Clearly, my ancestors knew how to have a good time.
♥ 20
May 2022
418 tweets
My grandpa too!
♥ 5
Replying to @null and @amyvdh
Haha 🤗♥️
The socks are everything.
♥ 1
Replying to @innovimax and @ibjacobs
Jane! It would be amazing if it were the same family!
Replying to @BrettFrischmann
Is it knowledge reduction or is it control/management? Speech that reduces your ability to do what's right for yourself, and in that increases your availability for management by others.
Replying to @null and @amyvdh, @innovimax
Reggie isn't having it.
Replying to @null and @amyvdh, @innovimax
Scottish 🤣 But yeah.
Replying to @evgenymorozov
He looked cook much of his life!
Replying to @e2dot7182818284
I saw that when it came out but unfortunately I am unable to listen to podcasts so I'll have to discover @big_data_kane's views through other means!
To the broader point, I think that the most interesting parts of complex systems cannot be rendered *legible*...
♥ 1↺ 1
...(in the SLAS sense) and therefore not readily digitisable in the way that we would normally digitise things (by categorising and capturing regularities).
But that doesn't mean that a digital environment cannot be complex, the Web (in a capacious sense) is!
♥ 1↺ 1
But indeed, the people who wish to act on the web as modernist city planners of old proceed through legibility and categorisation. It can't work! Though they can "succeed" (in imposing their ways) because they have the power.
♥ 1↺ 1
Replying to @null and @dansereduick, @montezumachavez, @RobertJBateman, @floorter, @JBAGerritsen, @PrivaCat
I haven't read the rest, but they seem to be considering that there is a lot more data sharing happening?
♥ 1
Replying to @e2dot7182818284 and @big_data_kane
Oh don't transcribe it on my account! I'm sure that I can find another way to find out about his thinking. You are too kind.
Systemic collapse is precisely one of my concerns, from excessive simplification.
♥ 2
Replying to @e2dot7182818284 and @big_data_kane
That's a lot of good stuff. It's why I've always loved the complexity world.
♥ 1
Replying to @_alialkhatib
I wish we could do meetings this way.
Replying to @varlogsimon and @gsterling
It's... only for Chrome 😭
Replying to @MaxGendler
If only it were limited to that 😁
Data brokers are one concern. But what about legal subpoenas? The pervasive monitoring for instance in Chrome and Android (to cite just the two biggest threats) makes listing who has been to a clinic, illegal abortionist, or abortion site easy.
What's the protection?
Quoting a tweet by @motherboard ↗
♥ 13↺ 3
Replying to @null and @benharnett
What year are you tweeting from?
Replying to @swodinsky
If you're pregnant (or even just your regular cycle skips a couple of times), and then you're not that could be enough — at least probable cause.
♥ 23↺ 5
Replying to @kashhill and @swodinsky
Law enforcement might not go dragnet, but anti-abortion groups will whenever they can, no?
♥ 10↺ 1
Replying to @kashhill and @swodinsky
A blend of contextual/interests targeting and data buys, honeypots, if well-funded or technically sophisticated enough possibly bidstream-level?
♥ 10
Replying to @swodinsky and @kashhill
I'm not thinking that the app would segment that way, but get the pregnant audience and retarget that segment against abortion keywords is one thing I have in mind. Or delta the pregnant audience with new-mother audience.
♥ 22↺ 3
Replying to @swodinsky and @kashhill
For honeypots, yeah, AFAIK it's FB+search. But with a pregnant segment you could get creative. I also assume that the most at-risk populations are less likely to be iOS, and so have a broad ID graph.
♥ 1
Replying to @swodinsky and @kashhill
Thanks. I'm not sure at what point I thought that becoming good at dystopia was a healthy life choice, but I appreciate the kind words :)
♥ 5
Replying to @swodinsky and @kashhill
I think you want to think like a marketer. What's your illegal abortion funnel? (Threw up in my mouth a bit there.) Keywords will be noisy — people thinking about it not following through, unrelated searches, etc.
♥ 6↺ 1
Also consider that in some states it's going to remain legal if it takes place within some (ridiculously short) time frame — having a pregnancy date would make it easier to pick off who is out of bounds.
♥ 2↺ 1
Replying to @swodinsky and @kashhill
Starting off from an AdID also probably gives you a much higher reidentification rate than cookies, I think.
♥ 4
Replying to @swodinsky
Sorry :(
One thing that's disheartening (apart from the whole fucked up thing in itself) is that this is pretty much exactly what privacy folks have been warning about since basically the dawn of time.
♥ 14
Replying to @swodinsky and @kashhill
Cross-device matching also means that you can match against IP addresses for people who thought that "Incognito Mode" would protect them.
♥ 6
Replying to @_alialkhatib and @swodinsky, @kashhill
That is true, but what specifically worries me about pregnancy app data is that in conjunction with relatively little other data I guess you would have cause to subpoena more. And then Android location + Chrome Sync are the helpful automated Stasi — you're done.
♥ 1
Replying to @swodinsky and @_alialkhatib, @kashhill
Yes exactly! Some fascist millionaire could easily bankroll this, and there are plenty of people with the marketing skills. Then just give the (complicit) cops a list with the evidence you have and they can do the rest.
♥ 4
Or better: you can just roll abortion detection up into predictive policing software.
♥ 1
Replying to @MaxGendler and @swodinsky, @_alialkhatib, @kashhill
I don't know man, I'm still not sure that Sauron got fair coverage.
♥ 3
Replying to @RobertJBateman and @swodinsky
Thanks Robert! It's draining, so it's good to hear that you find it worth doing.
♥ 3
Replying to @histoftech
I keep wishing we had a nonprofit that did nothing other than write good-looking, user-friendly free apps that don't do dystopia.
♥ 11
Surveillance technologists from tech companies big and small have delivered amazing support infrastructure for fascism. @swodinsky helps you navigate this world.
gizmodo.com/how-to-get-an-…
♥ 9↺ 4
It's important to read Aram's thread about the very real threat that adtech poses to uterus-havers post-Roe, and consider it in the context of the narrative that the adtech industry (at least those who still refuse privacy) has been pushing around this tech.
Quoting a tweet by @Chronotope ↗
♥ 31↺ 16
One key tenet is that "people want relevant advertising" and that is somewhat true (within limits).
But no want is absolute, everything has a cost. Do people want relevant advertising that puts women on death row?
Quoting a tweet by @nicoleperlroth ↗
♥ 9
Another common point is "but there are plenty of *good* people in adtech, I sleep well at night, we don't do those things." And often it's true.
But unless the only market in which you operate is Care Bear Land, we still protections against the many bad actors.
♥ 5
Replying to @publictorsten
Yes and yes.
Replying to @salome_viljoen_
The level to which this is *exactly* what privacy folks have been saying forever is, quite frankly, depressing.
♥ 1
Replying to @mgahntz
Yes, I've been thinking about that. And same for state-level legislation in those states.
Replying to @publictorsten
Yes.
Replying to @null and @blakereid
You're not cut out to be an adtech lawyer. That's actually the position: you need to be really sure, GDPR-level consent, that people want to opt out of their data being shared with third parties.
♥ 4
Note that this isn't new, the IAB advised the same position. You've gotta admire the ethically-spineless YOLO shamelessness to it. I don't know if people are born that way or if the money really is worth it.
♥ 3
@TomKemp00 is right: data brokers are evading CCPA enforcement. One trick they have is to get publishers to list them in notices so they can claim to be just third parties. That loophole needs closing.
♥ 2↺ 1
Replying to @PrivacyMatters
That position was made in the ongoing CPPA stakeholders panel. It's… creative I guess?
♥ 2
Replying to @megangrA
Thank you!
The CPPA stakeholder session on opt out has an interesting dynamic: publishers aligned with consumer orgs to support privacy — better for both our readers and our business — and legacy adtech with increasingly desperate arguments to maintain their chokehold on advertising.
♥ 17↺ 5
I know that many in adtech want this system to change. A growing number of adtech companies understand that they don't have to be user-hostile to deliver.
If I might ask: could you tell your trade bodies? Right now, the @iab, NAI, DAA are really holding you back.
♥ 20↺ 8
@Chronotope showing code on the record: the seven lines of code that WaPo used to implement GPC. You can even steal them.
That's what the adtech industry groups call "complex."
♥ 20↺ 5
Replying to @null and @catsoo
I know! The French jokes just write themselves.
Replying to @ZackNewsMan and @swodinsky, @_datamimi
Yesterday, I didn't think it could get much worse, but I hadn't considered bounty hunters.
♥ 2
Replying to @ZackNewsMan and @swodinsky, @_datamimi
Yup. Or what car companies do with the geolocation that they constantly collect!
♥ 2
Replying to @ZackNewsMan and @swodinsky, @_datamimi
The part of the story they don't tell you in "curiosity killed the cat" is that the cat was actually curious about adtech, and asked @swodinsky.
♥ 1
Replying to @swodinsky and @ZackNewsMan, @_datamimi
They should publish your collected works in a nicely bound volume :)
♥ 3
Such a Princeton experience: a presentation for all elementary school parents that's full of maths, p-values, error bars, and has a references section. 😂
♥ 7
Replying to @null and @MadronaCitizen
It was actually very clear, but the style is definitely unusual!
Replying to @privacyguru and @swodinsky, @kashhill
That might be an option, yes. I see two challenges: 1) this has become harder than it used to be, you would need real scale, and 2) the people doing that in those states might be at legal risk.
But maybe our AdNauseam friends have idea? @HNissenbaum maybe?
♥ 6
Replying to @daniellecitron and @wwnorton, @UVALaw, @UVALawTech, @VonCleland
Oooooh, it looks amazing too! Looking forward to reading it!
♥ 3
Replying to @montezumachavez and @keir_lamont, @tatiianarice, @staceygraydc, @CalPrivacy
They were!
♥ 1
Replying to @Chronotope and @swodinsky, @Bryson_M, @daphnehk, @ireneista, @mberk, @monteiro, @monteiro
is great! I also recommend ruinedby.design.
♥ 4
Oh, and this isn't adtech ethics overall but specifically on consent I recently made this: youtube.com/watch?v=y3k4sd…
I'd totally be down to put together some ethics for adtech thing, it could be hilarious and grim in equal parts! We'd use your collected works as cases.
♥ 6
Replying to @tariqajyusuf and @ireneista, @swodinsky, @mberk, @Chronotope, @Bryson_M, @daphnehk
It's an early draft (not officially announced yet) and a bit all over the place in quality but I'd love this group's feedback on w3ctag.github.io/privacy-princi…. It's state of the art privacy framed for Web nerds, or at least that's the goal.
♥ 6
your expertise in making things readable for humans would be particularly welcome if you have thoughts!
♥ 5
Replying to @tariqajyusuf and @ireneista, @swodinsky, @mberk, @Chronotope, @Bryson_M, @daphnehk
Ouch, my sympathies, been there a few weeks back!
♥ 3
Replying to @ireneista and @tariqajyusuf, @swodinsky, @mberk, @Chronotope, @Bryson_M, @daphnehk
Do you mind filing these in GitHub so the group can discuss? Otherwise I'll do it. Good points all, thanks!
♥ 2
There are multiple antitrust bills in play right now. Does anyone really expect Google or Facebook to choose women over Republican support?
Quoting a tweet by @AASchapiro ↗
♥ 5↺ 2
Replying to @ireneista and @tariqajyusuf, @swodinsky, @mberk, @Chronotope, @Bryson_M, @daphnehk
Thanks a lot! No rush, it's the weekend 😁
♥ 2
Replying to @emilybell and @justinhendrix
Avocado is particularly vicious in how quickly it flips between the two states.
♥ 1
Replying to @tariqajyusuf and @ireneista, @swodinsky, @mberk, @Chronotope, @Bryson_M, @daphnehk
I have a glass of whisky in the other hand, so clearly this must be relaxing times!
♥ 3
Given the institutional process in Chile, I kind of want to go work for that DPA. ❤️
Quoting a tweet by @montezumachavez ↗
♥ 5↺ 1
Replying to @ireneista and @tariqajyusuf, @swodinsky, @mberk, @Chronotope, @Bryson_M, @daphnehk
Thanks a lot!
♥ 1
Replying to @montezumachavez and @alextcone, @JEsquenaziMX, @PrivacyMatters, @EU_EDPS, @VeroCimina, @CMAgovUK
I would wish that the synergy would lead them to *not* have overlapping events in different countries!
♥ 3
Replying to @charlesspray
That's a really cool idea. I wonder if it could be like a home implementation of the Solid protocol stuff?
Replying to @charlesspray
Yes! Sorry I wasn't clearer. It's been a while since I looked but your idea reminded me of pods, and the ability to switch between home storage and online pods feels like it could be useful?
Replying to @AureliePols
J'ai mes doutes sur l'IAB parce que leur marque de fabrique est d'être pire que leurs pires membres histoire de les couvrir, et sur les groupes buy-side qui sont à la ramasse niveau tech. Mais le Tech Lab et les groupes de publishers bougent, donc tout peut changer!
♥ 2
Replying to @AureliePols
The initial data controllers, but not always in the financial position to make enlightened decisions. That's why it takes time for everyone to get with the programme.
♥ 1
This (and the rest of the thread) is also true for many other issues in tech today.
Quoting a tweet by @random_walker ↗
♥ 3↺ 1
Replying to @PulitzerPrizes and @QuantaMagazine, @nattyover
This is *super* deserved, congratulations!
♥ 1
In general, reading Ostrom or around is something I have found very rewarding. A good book length intro is bookshop.org/books/the-unco…
♥ 2
My favourite from Ostrom herself is bookshop.org/books/understa…. Then there's a lot of great academic work.
♥ 2
But yes, fundamentally you only get a tragedy if the agents are made to be in a prisoners' dilemma. But that's rarely the case unless it's created! It's a characteristic of online mechanism design to falsely create PDs that don't need to exist. Google are experts.
♥ 2
I forgot: full disclosure is that I want to make an advertising commons, and also use that elsewhere on the web.
♥ 2
Could the Web be governed better? It needs to be.
Can we reform the W3C? It's happening as we speak.
So could we reform the W3C to bring about better governance for the Web? It requires change, but I think it's doable. Read on for some thoughts!
berjon.com/w3c-governance/
♥ 75↺ 27
Replying to @karlgroves
The problem with standards is that a lot of the real digging involves talking 😁
♥ 1
Replying to @null and @dauwhe
Thanks 🤗
Replying to @yoavweiss
Yeah, I'll admit I ran out of steam around 5k words… I have another post on commons teed up, I think it might help explain domains and polycentricity. Or we need a whiteboard.
Re funding you're totally right, but I think the core point is there.
♥ 2
Replying to @alextcone
I wish I were more disciplined about writing things down — it's fucking hard!
♥ 2
Replying to @alextcone
That's a good question. The Web does not have a well-defined scope, which makes for interesting issues. Technically, any tech that abides by the values could be in scope. I'm comfortable leaving things open ended and seeing what happens.
♥ 2
Replying to @robleathern and @btsavage
What part of "you don't want to know" isn't clear, Rob?😋
The W3C doesn't exist, it's a contractual arrangement between four hosts (MIT, ERCIM, Keio, Beihang) that collect dues and pay for stuff. It's messy.
♥ 3
Replying to @rdeltour and @innovimax
I think there is more money in Internet governance structures than people realise, that's one source.
Then I do think that vertical domains should use (very small) taxation. GARUDA has that for the ads domain; I think the browser/search space could use the same.
♥ 4
Replying to @hober and @robleathern, @btsavage
That's on a good day and before I get to the part about the percentage of donations that some hosts take for themselves…
♥ 2
Replying to @robleathern and @btsavage
Yeah, it hasn't changed yet, but it's in the process of changing.
I've been looking at DAOs too for this, and there's a lot of interesting work, but I want to focus on figuring out the governance first. The rest is implementation details.
♥ 2
The impression I'm getting from a lot of co-op/blockchain world is that the blockchain part will either turn out to be unnecessary or a convenient implementation detail but only rarely core. We'll see!
♥ 3
Replying to @mc2hampion
I don't know, but it's a good question. I think this corresponds to what the W3C has been trying to be, and I'm pretty sure that it will be much more successful pivoting in that direction — but it's true that it will take some convincing.
♥ 2
Replying to @robleathern and @btsavage
That's absolutely the appeal, but that only works if the contracts are exactly right. Real-world governance works *because* there's imprecision and wiggle room; perfect contracting is a fiction. Society breaks in the ordered regime.
♥ 5
Replying to @mc2hampion
Take a break so you recover from the burn out, I'll cover the stewardship issue in the meantime, and we can join forces when you come back refreshed!
♥ 4
I mean either that or we just fork W3C — I see no value in a Consortium that doesn't work to solve the Web's problems.
♥ 3
Replying to @mc2hampion
I think that's basically where I want to land eventually too!
Replying to @samspearsevans and @yoavweiss, @cwilso, @hober
We have plenty of those. I'm just pointing out that they're a bad fit for the Web.
If you're into marketing analytics (or even just analytics, and curious about marketing), I have several open positions on a mighty good team!
• Analyst (Owned Media) nytimes.wd5.myworkdayjobs.com/en-US/NYT/job/…
• Senior Analyst nytimes.wd5.myworkdayjobs.com/en-US/NYT/job/…
• Director nytimes.wd5.myworkdayjobs.com/en-US/NYT/job/…
♥ 1↺ 1
Replying to @RIngthorss
I get ads for periods underwear and sports bras, because "Robin" is a woman's name for Americans, so…
♥ 1
Replying to @cwilso and @yoavweiss, @hober
You're absolutely right that we have to transition — but we have to transition to something that works and has a sustainable future. That requires working towards a Consortium that offers more than a place for Apple and Google to sort things out.
♥ 1
I am keenly aware that outsize influence comes from control over infrastructure. That's why I propose to solve this problem instead of just leaving it by the wayside.
We can broken browser/search dynamic or we can decide the Web is Google's other proprietary system.
There isn't really an option that doesn't simultaneously keep the Web open and doesn't address monopolistic and unfettered control over the infrastructure.
I mean, if we don't, what is it that a standards organisation does exactly? Rubber stamp things?
You seem to be assuming that I don't know that this is hard, and that I don't understand that it needs money. I do. I have more detailed plans on the infrastructure funding side for ads, search/browser, and (partially) social.
This is the simplest change that I have been able to devise that doesn't pretend we don't have a power problem. I would be absolutely overjoyed to consider better options, I'm sure they exist. But what I don't see is the value of having a W3C that's just decorative.
Why bother?
Replying to @cwilso and @yoavweiss, @hober
I don't think that having Google engineers working on the platform is a problem at all. I think that Google having the power to get away with Chrome Sync, AMP, CWV, etc. is the problem. Even with the best intentions, that kind of power is clearly being misapplied.
Replying to @cwilso and @yoavweiss, @hober
Option (3). It requires lightweight regulatory intervention to work, though. The only downside for Google is that you would be forced to compete on the merits in Search.
Replying to @cwilso and @yoavweiss, @hober
These are different times. Microsoft never used IE to push an identity system across the whole Web, they never collected data about other people's businesses. Adjacent dominance doesn't require 98% marketshare to apply excessive power!
♥ 1
Replying to @cwilso and @yoavweiss, @hober
I'm not sure how Chrome Sync isn't a problem? Forget the fact that it's a privacy nightmare and that it pioneered the use of dark patterns in browsers, you have no business collecting information about behaviour on my site.
But if you prefer other issues, the fact that Chrome pushes Google identity across the whole Web (including reviving the cookie that powers it) is another one.
Replying to @cwilso and @yoavweiss, @hober
Fair. They didn't get anywhere with that though! The Microsoft precedent has often made me wish Google would focus on Android.
Replying to @cwilso and @yoavweiss, @hober
Does Chrome need Google, or just the money?
Replying to @cwilso and @yoavweiss, @hober
And there's no question that the money exists, and I think that it's not that hard to get to in the grand scheme of things. Indie Chrome would be a pretty badass browser!
♥ 1
One big difficulty is preventing a search app with in-app browser from taking over. But that's already a major threat today anyway.
♥ 1
Replying to @cwilso and @yoavweiss, @hober
Oh, my solution is market based baby. Or at least it's not some fluffy hope that money will materialise. I mostly replace the right to buy search dominance with the right to be an option in a choice dialog.
Give me a bit of time, I have other things to write first.
♥ 2
Replying to @btsavage and @robleathern
Agreed! That's the thought. It would be hard because consensus is painful, but we can't both complain that regulation has issues and sit by the sidelines. But making it work means it can't feel like it's the same as lobbying think tank fodder.
♥ 5
Replying to @WKCosmo
It was good! Thanks for writing it. I think it's the first time I've seen someone paint that complete a picture of this topic at once, and I feel less dumb for having read it.
♥ 1
C'est @criteo qui vient sur ton site — et PAF la vie privée!
♥ 4↺ 2
Oui, ceci est une blague extrêmement adtech :)
Contexte: @criteo a rejoint le mouvement réactionnaire anti-vie privée "SWAN". SWAN est la risée du monde des éditeurs et de l'adtech depuis des mois, d'où sans doute le changement de nom vers "PAF."
♥ 3↺ 2
Comme SWAN/PAF n'a aucune crédibilité, ils ont manœuvré pour s'adosser à @prebidjs (le "P" de PAF), un projet pragmatique et techniquement compétent.
Les gens que je connais qui sont impliqués dans Prebid sont verts de rage. Ils craignent (à raison) pour leur réputation.
♥ 3↺ 1
Replying to @robleathern and @btsavage
Looking forward!
Un article très fouillé et très intéressant de @pixeldetracking sur le consentement, le TCF, et l'exemple Sirdata.
pixeldetracking.com/fr/le-consente…
♥ 6↺ 3
Replying to @porteneuve and @nhoizey, @criteo
Oui mais par contre c'est lent et fragile! 😂
If you want to build up your ability to understand and think about DAOs intelligently (not just 😘 or 🤮), I highly recommend this thread from @robleathern).
Adding a few notes on the topic below, but if you read only one read his thread:
Quoting a tweet by @robleathern ↗
♥ 11↺ 3
Why is there so much interest in governance?
We built this whole digital world in a mostly ungoverned way. What happens when you don't have governance is that might makes right and & rich get richer. Governance helps fix imbalances of power.
♥ 1
Establishing a better balance of power isn't something worth doing just for the aesthetics and fun of it. Imbalances of power stifle intelligence and discourage investment and innovation.
Even if you're amongst the powerful, smothering the Web kills the goose.
♥ 2
This is the truth. This is why people who work for a better world talk about being tired all the time. It's why folks give up on standards.
Quoting a tweet by @spamaps ↗
♥ 11↺ 2
Replying to @null and @JudMarc2, @WKCosmo
Maybe it's mind reading, maybe there's a cause in common!
Why isn't "Historians Calling Bullshit" a core feature of, like, basically everything?
Quoting a tweet by @Cevasco_Carla ↗
♥ 7↺ 2
Replying to @mguariglia
I know! And it's delightful. What I'm saying is basically that it's an art form seen on Twitter that should be in way more places!
♥ 2
Replying to @mguariglia and @madebyhistory
Yes! There should be a hotline that you can call when you're stuck in a dinner or conversation with someone with BS takes.
♥ 3
Replying to @porteneuve and @nhoizey, @criteo
Ça ne m'étonne pas, j'ai vu plusieurs présentations sur SWAN et c'est complètement un véhicule idéologique construit autour de la haine de ses auteurs envers la privacy, ça n'a jamais été conçu pour être utile.
Replying to @DerekNonGeneric
That's not the core issue. The Host system has been problematic for a long time and switching away from hosts is good and should have happened sooner. MIT is just walking because the transition is taking forever.
♥ 1
Replying to @roessler
The part about the Tories doing it too is hilarious.
Replying to @reidmore_online and @wendy_a_reid
Thank you, that's too kind!
This is what people who don't work on standards don't understand.
Quoting a tweet by @bradfitz ↗
♥ 22↺ 3
The @w3ctag has just published a first draft of its "Privacy Principles." I take a look at what the task at hand is and what my personal view is with respect to privacy principles for today's digital environment.
berjon.com/principled-pri…
♥ 27↺ 15
Replying to @gizmomathboy
Me too!
Replying to @doctorow and @bengoldacre
See also papers.ssrn.com/sol3/papers.cf… if you haven't already.
I you were interested in my thread mentioning concerns (and promise) with DAO-style tech and @robleathern's discussion, I highly recommend this long but rewarding thread from Cory.
Quoting a tweet by @doctorow ↗
♥ 12↺ 2
My thread:
♥ 1
Replying to @smwat
Implement?
Replying to @null and @CodeforPDX, @avilarenata
Good question! I would say you need to have a clear governance model that works (but might be clunky) without the DAO, and only go DAO once that's done and you know it's implementable.
Replying to @GeradinLaw and @DinaSrinivasan
The ad industry way to think about it is that whether it's more or less than 5% is immaterial, you pick the auditing mechanism that will produce the result you want!
♥ 5
Euh...
♥ 2↺ 1
Replying to @null and @dansereduick
ouaish ouaish quoi
Replying to @kateconger and @paraga
I don't know, maybe it's the comms style that's tripping me but it feels like he's spreading a tweet's worth of information over a whole thread? Like, "Deal or no deal we're not on pause, it's all BAU."
♥ 2
With respect, this is a bad take.
If you make a law that requires businesses to annoy users in order to access their primary source of revenue, that's exactly what you'll get. Incentive structure matters, enforceability too.
Quoting a tweet by @MissIG_Geek ↗
♥ 28
I want to get rid of all this crap as much as @MissIG_Geek does. Consentgating it is a terrible way to get there. Unless there is a way to be competitive entirely under legitimate interest (and there really isn't) you're going to get the hellscape we got.
♥ 6
That was 100% predictable.
To get results, we need outright prohibition of the worst stuff, technology that enforces purpose limitation, and the means to deliver ads without bad UI requirements.
♥ 13↺ 1
Replying to @acruiz
You're right, I completely agree that it shouldn't. Which is why consent is a terrible way to get there.
♥ 2
Replying to @null and @TrialByTruth
I hear you, but I'm not sure that this is the best case to make? Though to be honest I'm not sure that there's any rational case to be made for those muppets...
Replying to @null and @MadronaCitizen
I don't think that that's the dichotomy. A problem can be entirely real and yet the solution bad.
We could fight climate change by requiring every polluting vehicle to blast the macarena top loud. But we'd just get a lot of macarena.
I'm definitely not saying that the GDPR is alone responsible. But we need to make it work better. It's possible within the GDPR framework, but it needs DPA changes.
Replying to @null and @TrialByTruth
I'm aware, and it sucks big time. Correct me if I'm wrong, but I think there are two problems. One is to make privacy work, the other is to convince your government not to be stupid.
Let's leave the former aside for now...
♥ 1
I honestly have no idea how to make Brexiteers have intelligent opinions, but might it be a better approach to point out that the GDPR does not require consent in this way, that's largely a DPA decision? The ICO could keep the law and decide to make it work better.
♥ 1
Replying to @null and @MadronaCitizen
Yes! You can build it, but you can't make money. We can fix that though!
Replying to @null and @MadronaCitizen
It's not what caused the business need, but it's what was offered to solve its problems - and it doesn't. Institutions design markets.
The incentives are, as often, to race to the bottom. Effective law is supposed to stop that, otherwise it's just wishful thinking.
If we did food safety the way we do privacy regulation everyone except a fringe food-educated elite would be eating cyanide-laced asbestos wrapped in lead paint because they agreed to do so when they walked into the supermarket.
That's not a good system and we can do better.
♥ 11↺ 5
You're all making jokes, but how are we going to replace every single statistician on Twitter who just died of facepalm-induced trauma?
Quoting a tweet by @elonmusk ↗
♥ 15↺ 1
Seriously, though, @Twitter, if your content moderation policy doesn't include banning someone for asshat stock manipulation, add those terms and kick him out.
♥ 16↺ 1
Replying to @gabbistubbs
Yes, it's a cautionary tale about quite a few things we shouldn't be tolerating as a society!
♥ 2
For all you morons who've wasted your lives agonising over random number generators, the ontological meaning of randomness, or proper sampling methodologies.
♥ 26↺ 3
Replying to @e2dot7182818284 and @Twitter
I'm pretty sure they've all died of facepalming too by now.
♥ 3
Replying to @clancynewyork
That's why I kept it in the frame... I'm laughing, I'm crying, it's everything.
♥ 5
Replying to @null and @MadronaCitizen
♥ 1
Replying to @AveryOfBirds and @struthious
You can make a wall of lava lamps and use that as a decent randomness generator, so I would say that lava lamps have social utility.
♥ 2
Replying to @AveryOfBirds and @struthious
It's kind of you to keep defending him, but I still think the random goo would do better.
♥ 1
Replying to @clancynewyork
I suspect that might be why neither of us is a billionaire 😁 (And I wouldn't have it any other way!)
♥ 2
Replying to @AveryOfBirds and @struthious
AND they give off that soft, snug, glowing warmth. You've really sold me on this whole thing of lava lamps running businesses.
♥ 1
Replying to @AveryOfBirds and @struthious
No, no, no, we have to seek a better world, we have to dream! The LED may match those laser eyes of the bitcoined, but we must soar at least to phosphorescent!
♥ 1
Replying to @MaxGendler
Max, say hi to @fran_mady when you get that job as Twitter Head of Bot Detection!
♥ 3
Replying to @MaxGendler and @fran_mady
Replying to @null and @TrialByTruth
Yes, agreed on all point. That's why I feel helpless: it's where they want to go and they are impervious to facts or rational arguments. Do you see a way to stop or steer that?
Replying to @null and @TrialByTruth
I only hear you as constructive, and I'm sorry if my attempt to engage constructively didn't work or worse backfired. I would prefer a different core message, but also might it not work better strategically?
"It's not the GDPR's fault" or points about ePrivacy feels like exactly the feedback they would expect, and is exactly where they thrive. Using a message they don't expect might carry more weight?
I realise I'm saying this safely in bed in another country.
Replying to @floorter
That is true, but enforcement strategy is part of regulatory design. You can't build law as a Platonic ideal and hope that when it comes into effect everyone will magically align to the intended outcomes. It interacts with reality.
♥ 4↺ 2
The idea of consent is that you rely on the party that has an interest in doing it badly to do it well, knowing that it's a complex, nuanced HCI topic that people often fail at even when they're trying to do helpfully!
♥ 2↺ 1
People tried asking the user for security purposes, a space in which the risks are vivid and relatively easy to explain to people. It failed spectacularly. Why would it work, even with good intent (that isn't there), for privacy which is harder to explain?
♥ 1↺ 1
Replying to @_aurelienfabre
Bien meilleur que Ouai. Je vois une gamme: "Putain...", "TMTC", que du lourd.
Replying to @null and @dansereduick
I *completely* agree, my point is that addressing that throught consent will fail predictably.
♥ 1
Replying to @EerkeBoiten
It's not about accepting the premise, it's about the fact that if you build a law in such a way that the people who have the least interest in doing it right are in charge of deciding who bears the cost, they'll offload it. Here, to publishers and people.
Replying to @PrivacyMatters
I think we can blame the law for not fixing the problem it purports to fix? A law that doesn't prevent bad actors from racing to the bottom and externalising costs isn't a good law in any domain, why would we expect it to work here?
♥ 3
Replying to @PrivacyMatters
I know! Which is why the outcome was predictable, no? I feel we went through a moment of "Consent is like violence - if it doesn't work, use more!"
Why not just ban the fuckery?
♥ 2
Replying to @minabird
Various parts of ad delivery, for one. But I have hope that things are looking promising in the PATCG here.
Replying to @mireillemoret
I know, but 1) the definition of "necessary" doesn't include making money and 2) that does nothing to address the incentives of bad actors.
♥ 1
Replying to @MissIG_Geek
Oh, you were only making the GDPR/ePD distinction? My sense is that most consent prompts that people are annoyed with are not just for cookies.
I certainly agree that surveillance is the problem, fixing it with transparency & consent too.
♥ 1
Replying to @MissIG_Geek and @TrialByTruth
Sorry, my bad for clumsily stepping in. I'm interested in the asteroid strike thing if there's room to help.
♥ 1
Replying to @floorter
I think your second sentence is the reason for the first! I mean, I agree, with A LOT of resources we could make consent work. We'd basically need an IRB for the Internet. I just think that we're not going to get those resources and so can figure out more frugal ways.
Replying to @floorter
Yes: I think we should only use consent in rare cases like immediately capturing data in a form. Other processing should be regulated through more effective means.
♥ 2
Replying to @minabird
That could work for direct sold (it's not completely far from what we do in Europe), but you need browser support to solve programmatic.
Replying to @MissIG_Geek
That's fair. My (very compatible) rant is that we need privacy laws to think more adversarially about bad actors. I recently heard someone claim that the ATT prompt is a dark pattern because it makes tracking look bad and defaults to safety.
People like that exist in the same intellectual universe as Brexit. Reality doesn't matter, people don't matter, only pushing their views matters. They don't care that they're building a world of monopolies and toxic data leaks. We need enforcement leverage.
♥ 1
Replying to @_aurelienfabre
Et la crème de visage s'appelle juste "Facial" parce que ça fait toujours rire un peu.
Replying to @null and @dansereduick
I think we can make the GDPR work, the trick is to be opinionated about Art 25 and build that into the tech layer. We can't leave it all to DPAs, we have to fix the architecture.
♥ 1
Replying to @MissIG_Geek
I don't know, I think we're making progress 😀
Replying to @MissIG_Geek
I relate to that. I often consider cutting my losses, abandoning tech to its totalitarian overlords, and switching to something where I can make a difference. But we can't afford the societal collapse that would bring and I still see a narrow path.
♥ 1
Replying to @floorter
Agreed. People confuse the necessity of the outcome with the necessity of the processing.
♥ 2
Replying to @PrivacyMatters
You should blame both!
I agree that enforcement is abysmal, but what kind of budget would it require to enforce properly with today's tech stack? A really big one I think. That's why I think we need to move enforcement into the tech layer.
♥ 2
Replying to @MissIG_Geek
I very much doubt that your moral virtue is illusory!
Replying to @MissIG_Geek
We're going to have to disagree on that! I think there is such a thing as the good life.
Replying to @null and @fran_mady
Ouai voilà.
Replying to @null and @fran_mady
And that's just for dates!
Replying to @PrivacyMatters
Agreed, agreed, that's why we're trying to provide some clarity for Web folks at least.
♥ 2
Replying to @null and @fran_mady
And there are standards in everything, and they all have issues. The closer you look, the more amazing it is that the world doesn't just fall apart 😂
Replying to @null and @benharnett, @fran_mady
You bring up a good point.
Replying to @null and @fran_mady
There are many reasons. The initial standard may have been poorly specified, or didn't exist. Or a major implementation had bugs. Or someone annoying on the committee pushed something stupid in. And once there's content in the real world, you often have to keep supporting it.
We have jokes about it, thankfully! datatracker.ietf.org/doc/html/rfc69…
See also xkcd.com/927/.
Finns have a joke in which God creates the peoples of Earth, giving each a balance of good and bad traits. But then He makes Finns perfect in every respect. An angel asks why the exception?
God: "You should see the neighbours I'm giving them."
Quoting a tweet by @minna_alander ↗
♥ 6↺ 1
If you're looking for extensive #Eurovision coverage with live updates: nytimes.com/live/2022/05/1…
♥ 2
The best part of the Portuguese song is the hat. #Eurovision
♥ 2
If you ask me, @NATO could use some of that metal. 🇫🇮 #Eurovision
♥ 2
TFW you Brexit from #Eurovision
↺ 1
Gotta trust the metalheads to vote! 🇫🇮 #Eurovision
♥ 1
UKRAINE 🇺🇦!!! #Eurovision
♥ 4
On purely sartorial grounds, I stan Italy's win. 🇮🇹 #Eurovision
♥ 2
Replying to @null and @dansereduick
Brilliant.
Replying to @JEsquenaziMX
It does. A law that would address issues from drug trafficking by requiring cartels to put their produce in hard-to-open bags with a health warning on them would be pretty damn useless.
♥ 1
Replying to @JEsquenaziMX
Not banning - we don't ban drugs either, that would be extremely dumb - but regulating usefully.
Replying to @trekonomics and @moiragweigel
I agree the quai are better, Luxembourg involves breathing a lot of dust.
♥ 1
Replying to @trekonomics and @moiragweigel
Oh, I'm afraid I don't have much in the way of good tips for morning people 😁
♥ 1
How Australia Saved Thousands of Lives While Covid Killed a Million Americans nyti.ms/3LeRDv9
♥ 3↺ 1
The feedback on my sketch for reformed W3C governance has been very encouraging so far. In response to some of it, I have made a few changes to the proposal which I call out in the thread below.
berjon.com/w3c-governance/
♥ 3↺ 1
Several people asked if it was modelled after the IETF. Not intentionally so, but some of the people I have spoken with are close to that and it was a very clear influence, which I now note. The IETF has shown that community governance works and is fundable.
♥ 1
asked for more information about the participation of public or governmental organisations. I added a fair amount of detail to the Regulatory Recommendations section. I notably think that we should collaborate with DPAs more to make privacy regulation work better.
♥ 2
helpfully pointed out that the @W3CAB has already been advocating a values-based approach. I've added a note pointing to that and also explaining why it's not the approach that I would recommend.
♥ 2
correctly noted that it's short on funding and domains. A lot of the meat on these is in governance of specific areas, for which blog posts are coming. I added a note to indicate that there is more and that it will address those parts better.
♥ 2
Thanks all for the feedback, keep it coming! I am lucky to have several chats with interesting people lined up, please don't hesitate to reach out!
♥ 1
All of adtech twitter right now. twitter.com/elonmusk/statu…
♥ 20
Replying to @null and @CharlieGermano
Replying to @cwilso and @W3CAB
I'm not sure that's different. What to optimise for or desire when you can't run the world in two separate configurations to determine which is best is actually a shortcut. That's basically all of virtue ethics (summarised in a way that would get my philosophy licence revoked).
Premature optimisation is the root of all evil (Knuth).
Optimising a complex system is always premature.
♥ 8↺ 1
It's kind of funny how Google's Q1 report about its @CMAgovUK commitments covering W3C issues with Topics forgot to include the substantial feedback about how the API structurally benefits larger players.
It must be in an annex that got lost in the mail.
♥ 10↺ 3
Replying to @hibaymj and @CMAgovUK
I mean, you wouldn't want to lie to a competition regulator about competition matters.
♥ 1
Replying to @chapell68 and @CMAgovUK
That's not the issue. The issue is how much it helps large third parties, not large sites. Sites with more traffic contribute more to the pot than they get out.
♥ 1
OH: "I liked cutting the head off of rats, but I didn't like the statistics."
♥ 3
The Advertising Unfuckers meet!
♥ 36
Replying to @onetrueorder
The ad is literally about Google...
Platforms are a mistake.
Two-sided markets should be owned by their users, everything else becomes grift.
Quoting a tweet by @WellsLucasSanto ↗
♥ 35↺ 6
Replying to @alex_kuhner
Yes, that's essentially infrastructure neutrality. The problem is that switching costs for two-sided markets are very high (each side has to be where the other is) so unless they are owned by their users they all eventually turn into leaches.
Replying to @kkomaitis
This strengthens my belief that we need to be very careful with how we architect adtech features in browsers. If they're complex, we'll only get quality implementations when the vendor has a vested interest in doing it right.
♥ 1
Replying to @m_older and @Uneheuredepeine
Moins cool (c'est dingue ce heist) mais rigolo quand même dans la série strategic reserve: vox.com/science-and-he…
♥ 4↺ 1
Replying to @null and @arainert
You don't seem to be much more on the encouraged to expected spectrum than I am 😅
Un excellent dossier qui explore en profondeur les sources de la misère haïtienne, notamment via le rôle central de la France. L'article est disponible en français et en kreyòl.
nyti.ms/3G3zNKB
♥ 7↺ 1
La classe.
♥ 3↺ 1
Nudgeocratic technocentrism is an apt description of our digital sphere, but read @LMSacasas's full essay, it's worth every paragraph.
Quoting a tweet by @LMSacasas ↗
♥ 5↺ 2
Replying to @montezumachavez and @CMAgovUK
Let's just say that they may have a selective memory 😂
♥ 2
Replying to @null and @ndw
That was considered, but I believe they went with binary RDF in the end.
♥ 1
Replying to @null and @ndw
I think they still have a few formal objections to go through.
The 8yo, sitting by a toy ice cream truck: "I'm playing this game where I had a bad day and I tell the ice cream guy while eating ice cream and milk shakes."
She gets it.
♥ 10↺ 2
Replying to @montezumachavez and @alextcone, @ICOnews
There's no way we can make the Internet sustainable without getting rid of them completely.
♥ 3↺ 1
Replying to @null and @nhoizey
Ha! Mais c'est vrai, je n'avais pas fait la connection.
Replying to @cyberbaloo_
Je fais la même mais au bistro😁
If you're attending @CPDPconferences, go say hi to @k_johnsn and @megan__goyette from the @nytimes data governance team, they're the best!
♥ 12↺ 2
Replying to @pbannist and @slayser8
The new fun twist on this one is people saying that "fingerprinting" is a term invented by privacy absolutists to make a perfectly legitimate practice look bad by association with criminals and, you know, fingerprints.
♥ 13
Replying to @slayser8 and @pbannist
Lost in our ivory towers, we forget that very basic desire of the common people. It's the most frequent complaint I get working on privacy at a publisher. "Why can't we just get tracked more," they yell at me, "it's the only genuine choice."
♥ 12
Replying to @humanpropensity and @slayser8, @pbannist
You only need surveillance if:
• Quality publishers (big or small) should be subsidising disinformation clickbait farms, or
• You have no other means of measuring/attributing.
Publishers, even small, with quality content that can retain an audience can use SDAs.
♥ 6
If we get cross-context privacy preserving attribution then we can anonymise everything.
This would really benefit quality publishing at all sizes, and we'll get relevant ads too.
♥ 1
If your content doesn't make people come back for more or you don't know your audience, the Internet doesn't owe you money.
♥ 3
Replying to @johnwilander and @humanpropensity, @slayser8
Addressable Rhapsody is such a good song.
Replying to @humanpropensity and @Log3overLog2, @slayser8, @pbannist
Gee, you people, now a guy can't have a Twitter take without also fixing the whole Internet? I can't wait for a billionaire to come clean up this place.
More seriously, I don't know that I can square the circle, but I don't think this is a big-pubs only take.
♥ 5
Random notes because I'm tired and also I have a book outline to write.
The incumbent system assumes that publishers are fungible sources of commodity eyeballs. That's great if you produce fungible content for one-and-done visits. It's less great otherwise.
♥ 3
I don't think that this is a big pubs/small pubs distinction. There are big pubs that do bland fungible content and don't build a relationship with their readers and small pubs who do. In the current system, the latter produce signals subsidising the former. That sucks.
♥ 6
Replying to @Log3overLog2 and @humanpropensity, @slayser8, @pbannist
I'm so extremely niche zoologists are still debating whether I exist.
♥ 3
Replying to @slayser8 and @humanpropensity, @Log3overLog2, @pbannist
I agree, that's my assumption for most niche pubs.
But I'm thinking of niche services like something that gives you meeting times in various TZs. They might have high retention but zero context (you can't just sell people calendars). Harder to SDA.
♥ 2
Also note I said *if* that revenue transfer is worth doing. That's hard to answer in the abstract. The assumption that advertising must be the answer to everything is just that, an assumption. This could be WebMon territory.
♥ 1
Replying to @Log3overLog2 and @humanpropensity, @slayser8, @pbannist
Hahaha, wait, which one are you?
♥ 3
Replying to @Log3overLog2 and @humanpropensity, @slayser8, @pbannist
Ah, the unbreachable protection of pseudonymity!
♥ 5
Replying to @TzviyaSiegman
Thank you! I've been looking into DAOs and am curious about this, very helpful!
New book from colleague extraordinaire Chris Wiggins (along with some others) just dropped: Data Science in Context. I'm excited to read it!
datascienceincontext.com
♥ 2↺ 1
It's GDPR Day again!
It's not all bad but there's clear room for improvement. Some thoughts about how we could realistically go about that, knowing that money isn't going to start flowing to DPAs any time soon.
berjon.com/fixing-the-gdp…
♥ 16↺ 12
Putain mais le vote par Internet, quoi.
♥ 3↺ 1
Replying to @olivierthereaux
"J'avoue, on a foiré la première tentative de mettre en place notre système mal sécurisé, et comme en plus c'est codé avec le cul il faut qu'on renvoie ça à tout le monde à chaque fois qu'on fait une mise à jour. Mais sinon genre trop désolés du désagrément occasionné, hein"
♥ 2
Foreigners in the US: can you remind me why we bother again?
♥ 14
Cool, level-headed thread about why dissolving the US is arguably the most promising policy option and why we should normalise discussing it as such.
Quoting a tweet by @BrynnTannehill ↗
♥ 12↺ 2
Replying to @null and @MadronaCitizen
The other things are more unworkable...
Replying to @khuey_ and @fantasai, @ChrisOrlin, @Rich_Harris
Yeah it's tragic how those British school massacres keep piling up.
Replying to @kkomaitis
"What if we found a more personalised way to blame people instead of corporations, but also in a way that allows us to repurpose the data?"
♥ 2
Replying to @joshgnosis and @mnot
I get a sense that Firefox's product roadmap is to slowly remove every feature that made it a distinctive browser worth using until it is eventually left with no competitive advantage. This time it was usable search shortcuts, I wonder what's next on the chopping block.
♥ 23↺ 8
Replying to @NiNanjira
Who knows? Firefox certainly hasn't been helped by its competitors' tactics, but its leadership team has had me baffled for years. They give a whole new meaning to "headless browser."
♥ 7
Replying to @fabricedesre
All the custom ones I set that don't begin with "@" have been wiped out. The list doesn't seem editable anymore. Trying to add one via clicking on a form field and bookmarks yields weird behaviour. Also, they're definitely dead and gone in mobile.
♥ 3
Replying to @realkpfleming
That too! It just stopped doing the right thing and started saving everything straight to the dumpster the way other browsers do it.
♥ 4
Replying to @shepazu
Oh, I actually thought that Thunderbird was doing OK and attributed that to the fact that it's under community management, not really Mozilla.
♥ 5
This study takes "follow the money" seriously when it comes to figuring out why there is so little climate action.
Quoting a tweet by @GregorSemieniuk ↗
♥ 10↺ 3
Replying to @braedon
Mine all vanished and I can't add them back. Looks like they're mirroring the removal in mobile...
Replying to @braedon
I haven't found the energy to look. When they removed them in mobile the folks I spoke with didn't see why they were better than tapping through a menu... It would be great if it's a bug, but given the trend...
♥ 2
Replying to @braedon
Oh that's not one I use. I died on the inside when typing to navigate links went away.
Replying to @braedon
I hadn't even realised that there were browsers that don't have a minimal tab width! I always thought people with those weird seesaw Chrome tabs had an extension. WTAF, how is that even usable?
♥ 2
It's like the people who made the Gmail UI were put in charge of tabs? What the fuck.
♥ 1
I miss Tab Candy. Like most Mozilla things, it got 70% of the way there and then they lost interest, but it was headed somewhere good. And the web would be different if they'd kept Persona.
♥ 1
Replying to @braedon
It's been a long time since I've had more than a handful of tabs in any of these.
People use Chrome because it's the default and they don't know the difference between browser and search, not because they compared. That's why it'll get obliterated by Google Search app soon.
♥ 1
Replying to @braedon
Same, I have an extension to name windows and send tabs between them, but candy would be make switching between contexts easier.
♥ 1
Replying to @braedon
It's addons.mozilla.org/en-US/firefox/…. Not super pretty and has the occasional bug, but I have around 25 windows all with dozens of tabs and it does the job.
♥ 1
Replying to @tante and @jonsjoberg
Exactly. There's no question that their competitors don't play fair, but that's no reason to be hapless.
♥ 1
Replying to @larsmb
Same. I just got to the front of @browsercompany's wait list. They have some interesting ideas.
Replying to @jonsjoberg and @tante
It's a mix, but knowing quite a few people who have worked there, the lack of organisation and focus are always cited. And lack of vision from leadership sometimes too.
♥ 2
Replying to @larsmb and @browsercompany
True true, it's going to be hard to check all the boxes!
Replying to @badboy_ and @tante
As far as I can tell they only work prefixed with @, all existing ones were wiped. I am not bothering filing because the response I got when they removed them in mobile was basically apathy.
Replying to @badboy_ and @tante
You can trigger bookmarks but they don't interpolate search terms, I tried.
Replying to @badboy_
Yup. I had a bunch, updated and they're all gone. Trying to re-add them I can't get the same behaviour, it looks like the same as when they removed it in mobile.
Replying to @badboy_
I figured out how to bring it back! 💕
• Keyword "b", URL bookshop.org/books?keywords…: did nothing
• Made it @b, loaded but not interpolated.
• Deleted, readded: would not let me add as "b" (disabled button) but accepted @b — worked.
• Edited the bookmark to "b" — works!
I'm cycling through all of them to fix them. I'm glad this was a false alarm (even if I have to do work to fix it here); I wish that were true of the overall trend, though :(
Replying to @badboy_
Yes — on mobile I've never managed to bring them back, I forget who I spoke to there but I was told that was intentional.
Replying to @badboy_
That is weird indeed. Could it be some corruption of the bookmarks data? I'm seeing some entries duplicated in there, though not that many.
Replying to @alextcone and @Chronotope, @Log3overLog2, @taubeneck, @pbannist
I'd like to return to this thing that @Log3overLog2 said. Is there any reason why we (collectively, I get that it could be hard for G on its own) don't decide that FLEDGE-like things are what adtech is from now on?
Quoting a tweet by @Log3overLog2 ↗
♥ 2
I say "FLEDGE-like things" because I don't think that necessarily means that everyone gets on exactly the same thing, but we could have a family of options that abide by some agreed-upon principles. (With governance yada yada you know my spiel.)
♥ 5
Every time I think about what an MVP looks like, it's mostly IPA+FLEGAKEET but I don't see Topics being in there. The design space seems too constrained, there's no solution that isn't bad for privacy, competition, utility, or all of the above.
♥ 2
Replying to @Log3overLog2 and @alextcone, @Chronotope, @taubeneck, @pbannist
💗💗💗
Where are those privacy advocates that I could have a chat with them? Because that's not the feedback I've heard from those I've spoken with. I think there's a path?
♥ 3
Replying to @null and @utrenkner
I know, I completely get that — there's been a lot of unfair fighting involved.
Replying to @Log3overLog2 and @alextcone, @Chronotope, @taubeneck, @pbannist
That's not the feedback I've been getting in those parts. Maybe policy folks are less focused on MPC/TEE than privacy people with an infosec background?
♥ 3
Total respect pour la candidature du Parti Breton dans la circonscription Amérique du Nord — ils sont décidément partout!
(À noter pour ceux qui cherchent: la candidate NUPES, @florencergr, est présentée sous étiquette LFI.)
♥ 6
Replying to @taubeneck and @Chronotope, @Log3overLog2, @alextcone, @pbannist
I'm not too worried about "feeling creepy." I think it's @EvanSelinger who made the point that people riding the first trains reported feeling all kinds of ailments at speeds that wouldn't constitute speeding in a school zone today. Those things change.
♥ 1
We're also dealing with the fact that what people feel is creepy actually is worrying in many cases. I think there's a different dynamic if it isn't.
I think that if we fix the security threat (think abortion laws) and the economic model that's a problem today, that's good.
♥ 1
Replying to @alextcone and @taubeneck, @Chronotope, @Log3overLog2, @pbannist
Thiiiiiiis. We probably don't need to fix this as the MVP inside of fenced frames, but I really really really think that we should fix it. Script injection shouldn't be a business model.
♥ 5
Replying to @fabricedesre and @florencergr
Oui, je n'aime pas dire du mal sans avoir testé.
Replying to @taubeneck and @Chronotope, @Log3overLog2, @alextcone, @pbannist, @EvanSelinger
Yes! But I don't think that we can develop a good privacy discussion before the security issue is addressed. There's a host of decisions to make about what the UA and 1P should be allowed to do but it's hard to have them because today there aren't really options.
♥ 2
The privacy discussion becomes a lot more practical once it rides on decisions made by parties that people can readily identify. I can't share numbers but I'm confident that our readers are comfortable with things if *we* do it instead of 3Ps.
♥ 2
If we could have a trustworthy (because purpose-limited) supply chain, that trust could become transitive. Getting to a place where being trustworthy is a competitive advantage for a 1P is a really nice outcome, and changes the privacy discussion IMHO.
♥ 4
Replying to @alextcone and @Log3overLog2, @taubeneck, @Chronotope, @pbannist
You do the pop refs, I'll work on the acronyms.
♥ 3
My expectation of philosophers of technology is that they should understand the implications of technology better than technologists.
Much of the focus on AI is delusional and high on the tech supply. @GiadaPistilli's thread brings me joy.
Quoting a tweet by @GiadaPistilli ↗
♥ 7↺ 3
Replying to @anthonykatsur and @Log3overLog2, @alextcone, @Chronotope, @taubeneck, @pbannist, @IABTechLab
I'm always up for looking at solutions!
♥ 4
Replying to @innovimax and @MonniauxD
Ce n'est malheureusement pas une banalité pour tout le monde, beaucoup pensent que la science ou la technique marchent de façon neutre, d'un point de vue objectif et inchangeant.
♥ 2
I highly recommend this piece on "Revitalising Global Internet Governance" from @kierenmccarthy. His diagnosis is highly accurate and certainly applies to the @w3c.
institute.global/policy/revital…
♥ 6↺ 4
For a high-level sketch of W3C reform that proposes a path to making it relevant to the Web's issues rather than perpetuating the narrower technical approach that is failing to address today's problems:
berjon.com/w3c-governance/