governance, innovation.
Replying to @johnwilander and @kyotonio
I must have missed that debate? I mean, there's an explicit link to Ads and they keep pushing it.
♥ 1
Replying to @hackylawyER
Looking forward to the sequel to The Social Network feat Haugen having technocentrist epiphanies.
♥ 3
@jinyanzang and @iamdhruv4 are most excellent hosts, I highly recommend this!
Quoting a tweet by @TechBecomesLaw ↗
♥ 1↺ 2
Replying to @swodinsky
I'd like to think René Carmille's response to this idea would have been very rude, even in French. en.wikipedia.org/wiki/Ren%C3%A9…
♥ 1
"Think of browsers as the next generation of operating systems, the infrastructure and connective tissue."
Great points from @protocol — but much in the browser world, from financing model to defaults in install & search, is wrong. Can we fix it now?
protocol.com/newsletters/so…
♥ 9↺ 3
For a second, I thought we were doing "Spread Firefox" again. I totally would! twitter.com/anildash/statu…
♥ 9↺ 2
Replying to @kkomaitis and @protocol
Very little participation, but a modicum of shared governance in standards. I think that has to be part of the fix, though.
♥ 1
It's not that I don't enjoy my job, but so far no one has sent me to SEVEN HUNDRED surf spots on the Atlantic to assess various approaches to common-pool resource governance without state intervention amongst surfers.
And that's unfortunate.
link.springer.com/article/10.100…
♥ 1
I've been trying @DuckDuckGo new Android tracking protection system that keeps apps from broadcasting data around, and so far I really like it! wired.com/story/duckduck…
♥ 20↺ 2
I get that it's a hard industry out there and apps have to do things they'd rather not, but I don't think leniency should extend to apps that support trust, like security apps or browsers.
So WTF is a marketing analytics SDK doing phoning home in Microsoft Authenticator?
♥ 6↺ 2
Replying to @shails
Well, yes and no, some stuff really has to work in the background, like listening to music while you're doing something else!
Replying to @alextcone and @shails, @YouTube
I use it rarely, but I'm thinking I might pay just so they stop badgering me to pay.
♥ 1
Replying to @fatemehx2
I know... The FAQ claims that it's deidentified, but... why does Microsoft need someone else to collect deidentified usage data?
♥ 2
Replying to @Chronotope and @fatemehx2
Amazon syncs with a ton of others, I presume for DMP/cookie sync purposes. Their site is a Christmas tree of third parties, or at least was last I checked.
♥ 2
Replying to @publictorsten and @DuckDuckGo
No, it's like a VPN except that it only filters the traffic.
Replying to @alextcone and @Chronotope, @fatemehx2
Has anyone dug into what Amazon does? Sometimes I see twenty third parties showing up on a page, and sometimes none.
♥ 1
It's interesting to see Twitter keep positioning itself as a healthier, friendlier, ethical approach to tech rather than try to NOBO everyone else out of existence. I hope they can keep pulling it off, I'm curious to see where it leads.
Here with AMP:
theverge.com/2021/11/19/227…
♥ 20↺ 4
"Even if I’m reading nonfiction books – which is 90% of my reading – I still view the content as an adventure movie."
Why curiosity > smartness.
durmonski.com/life-advice/cu…
♥ 4
Replying to @null and @benlikestocode
I have been suspecting the same thing. You can't innovate in AMP because Google decides everything, so ad-free can't be done there. That said, they could very easily have differentiated.
Replying to @null and @benlikestocode
In fairness, this is basically Scroll under the hood so they have to track visits in order to apportion revenue (outlets get paid by share of traffic). But yeah, my longer-term preference for this would be to have privacy-preserving ways of doing the same.
♥ 1
Replying to @slightlylate
Glitchy is good, I thought it wouldn't load 😂
Replying to @Chronotope and @alextcone, @fatemehx2
Audible is Amazon, no? They have all kinds of partnerships. Also, since when do adtech companies get data because it's useful?
♥ 4
Expat gatherings can be good, sometimes 😁
♥ 28
Football big fire, wheeee 🐯
♥ 1
Replying to @montezumachavez
Next time!
Replying to @btsavage
Would you argue that it's an improvement that one should agree to?
This would be a spectacular improvement. Combining data across unrelated services of a first party is not in any way better than combining data across third parties — and relying on consent solves nothing.
I hope this happens!
Quoting a tweet by @BeterOpDeFiets ↗
♥ 21↺ 3
Replying to @mnot
This came up at the Social Media as Crisis Discipline workshop. On first principles, you'd think that splitting the big social mosh pit into groups should help prevent misinfo, toxicity, polarisation — but when Facebook pushed for groups it was a train wreck.
I think that is likely because doing that with all the surrounding trappings of social media doesn't solve the problem: it just allows it to fester in hiding. I suspect that E2EE would suffer from similar problems.
♥ 1
Replying to @alextcone
That's the DMA's scope. But I would be happy with this being a general rule.
Replying to @alextcone
Yup, totally agreed, though I have a long list of things to do before I get around to defending the platforms 😁
♥ 1
Replying to @bigblackjacobin and @kellymakena
What the fuck, mashed potatoes might be one of the best possible dishes. Also, what the fuck is "meal filler," is it drab food for people who have enough to eat anyway? Why would anyone do that?
♥ 2
Replying to @mikarv and @RDBinns, @1Br0wn
The details will matter a lot. If the data isn't used for targeting on O&O, it can still be used to increase inventory by improving eg. Search.
♥ 5
Replying to @BrettFrischmann
It should be possible to train a model to suggest measured, factual variants for tweets. Could be a fun project.
Replying to @julien51
You need to poll 8 year olds. A bunch of other parents are going to be terribly biased.
♥ 2
Replying to @thezedwards
Yeah, we need to correct that misunderstanding. It's very much on purpose that it doesn't do that.
♥ 4
Replying to @thezedwards
Oh yeah, that's totally the plan. See berjon.com/gpc-under-the-…. *But* not interfering with local storage consent is deliberate. That's largely an ineffective and outdated mechanism IMHO, and one that can't be addressed well in a general fashion.
♥ 4
Replying to @publictorsten and @thezedwards, @NOYBeu
I do. I think it's an interesting idea, but I think it still suffers from being fingerprinting vector that unlawful actors can abuse. Also, the problem with automated consent is picking defaults. I prefer solutions from collective governance, less weight on subjects.
♥ 3
Decentralisation does not prevent the concentration of power; it can very easily make collective action harder until might makes right.
Build for polycentricity or GTFO!
Quoting a tweet by @Vinncent ↗
♥ 14↺ 2
Replying to @_abialas
It also rocks that they made an article with like twenty pictures but didn't turn it into a slide show. THANK YOU SIBERIAN FRIENDS.
♥ 2
Hey, I'll be speaking about "Getting Some Privacy on the Web" at Web Directions Safe in December!
The conference is super well organised, and it's set up so you can attend at a convenient time from anywhere the world around. Use discount code "robinbsentme" for 20% off!
♥ 18↺ 3
Replying to @_abialas
I did not, I also didn't read about the Siberian practice of bikini skiing because I know I'll just want to do that too. I must resist feeling too Siberian in my heart.
♥ 1
Replying to @MaxGendler and @_abialas
I mean, it's hard to resist: "On Sakhalin skiers came downhill banya-style, wearing towels or white sheets as if just out of the sauna." siberiantimes.com/home/voice-of-…
♥ 1
Replying to @gchampeau
Il n'y a pas de raison de ne pas les avoir dans SciHub; je crois même qu'elles ont un DOI.
♥ 4