Replying to @fatemehx2
No match for Suburban Dad.
♥ 2
November 2020
466 tweets
Replying to @TzviyaSiegman and @adambroach
You would think they'd get that right!
Replying to @peligrietzer
I think there's a way to balance things. Kids need to learn to find ways to have fun by themselves, they also need to be bored now and then. Everyone in the family gets to have their own life and time.
♥ 1
Replying to @fatemehx2
I haven't seeped in the chtonic powers of suburbia long enough to chop it, yet. 😄
Replying to @null and @gdiazc
Yes, yes, but it's not there yet. Another option is to get rid of this dumb constitution, like normal countries do. Running the same constitution for too long is like not doing your system security updates. Eventually, fascism happens.
How it'll play out:
* Google Search will put YouTube at the top for "election result" searches, because they're terrible at reporting & can't resist self-preferencing.
* YT will carry Trump claims uncritically, because it's money.
* Biggest misinfo op in the shortest time, ever.
Quoting a tweet by @evelyndouek ↗
♥ 7↺ 1
Replying to @ocdtrekkie
If that's the concern, why bother with state electoral law for federal elections?
Replying to @ocdtrekkie
Wouldn't some states still have a disproportionate share of the outcome, probably the same that already have a disproportionate share of the Senate?
I would add: if the proportion is fair, the electors serve no purpose.
Replying to @ocdtrekkie
How is that different? They could disenfranchise picking electors. The trick there is to make it a federal crime to disenfranchise, like they do in democracies!
Replying to @ocdtrekkie
I'm still not sure I get it. If you have 10 electors for your 10 million people, with disenfranchisement you're still sending more of the 10 or more of the 10 million?
There are many strategies theft or murder, too!
Replying to @sinanaral and @jeremyzyang
It's not on the Internet? All I could find is another mention but no link.
Replying to @ocdtrekkie
But that would make things worse: if you have 10 EV and 10m people, 5m for each candidate. You suppress the 5m you don't like. Counting votes, that gives you a 5m add. Counting EV you've gained 10! It incentivises suppression!
Replying to @ocdtrekkie
OK, we're going to need to meet with a whiteboard because the way I'm counting EV guarantees that when you suppress you still get max power, whereas counting votes decreases the effect in proportion to suppression!
Replying to @ocdtrekkie
Well, twice isn't suppression, it's making people up!
Agneau de 7h 😘
♥ 16
A bit of warmth in a lonely year, a bit of light for dark times.
♥ 13
Replying to @fatemehx2
One of the advantages of having spent my teens in the exurb of a banlieue of a boring small town is that something like Princeton still feels like a big city 😄
♥ 1
Replying to @benwerd
Same! Search is still a little rocky and they occasionally don’t have a given title, but overall it’s been really good.
While the intent is no doubt laudable here, this is mostly an example of how not to talk about privacy. A quick smell test is that it doesn’t have “Google” at the very top. A lot is wrong here:
• The idea that there is a %age of personal data one can collect.
(…)
Quoting a tweet by @ianbremmer ↗
♥ 19↺ 6
• No differentiation between first-party and third-party collection.
• No indication of purpose limitation (eg. Ikea might know your address _just_ to deliver, Facebook stole it from your address book for targeting & conversion measurement).
(…)
♥ 4
• No consideration of betrayal by software you’re supposed to trust, like a browser tracking your reading history (Chrome) or an OS your location (Android).
• Boring and sensitive data (eg. phone type and health) given at same level.
(…)
♥ 2
• No indication of pervasiveness (the more constantly you know, the more dangerous).
• Nothing about kids (eg. Google Classroom forces you to sign over your child’s data when it’s the only education they have in a pandemic).
♥ 3
Overall, an assessment of which companies are trustworthy with data and which aren’t would be very valuable to people. But unfortunately it takes more than just a spreadsheet and getting it wrong largely gives the wrong conclusions.
♥ 2
Replying to @gchampeau
Apparement il y a une invasion de frelons dans certaines villes de France où il y a trop de ruches sur les toits. (Parce qu’ils mangent les abeilles.)
♥ 3↺ 1
Replying to @baekdal
That’s not an article, Thomas, that’s a book.
It’s entirely possible that American democracy depends on this decision, yet the live tweet of the hearing has some hilarious moments.
Quoting a tweet by @RMFifthCircuit ↗
♥ 3
Some important information about the @nytimes Needle(s) this year.
Quoting a tweet by @Nate_Cohn ↗
♥ 2
My 7yo thought her T-Rex lamp could eat a banana.
Now put down your phone and go feed your own T-Rex.
♥ 12
There are also some pretty good books out there, waiting to be read. Why not do that? I mean, look at this dog! 🐶
♥ 5
Replying to @null and @SoSillyDaddy1
That's how we science in this family. There were actually huge bananas in T-Rex times. Proof!
Replying to @BofingerDavid
Sexy lingerie could be both. It's when you're targeted for non sexy underwear that you know for sure!
Good luck, my American friends.
♥ 36↺ 3
Replying to @alextcone and @nytimes
All I know is it was before my time, why?
Replying to @alextcone and @nytimes
We’ve definitely been using SSO for several years. I wish we didn’t, but I don’t think that’s a reasonable option now.
I think what you saw is One Tap. That does look way too much like a browser thing, I agree. But Google will blur that line for themselves at every chance.
♥ 1
Replying to @null and @SoSillyDaddy1
Extra cheese free today.
Replying to @alex_kuhner
Hell of a ride!
On November 6, join @johnnyryan, @Chronotope, @DinaSrinivasan, and yours truly for a series of presentations and a panel on “Adtech, Antitrust, and Audiences”. I’m excited to be part of this amazing line-up and look forward to the chat!
wanifraevents.eventsair.com/202011_fr-asso…
♥ 27↺ 9
Replying to @thezedwards and @johnnyryan, @Chronotope, @DinaSrinivasan
Note that that’s not the final summary for my session, it’s just a placeholder from the organisers. I need to figure out what I’ll talk about ;-)
But yeah — I’m psyched to get to share a stage with such a cool crew.
♥ 2
Replying to @thezedwards and @johnnyryan, @Chronotope, @DinaSrinivasan
I think we need more bird proposals! 🦜🦢🦩🐦🐧🦅🦆🦉
♥ 2
Replying to @thezedwards and @johnnyryan, @Chronotope, @DinaSrinivasan
I actually have an acronym that spells SWAN. I’m trying to figure out if it’s crazy enough to actually write it up as a proposal.
♥ 2
Replying to @thezedwards and @johnnyryan, @Chronotope, @DinaSrinivasan
Man, how did you guess!!??
Yes: we need harder requirements to evaluate against, much agreed.
♥ 2
Replying to @Edouard_b and @johnnyryan, @Chronotope, @DinaSrinivasan
I don’t know for sure but it looks like it. I guess they’ll let you in either way ;)
♥ 1
Voter suppression is bad, really bad, but whoever is causing my wifi to fail right now…
♥ 4
My election night cocktail is a "Judge Hidalgo". It's basically taking a sour and making it bubbly, sort of what Judge Hidalgo did in Harris County.
♥ 9
The Needle is coming. Well, the Needles.
nytimes.com/interactive/20…
♥ 5
Replying to @null and @benharnett
So much.
Replying to @USofDisaster
It's not tilting that way, right now.
My cat is telling me something about not getting too worked up before more votes are counted.
♥ 8
Replying to @jkohlmann
Also: hooman! See how good I look with black fur on moostard yellows! And weep.
♥ 1
Replying to @gabrielazanfir and @fmanjoo
I'm mostly surprised that there hasn't been a premature declaration of victory?
♥ 2
Replying to @hackylawyER and @Twitter
Different. Things looked bad faster. I recall having called it already.
♥ 1
Replying to @swodinsky and @jowens510
That, or most people do want to maintain some kind of underclass.
♥ 4
Replying to @swodinsky and @jowens510
We used to talk about the end of work, but all we really wanted was the end of having to deal with workers.
♥ 8↺ 2
Replying to @null and @eyesondesign00, @swodinsky, @jowens510
It's not specific to California, ballot initiatives happily blend everything that doesn't work in direct democracy with everything that fails with voting.
♥ 1
Tomorrow, and tomorrow, and tomorrow,
Creeps in this petty pace from day to day,
To the last syllable of recorded time;
And all our yesterdays have lighted fools
The way to dusty death.
♥ 13
Replying to @tim_libert
I'm still hoping!
Replying to @swodinsky and @eyesondesign00, @joejerome
🤗 The good news is that GPC works with our without Prop24.
♥ 1
Classic, expected, but felt good all the same. t.co/jogMNtY697
♥ 22↺ 3
Replying to @esd and @kthalps, @DemSocialists
They voted *Trump*, not just voted.
Replying to @esd and @kthalps, @DemSocialists
There's no rehab in sight if Republicans are more supportive of Trump than they were before.
♥ 1
Replying to @slayser8
That's a quality choice and some truth in feeling.
Oh, sure, let’s have two run-offs in Georgia that way we can extend this fun moment all the way to January!
♥ 5↺ 1
It will be interesting to see if the courts can stop regulatory capture. Those who work their best to abide by privacy legislation are facing competition from companies that don’t, as regulators mostly watch it unfold, unconcerned.
This creates unfair markets.
Quoting a tweet by @PrivacyMatters ↗
♥ 7↺ 5
No one wants to speak up for fear of regulatory retaliation, but several of the major data authorities are, through sheer inaction, turning the GDPR into a bit of a joke.
♥ 2
This inaction is particularly salient when it comes to big tech monopolies. Does anyone seriously believe there’s nothing blantantly unlawful there?
The adtech world is in the middle of sweeping reform. The time to shape the coming decades of digital advertising is *now*.
♥ 1↺ 1
What’s the biggest impediment to adtech heading in a consumer-friendly direction at this juncture? EU regulators let big tech do whatever it wants, so adtech companies have to compete in a race to the bottom with personal data.
And they face no cost for ignoring the GDPR, so…
♥ 3↺ 2
This is fixable. There are excellent people inside data authorities, people not looking for revolving doors — but they need to be empowered to act.
I raise my coffee to @OpenRightsGroup trying to unlock this situation!
♥ 5↺ 1
We often speak of working in service of our journalistic mission, but rarely is that as true as when @caityweaver writes. nytimes.com/2020/11/02/sty…
♥ 1
I don’t like to brag, but I just received an exclusive invitation to a webinar with a "customer experience futurist”.
I know, I know, what can I say, I’m just one lucky guy.
♥ 20
Replying to @FitzTechLawIE
Who knows what they can do in the future?!
No matter what happens when the pandemic ends, I hope that remote participation in seminars will remain a reality. I can be working and simultaneously listen to a workshop on Science Without Numbers featuring Field, Burgess, Yablo, Maudlin… This used to be inaccessible.
♥ 12↺ 2
Replying to @lukemulks
I may have! But I mostly ignore LinkedIn, I can only take so much thought leadership.
♥ 1
Replying to @null and @amyvdh
I don’t think I can pull it off as well as you would — let me forward the invite.
Replying to @null and @amyvdh
I’d do that just to see you going full River :)
Breeeeeeeeathe.
♥ 5
Brrrreeeeeeeeeeeeaaaaathe.
♥ 5
HHhhhhHHnphhpllllgggnnnnnblpjblpjttttpppff.
♥ 2
♥ 2
Replying to @alextcone
Hahaha, I also don’t think this is for-profit conference territory. It was free, open to all, there were many serious heavy hitters in attendance anyone of which is a star on their own, and there were… 50 people? 70 maybe?
Only so many care debate whether numbers really exist.
♥ 1
Replying to @null and @SoSillyDaddy1
Joining us live from Georgia.
youtube.com/watch?v=9jK-Nc…
♥ 3
Or maybe we could understand that the people who've been telling us about endemic racism weren't making it up?
Quoting a tweet by @AndrewYang ↗
♥ 8
Replying to @publictorsten
Hahaha :) Did you also see Johnny before?
I had a wonderful time speaking with these excellent minds, many thanks @E_Perotti for bringing us together. There is so much to think about in looking towards the future of media, I would gladly have stayed in that Zoom discussion all day long.
♥ 3
I forgot to say this, but if you fine people have some recommended reading on bureaucracy, especially somewhat theoretical, I'm very interested.
Replying to an earlier tweet of mine
"The Utopia of Rules: On Technology, Stupidity, and the Secret Joys of Bureaucracy", by David Graeber. This book makes thinking about bureaucracy fascinating. It does not always deliver as much as I would have liked it to on its promises, but it suggests great avenues to explore.♥ 5↺ 1
♥ 12↺ 2
Replying to @mblafkin
I have, and it’s on my list — thanks. It seems very focused on government bureaucracy (whereas I’m curious about bureaucracy in general), but I assume it can be seen more broadly. Have you read it, and if so do you recommend?
♥ 3
Replying to @daniellecitron and @ariezrawaldman
Oooh, now I'm excited! Does it have a name yet? Is it on preorder? Thanks for the recommendation!
♥ 2
Replying to @jason_kint and @slayser8, @ashk4n, @JustinBrookman
This is just a gut feeling but the rapid alliances feel more like desperation than strategy? The SSO leg of the endeavour also feels very optimistic, to put it nicely.
I agree there remains a need for audience matching but is this the best way?
♥ 3
Amongst other things I would worry about the weakness of email hashing. It's more reversible than people expect. And if email is the key, that'll be the next thing to go.
♥ 3
Wait — so it wasn’t about the emails?
♥ 7
Replying to @future_email and @MalloryKnodel
Found it! Checking it out now, thanks a lot! economist.com/open-future/20…
Replying to @__apf__ and @v0max
Not to necessarily side with Doordash here, but that seems like an accurate election night delivery.
♥ 3
Replying to @MarietjeSchaake
* We Will Rock You
* We Are Your Friends (Justice & Simian)
* will.i.am's Obama thing from 12 years ago
* Leonard Cohen's Anthem & First We Take Manhattan
♥ 2
Fort Greene ♥
♥ 10↺ 3
Replying to @null and @GamingTrent, @MalcWilloughby, @pattonoswalt
I have trouble connecting to the sadness.
♥ 1
Replying to @null and @GamingTrent, @MalcWilloughby, @pattonoswalt
I know, I know, but still, all I'm feeling is "hahaha".
♥ 1
Just because it's mobile doesn't mean you can't reach for the big type.
♥ 22
Happy Sunday everyone!
♥ 23
Replying to @fabricedesre and @carissaveliz
Something is weird with this one, I couldn't get it from @Bookshop_Org and had to use Amazon.
♥ 1
Replying to @gpxl_ and @carissaveliz
Oh wow I hadn't heard of that one, I love the premise! I added to my next order batch, not sure when I'll get to it but thanks!
Replying to @fabricedesre and @carissaveliz, @Bookshop_Org
Maybe Carissa can clarify if she knows of availability issues or such, when her time zone comes up!
Replying to @CarissaVeliz and @fabricedesre, @Bookshop_Org
Fingers crossed that t things will be such that you can do a book tour!
♥ 1
Article très intéressant sur la base d’études de terrain: les musulmans de France sont (sans surprise) généralement très bien “intégrés”; les chantres de la laïcité identitaire créent (aussi sans surprise) plus de problèmes que de solutions.
Quoting a tweet by @THofnung ↗
♥ 5↺ 1
There's a whole decade during which Google was the Democrats' Halliburton. Clearly, it would be a bad idea all around to let that happen again.
Quoting a tweet by @mer__edith ↗
♥ 4↺ 1
Are other conferences & societies considering doing the same?
Quoting a tweet by @BrittanyKamai ↗
♥ 15↺ 6
Replying to @johnallsopp
Right, I've seen that, but this takes it further.
Replying to @AmeliasBrain
Exactly what I was thinking. I've seen it done but not at the city level and not as a clear policy. This almost invites standards that conferences could adhere to.
♥ 1
The primary reason I gave up on Google Search is because it turned from a search engine to a portal with a random jumble of boxes and ads looking like results. I’ll still stick with DDG but this cool extension brings the usable Google back, it’s neat!
Quoting a tweet by @JuliaAngwin ↗
♥ 17↺ 1
Replying to @tim_libert
It totally did! And then some!
Replying to @antti_antinoja
This is @themarkup, you can trust them :) I haven’t audited the code but I would be surprised if this were anything other than a local transformation.
♥ 2
Replying to @julien51
You’d have to factor in the ARPU of how they leverage that to dominate all the other markets. It might be more money than you think :)
Replying to @julien51
World domination!
Replying to @JYSexton
In addition to all of the above, this also makes voter ID checks a major topic for their base — and that’s key to suppression efforts.
♥ 17
Replying to @mquijoux and @temptoetiam
C’est aussi un bon article! Je le référence régulièrement :)
♥ 1
Google announced that they will be changing Search (in May) such that AMP will no longer be a requirement for results to appear in Top Stories.
Depending on how that’s done, it could be a welcome step forward, or it could make no difference whatsoever. 🧵
webmasters.googleblog.com/2020/11/timing…
♥ 21↺ 2
A path out of AMP is important because it’s nothing but problems:
• Severely decreased revenue for publishers (in both ads and subscriptions).
• Decreased user experience (which we can see from the fact that it converts to subs far less).
• A lot of extra work to support.
♥ 3
• Publishers are forced to do AMP.
• The tech is brittle and hacky, and limits innovation.
Could "Core Web Vitals” be a way off? Maybe, but it depends on details that Google is remaining unclear about.
One option is that CWV is a *threshold* for appearance in Top Stories (ie. what is fast enough gets in), and beyond that ranking is just based on the other signals.
That would work and we could all thank Google for fixing the problem it created five years ago.
♥ 2
The other is that CWV remains a *ranking* factor even in TS. Ranking is very competitive. There’s no beating the 0ms loading time of preloaded content hosted with Google (through AMP Bundles).
Under that case this change makes no difference since we all have to keep doing AMP.
I hope it’s the former (I would like to spend less of my time on Google-created problems!) but fear it’s the latter. (If it is the latter, the anticompetitive issues around AMP don’t change, so that might be a factor.)
Either way it would be great to clarify in the post!
♥ 5
Haha `rm -Rf */node_modules` in my projects dir just saved me almost 30Gb - and I don't even work as a developer anymore 😂
♥ 14
Replying to @null and @SoSillyDaddy1
The better to code you with!
Replying to @can
JS on a chip!
Replying to @can
Oh I'm not really joking, I've been in serious discussions about XML on a chip!
♥ 1
Replying to @can
Just like the finest wine, my friend!
Replying to @jaffathecake
By conversion I mean to subscriptions. We don't do push grabbing at all. We only show a paywall when someone has seen enough content that we reckon they should pay. In theory, AMP does that too.
But maybe you're right that it's not the experience, that it's due to the strip mall presentation. Why subscribe if there is no differentiation between sources?
Either way it's a loss of revenue for people who need it.
Replying to @azeem
I can’t wait for the “it’s electric therefore we must make it look like some ugly futuristic pulp cover car” phase of electric cars to end.
♥ 1
With sync in Chrome, your data (history, etc.) is by default stored in cleartext unless you set a passphrase, even though you have to be logged in for this to work in the 1st place. Other browsers just do E2E.
Could someone in Chrome or Google security explain the rationale?
♥ 24↺ 8
Replying to @fchollet
The answer is “yes.” It would make sense that one would feed the other.
A good post on this: conversationalist.org/2019/11/08/soc…
♥ 2
Replying to @_alastair
I can’t believe we got scooped on this one.
Is there a way to get more than 3 header levels in @RoamResearch?
Replying to @fatemehx2
I know. That’s why I have to ask the question. On the face of it it’s such a bad decision, I’d like to give someone a chance to answer. There’s a point at which I wonder if a diligent CISO shouldn’t be required to mitigate this. I also wonder if it violates the consent decree.
♥ 1
Replying to @stimpee
This is purely Chrome, I doubt any other browser does the same thing. It’s… really hard to justify?
Replying to @jeroenev and @intuxikated
The next question then becomes: what legal basis are they doing that under? :)
Replying to @hexalys
Who knows? It’s not like reading history isn’t sensitive enough already.
Replying to @timothyswt and @fatemehx2, @GoogleChromeDev
This actually had to be built specifically to be this way, so I don’t actually buy either reason — especially as other browsers just do it right with smaller teams and the same debugging needs.
Replying to @jamie_gaskins and @intuxikated
Yes, GDPR enforcement is a running joke. I don’t think the FTC is doing much either.
♥ 1
Replying to @erikphoel
Except LaTeX isn’t universal, it’s about 15% of papers — most of the rest is Word. Both Word and LaTeX suffer from being rendering formats rather than semantic, which is what you’d want for this to work.
Not to say this is impossible — I know it can work because I built it once.
♥ 1
But getting authors to switch to better formats is nigh impossible (especially in eg. physics where people think that LaTeX is actually a good format and make fun of Word, even though they’re both pretty terrible). So that means processing a standardised form of some sort.
♥ 1
If people are willing to give up their local LaTeX tweaks, a standard LaTeX-alike should be possible. The same might work with Word. Both would work better if you’re willing to spend some time tweaking the submission in the tool (because neither is good at semantics).
♥ 1
From there it should be possible to output JATS (maybe JATS4R) and be more or less in business — for rather generous values of “more or less” given how pigheaded publishers and journal editors can be about their personal little preferences.
♥ 1
Replying to @HalSinger
There's an election in 2022 with notable odds of autocracy, though. New Zealand?
♥ 1
This is not particularly hard to address technically. The reason it’s happening is because Google wants the money from advertising on viral disinformation videos, no matter what it does to democracy.
Quoting a tweet by @mhbergen ↗
♥ 31↺ 16
Replying to @tim_libert
I don’t know, I take the gross incompetence excuse for the first year or two after the problem is raised, but it’s been much longer than that now! They seem more interested in stoking the dumpster fire than anything else.
Replying to @tim_libert
That is certainly true — even when sober.
Replying to @tim_libert
You’re right — I hadn’t thought about it that way, but it makes sense that they’re just a successful Limewire and have the ethics to match.
♥ 1
Replying to @montezumachavez
Hardware makers trying to catch up to the surveillance practices of Big Tech are the worst.
♥ 1↺ 1
Privacy Is Apple. Unless, you know, you want some actual privacy, like, from Apple. You can’t get that why would you want that?
sneak.berlin/20201112/your-…
♥ 58↺ 23
Replying to @null and @benlikestocode, @msdixon
You might say that JOBOL is just a different way of running COBOL.
♥ 1
Replying to @null and @HelloFillip, @Apple, @tim_cook
That’s what I thought too. This isn’t just a bad idea, it’s a bad idea implemented spectacularly poorly!
Replying to @frank_massanova
I’m not sure what’s sensationalised here? There’s a huge difference between collecting local log data and transmitting it remotely: the former is purposeful, the latter is a violation of privacy.
Doing it unencrypted is additionally woefully incompetent.
♥ 4
Now, of course, if your benchmark for sensational compares to, say, Google, then evidently what Apple is doing here is nowhere near as bad as permanent monitoring of geolocation and collecting a user’s full URL history. But Google is a very low bar in data protection.
♥ 3
At the end of the day, the one question that matters is: who is the OS working for? The OS is expected to be the user’s agent, it must by definition be trustworthy and loyal. Breaking that promise is dangerous ground.
♥ 3
Replying to @tulchin
This isn’t local. This is stored remotely. It means that Google has access to your history.
Replying to @tulchin
It’s documented as such. Why would it matter that it’s encrypted on their disks?
Replying to @null and @pes10k, @thezedwards
I don’t think this is “assuming the worst” though. The data is sent in cleartext — that alone is ridiculous. If the purpose is usage, I would expect some DP to be involved, eg. something like Prio.
♥ 1
Replying to @chrismarrin
Transmitting analytics unencrypted, apparently without DP, and preventing user space firewalling aren’t “baseless conspiracy”. It’s either deliberately bad or shockingly incompetent. Either way it’s bad privacy.
Replying to @chrismarrin
This isn’t about me. It doesn’t matter what settings technically competent people may pick. What matters is how data is handled on behalf of people who can’t make that call themselves.
At no point did I say “nefarious”. It may be nefarious, but it may also be negligent.
♥ 1
Replying to @null and @podopie
Could you repurpose Zotero for that?
Replying to @thezedwards and @pes10k
These are good points, is the author on Twitter? Can't find his name. Would be good to compare evidence?
♥ 1
Replying to @thezedwards and @pes10k
Ah, crap! Well, let me at least push your super helpful look. I wish I got that kind of answer with Google issues 😁
♥ 1
It's worth looking at the fact that @thezedwards has a different read from the original post. And he may be right indeed. twitter.com/thezedwards/st…
♥ 2↺ 1
Replying to @slightlylate
It's worth looking downthread though, the unencrypted claims are called into question.
♥ 1
Facebook: we're the best at making money discrediting democracy!
Google: hold my beer.
Quoting a tweet by @ubiquity75 ↗
♥ 9
I'm looking for a gif that says "threw up in my mouth" but I'm coming up short.
Quoting a tweet by @ubiquity75 ↗
♥ 15↺ 4
Replying to @null and @dansereduick
It's not their fault if Nazis are winning at the marketplace of ideas, they're just a very engaging bunch!
Replying to @null and @sparkstack, @thezedwards, @pes10k
Well, it shows quite clearly that informed consent is rarely a thing with computers.
Also: if this is for analytics, given the risks it really is an asinine implementation.
♥ 1
Replying to @tomayac and @pes10k, @thezedwards
Why would they need to push this kind of event for antimalware?
♥ 1
Replying to @mariafarrell and @DorotheaBaur
The privacy thing is still being dissected (it's possible that it may be something legitimate executed poorly) but at any rate unless there's a feature that's critical to you I would strongly recommend letting others find the bugs first!
♥ 2
Replying to @tim_libert
That's pretty cool, it sounds like a real challenge, I'd love to have that as my job.
♥ 1
Farewell old friend!
♥ 19
Replying to @fraserdeans
Gotta use that French card 😉
Replying to @johnwilander
Nothing fancy, just a sturdy 16in MBP! My interest in hardware just extends to making things work. I get a little sentimental when a laptop has seen me through this much change and has endured quite some weary abuse in the process is all 😁
♥ 2
Replying to @cdanuloff
Noooo, haha.
Fun fact: they tried to! But everything ended up labelled 943.086 so it wasn't worth it.
Quoting a tweet by @sys4dmin_ ↗
♥ 5
Replying to @null and @nhoizey
I should be able to get to it around July 😜
Replying to @null and @benlikestocode
Replying to @null and @nhoizey
Thank you, fingers crossed indeed!
Replying to @null and @podopie
The hair is a tell, though.
Replying to @Myles_Younger
Hahaha, me too, I'm mostly reading to prevent the unread pile from collapsing and killing me!
Replying to @stimpee and @johnwilander
We'll see, I had that kind of problem with my MBP13, it would panic once or twice a week. So far the 16 is handling better.
Replying to @stimpee and @johnwilander
I only use Chrome when Google makes a thing that’s Chrome-only, so I can’t say I’ve been affected. I don’t like the tabs and the dev tools, and it’s a privacy shit show so ¯\_(ツ)_/¯.
♥ 1
I agree with the problem statement, but less with the suggestion that this is solved just with honest journalism. It's imperfect, no doubt, but there's already more "honest journalism" than most could read each day. How does more help?
Quoting a tweet by @craignewmark ↗
♥ 7
Replying to @heidilegg
I don't know for sure, just some ideas! And I'd love to hear other ideas too. There's the usual bunch. One is more funding so that local journalism can exist again. It's harder to distrust your neighbourly reporter.
♥ 1
Policing the digital ad market so that Google stops cheating. The header bidder natural experiment indicates that that could bring 30-100% more money to newspapers. Then some better funding models, eg. Scroll.
♥ 1
Less usual, I think we need to think about how news is presented. With aggregation, everything looks uniform: you get a big blob of results that all look like they're talking about the same thing (because search). The "MSM" looks like a real entity.
I think this would warrant some user studies to see what the effect of aggregation on trust and the perception of news is
There's also work on depolarisation that I'd like to look into. I think some could be used by news products.
♥ 1
Finally, we need to sort out Facebook and YouTube. With them, or against them. They rooted out radical Islamic terrorism (and porn), they could root out white supremacy, antivax, etc. that kill far more.
♥ 1
Replying to @heidilegg
I definitely think we can do much better at presenting opinion as part of the paper (I wanted to pitch something internally but short on time). The same in aggregators would be neat.
♥ 1
Replying to @heidilegg
We do, but with what means? I hope to experiment some this coming year, but only so many outlets can afford to invest in that.
♥ 1
Replying to @ACBrewst
Sure, but the problem there isn't production, it's remuneration. Calling for more won't make it happen.
I've been wondering why this isn't more publicised. Covid-19 can affect your testicles and may cause erectile dysfunction.
You'd think we'd use this to make the mask toxic masculinity crybabies wear a mask?
Quoting a tweet by @DrEricDing ↗
♥ 26↺ 10
Replying to @null and @arainert, @can
New York, always drama drama drama.
At this pace the entire US is positive in, what, two weeks? The pace probably can't hold, but it would be good not to find out.
Quoting a tweet by @goldencell ↗
♥ 6↺ 1
For what it's worth, "two weeks" is an actual quick and dirty approximation of the actual rate - it's not meant as an exaggeration.
♥ 1
“MAdTech: Back to the Future” — this looks like it could be a fun event, I’m having fun trying to cram what I want to say into the 5 minutes of the lightning talk I have 😀
zoom.us/webinar/regist…
“Data also consolidates power. If you worry about Google, Facebook and Amazon having too much influence, you should be aware that at the root of their power is control over reams of information on where we go, what we do and what we like.” — @ShiraOvide
nytimes.com/2020/07/15/tec…
♥ 83↺ 37
Replying to @tim_libert and @ShiraOvide
Some — but then you get into multipurpose data (which it almost always is) and if you’re large enough you can anonymise while still using that data anticompetitively. It could help, though!
My quick fix: user agents are where the data come in. Make them fiduciaries.
♥ 2
Replying to @tim_libert and @ShiraOvide
Oh, I’m not worried about that — for those purposes you don’t ever need to keep the data much.
Replying to @coolharsh55 and @ShiraOvide
Maybe that was from _Data Feminism_? I know I’ve seen it too.
It’s yet another thing that benefits from thinking in terms of legibility, as in Seeing Like A State!
♥ 4
Much-deserved recognition. If I had a say, this one would also carry a Best Paper Title award!
Quoting a tweet by @histoftech ↗
♥ 7
Ok, this is completely addictive (via @masooga). randomstreetview.com
♥ 5↺ 2
If more people read it & link to it, it must be more relevant, right? Wrong.
Algorithmic curation should enhance autonomy — but the norm is to implement it in ways that instead substitute for it. Poorly.
Quoting a tweet by @AnnenbergPenn ↗
♥ 24↺ 5
When looking at a product the question to ask is: “Could I become *good* at this?”
If the answer is “yes” then you’re looking at a tool. If not, you’re facing the modern equivalent of a TV channel: your brain is supposed to be off.
♥ 12↺ 3
Companies don’t do this from malice, they do it because it’s easy. Run some A/B tests, tweak some hyperparameters: watch engagement go up and down!
Tools on the other hand require thinking and (*shudder*) listening to actual people rather than just measuring them. Ewww!
♥ 3↺ 1
Replying to @kingsburywrites and @AKingsburyNYT, @masooga
I KNOW RIGHT? For a short while there, this almost felt like the Web used to be.
Also, I have questions. Like, why do Streetview pictures have such crap saturation?
Replying to @null and @sivavaid, @ShiraOvide
Gee, I wish I knew someone who’d written a book on this.
♥ 1
Replying to @coolharsh55 and @AutoArtMachine, @ShiraOvide
Exactly, and the data economy does this at unprecedented scale. Data collection is just how it starts.
♥ 2
Replying to @justinph
Awww, thank you sir! 🤗
Replying to @AmeliasBrain
We try to optimise for the power user too 😉 We could do more in that direction though, I wish I had the time to look i into that too.
♥ 1
Replying to @ColinHayhurst and @AmeliasBrain
Thanks! A few sometimes are 😁
Senator @MalcolmByrne makes a good point about @DPCIreland: if the OECD made a data protection equivalent of its list of countries friendly towards money laundering, Ireland would hold a choice position on it.
Quoting a tweet by @maxschrems ↗
♥ 3↺ 3
Replying to @cwarzel
You’re in great shape, Charlie.
For fuck's sake @Twitter, if I wanted a shit interface of full screen content that's painful to navigate and won't stop moving I already know where to find Instagram.
We. Just. Want. An. Edit. Button.
♥ 24↺ 1
Seriously, though. The most baffling thing is that the stories UI is just plain crap. If you can't innovate, at least copy the good stuff.
Quoting a tweet by @hermit_hwarang ↗
♥ 1
Replying to @lukemulks and @Twitter
First thing I looked for, couldn't find. I guess it's like all the trends bullshit you ignore on your way to search.
Replying to @null and @dansereduick
Got to give him that he's creative at it!
This is a good article on AMP, @themarkup, but a more factual headline would be “As Antitrust Pressure Mounts, Google to Stop Forcing News Sites To Use Its Proprietary Technology”.
There was never any “benefit” from AMP, except to Google!
themarkup.org/google-the-gia…
♥ 59↺ 20
Replying to @nvbalaji and @themarkup
One of the reasons you get weird popups in the first place is because of the decreased revenue — in no small part due to AMP.
Also, the downgraded experience means you’re not getting the more innovative content.
♥ 1
Replying to @nvbalaji and @themarkup
As I said, AMP is a contributive factor, not causative on its own. There are other causes — Google systematically cheating in the ad market for instance — but AMP is making it worse.
♥ 1
Replying to @nvbalaji and @themarkup
There were ways of designing faster web experiences without this kind of brittle hack hosted at Google, though.
Replying to @neilturkewitz and @hypervisible
Nor does it say how long it stays up before they remove it.
♥ 2
Not correcting the spelling is a really nice touch, enjoying the editorial decision here.
Quoting a tweet by @nycsouthpaw ↗
♥ 10
Replying to @nvbalaji and @themarkup
What you describe is cheating already, but see papers.ssrn.com/sol3/papers.cf…
Replying to @nvbalaji and @themarkup
The caching is the problem, though, pour at least the core of it. Perf based search rankings have been possible for years (and what many of us advocated for). AMP's only advantage over that is Google controls and can reuse content.
Replying to @nvbalaji and @themarkup
This is a completely different market though. For comparable markets, people go to jail for these practices!
Replying to @nvbalaji and @themarkup
Not really. AMP does not force a clean design, it limits things but not that. And the caching is from specific constraints that are not necessary for sufficient performance.
Replying to @AmeliasBrain and @briankardell
I thought the same. It's almost like he's monkeying trumpisms on purpose.
♥ 1
Replying to @lukemulks
Hahaha new revenue stream, that's a good one.
Replying to @donohoe and @lukemulks
That further does not account for decrease in conversion rates, lower attractivity due to limited expressive power, loss of analytics AND loss of privacy at once, limits to business model innovation, and the maintenance costs of getting involved in brittle, hacky tech.
♥ 3
Replying to @donohoe and @lukemulks
I know. Web performance has taken so many setbacks for this shit show.
♥ 5
Replying to @slayser8 and @Chronotope
Oh! I had no idea whotf he was either but now I want to know him better!
♥ 2
Replying to @can
Can’t even get your Protestant work ethic right!
Replying to @mariafarrell and @BeireannBuanBua
Ok, you almost made my coffee come out of my nose :)
♥ 1
Replying to @Chronotope and @can
I’m ready for a country that doesn’t treat sociopathy as a legitimate political platform.
I mean, look at these folks! en.wikipedia.org/wiki/Katr%C3%A… or en.wikipedia.org/wiki/Gu%C3%B0n…
♥ 1
Replying to @Chronotope and @can
Is that a hot dog stand that’s over there? They have more than one? I’ve only been to the airport (though I’ve been there quite a lot…) so I sort of envy your hot dog story. (Also, no one can ever figure out where I’m from so I should be safe from that.)
♥ 1
Replying to @Chronotope and @can
Oh just that — and there I was picturing you being like @awkwafina on the Reykjavik subway!
youtube.com/watch?v=0-taYS…
♥ 1
What saddens me most about the adtech world is that what passes as innovation there is basically just monkeying Google — already not an innovative company — but without the scale.
Is this a good strategy? 🧵
Quoting a tweet by @lukOlejnik ↗
♥ 16↺ 1
Google is hands down the biggest privacy violator, easily worse than Facebook and most of adtech put together. And the alternative that the adtech industry is proposing is… let’s try to be just as bad and user-hostile!
♥ 1
This would all be quite funny (almost as funny as proposing a “Transparency & Control Portal” as a privacy measure) except that we actually need advertising to work. Societally. This is a race to the bottom that we’re all losing.
♥ 1
The logical next step would be for adtech companies to align with users instead of against them. That way regulators (and buyers!) could intervene against the platforms for their egregious data practices without flattening the whole of adtech at the same time.
♥ 1
But that creates a collective-decision problem: the current market has certain expectations and all would benefit from change (adtech, publishers, and users), but most of adtech would have to change at once.
That’s a problem for a standards organisation!
♥ 3
Unfortunately, standards in adtech are still set by the @IAB. So we seem to be stuck with losing strategies like “Unified ID 2.0”.
I hope more non-adtech web techies will get involved in trying to fix this mess — because that’s the only way out of this nonsense!
TGIF!
♥ 3↺ 1
Replying to @EdKick8011 and @DukeSilver80
Oh, they beat you to this one! ana.net/content/show/i…
Watched a bit of The Pack and I don't quite get it. Why do all the stuff with the competition and the humans instead of just spending the time with some good doggies?
♥ 1
Wait. There's a guy who found his dog because he walked past a guy who was telling his wife "I'll deal with you when I'm done killing this dog." And so he stepped in to save the dog. What about, err, the woman? WTF @amazon?
♥ 1
Replying to @EdKick8011 and @DukeSilver80
If you squint right it spells "ethics".
Replying to @slayser8
Completely agree. But by trying to be as bad as Big Tech they're giving it a lot of cover. An alignment of adtech and consumer rights would be a force to be reckoned with; instead Big Tech can just play them off one another.
♥ 1
Replying to @bayesianboy
I gave up and got some noise protection ear muff things. It's not perfect but at least I can read.
♥ 1
Replying to @slayser8
I know, which is why the market won't self-correct. It's also why agencies don't care too much about fraud so long as it doesn't tank the numbers. This can only change if we can get good regulation to kick in (or tech as a forcing function).
♥ 1
The interest in small adtech to get regulation to kick in is that they can't compete on the invasion of privacy market. Google gets location pings by the minute on a couple billion phones - and that's just one entry point. This can't be beat, ever.
But so long as they paint themselves with the same brush as Big Tech, it's hard to block one without blocking the other.
Replying to @slayser8
I increasingly doubt if that's so clear cut, though. You do need to measure if your channel works, but I am less and less convinced that doing it at the individual level really helps that much. People are noisy!
♥ 1
We cut down very heavily on how much data we provided to martech companies precisely for that kind of op. They're off all but our subs landing pages. We've grown something like 2-3 million subs since? And the spend has remained effective.
Replying to @slayser8
I think we're largely in agreement, but the problem is that pubs are telling regulators that forbidding the bad stuff that Big Tech does would also destroy smaller adtech, because smaller adtech shows no signs of reform. Something has to give first!
♥ 1
Replying to @slayser8
Sure, I meant this purely from the marketer pov. I know not everyone can kick programmatic like we do, but as marketers we're in the same boat as everyone else! And we're driving a strong marketing op with a lot less data sharing than most.
That's what makes me doubt that marketers need this as much as the conventional wisdom has it. People get a rush of power with microtargeting, but broad+MMM can be pretty sweet! Also stuff like TAFI.
You know what would make email substantially more energy efficient without cutting back on the number we send? Default to plain text and - most importantly - don't quote the source in replies. Those of us who recall how to do it right live in hiding anyway.
Quoting a tweet by @hoofnagle ↗
♥ 17↺ 1
You know who could make a real difference here? The Gmail folks. Energy efficiency could be a nice way to make up for that UI!
♥ 5
Replying to @EdKick8011 and @DukeSilver80, @slayser8
I think that people who buy from 3P data should get to see not just the segments they're in right now but how those segments have shifted randomly over the year prior. It never makes any fucking sense!
Replying to @gnat
This is a rule in Gmail design overall. It doesn't tell you who's copied on emails you get, just the (inferred) given names, you can set the font size you write at people but not the one you read with, etc. It's to send email, not communicate.
♥ 3
Replying to @EdKick8011 and @DukeSilver80, @slayser8
Or at least it's algorithmic guessing. 😛
Replying to @gnat
Frankly, I never thought I would say this but after a year being forced to use Gmail, I would *beg* to use Outlook.
♥ 2
Replying to @mnot
Just not citing in the response would work better, no?
Replying to @crenshaw_dev
Haha!
Replying to @mnot
Not necessarily, change the default in the mail client and watch it stop!
Replying to @null and @HelloFillip, @lukOlejnik
And Google's proposals ring shallow so long as they're not paired with actual changes in data practices elsewhere. It feels a lot like "do what I say, not what I do."
Replying to @tobie
♥ 2
Replying to @null and @HNissenbaum, @hoofnagle
Evidently, we should do both, and indeed advertising would have a greater impact. The advantage of looking at email is that, even if the i impact is small, a single provider changing one or two defaults would have a measurable impact.
Of lesser relevance to the future of the planet, it also vindicates aging geeks like me who learnt email the ancient netiquette way 😉
Replying to @swodinsky
I know, I know...
Is "burn it all down to the ground" not actually a sensible policy option? Why don't we discuss it more?
♥ 2
Replying to @swodinsky
I know, but it's a gift that keeps on giving! What I don't get is why Google isn't getting more mud from this. They're an easy target.
♥ 1
Replying to @swodinsky and @Chronotope
Not really. That's always been shit, they much prefer the tracking version, they get so much more data. But I guess that with 1P cookies they set plus link decoration or even just timing they have enough of a workaround.
♥ 3
Replying to @Chronotope and @swodinsky
But so what's the play? At this level it can't just be spite, no? Could it be that, for FB, Section 230 reform is far more dangerous than privacy legislation and so it's in their interest to make it clear it's a total shit show by dragging Apple in?
♥ 2
This is an excellent @doctorow thread summarising @DinaSrinivasan's already classic paper on how Google cheats in the ad market (with all the negative impact on democracy, etc.).
Just a couple of notes. 🧵
Quoting a tweet by @doctorow ↗
♥ 33↺ 13
The "privacy" improvements that Google claims in its ad stack don't exist. Yes, they've prevented others from reidentifying - but not themselves! This means no better privacy for users.
It's like saying your neighbour no longer spies on you, but the secret police still does. 🤷♂️
♥ 7
There is a path that avoids financialisation: slow down the bidding. The only reason to have real time bidding is if you're doing individual targeting. But with privacy, you can't. (And even today, it only works so well.)
♥ 8
That's one of the avenues which I think may be the most promising to get out of this bind. Dina's analysis is impeccable, but the risks of financialising such a key piece of infrastructure are high. (@timhwang has a good book on this.)
Let's see if we can figure that out! 😎
♥ 6↺ 1
Replying to @ShannonVallor and @OUPPhilosophy, @Lucy_Randall_
Is there a date set for it?
Replying to @ShannonVallor and @OUPPhilosophy, @Lucy_Randall_
I'm more into paper, I'll wait a bit - 2021 isn't *that* far away 😁
Replying to @LogicalAnalysis
I know that one! It's when your kid is acting like a teenager despite only being eleven.
♥ 1
Replying to @donohoe and @aiddya
It's not infrequent users who matter, though. You can rabbit hole yourself through the night and back on YT videos.
♥ 2
Like @swodinsky I worry about this. Like it or not, the great equaliser of access is advertising. Adblocking is a privilege, subscribing is a privilege.
But because the digital world was built by privileged people we got advertising wrong. Or ignored & let others get it wrong.🧵
Quoting a tweet by @swodinsky ↗
♥ 18↺ 5
For the first few decades of the Internet, the idea that it could have commercial use was shunned. I sort of get the draw, if I squint right, but let's face it: that's rich people thinking and it's privileged as fuck.
♥ 2
So we built this amazing worldwide information thing, but we spent zero time working on how to make it sustainable. It's supposed to run on... charity? Goodwill? Some guy in Nebraska?
♥ 3
Replying to @briankardell
True, and there will probably always be more than one, but I think that we can invent a good model that dominates as the default.
♥ 1
Replying to @steltenpower
You should try Scroll, that's pretty much what it does!
Replying to @steltenpower
Yeah, ads can work better, no doubt there.
Come for the history of tech, stay for the bad puns that actually explain things.
Quoting a tweet by @histoftech ↗
♥ 22↺ 4
Replying to @julien51
That's a great way to put it! I'd like to think that it's still possible to fix it. It's bad, but not over.
Replying to @julien51
Yes - I wonder if anyone has looked at how much that maps to on the electricity bill. I know it's been measured for bandwidth in places with metered access, but I think it is likely real money everywhere for some people.
Replying to @briankardell
It depends on which medium. These days it's subscription but to very centralised systems, often that don't pay artists enough (look at music for instance).
♥ 1
Replying to @julien51
If it were only the ads, but the energy they sink into useless problems... Products that do things for you badly instead of help you do things better and your way, trying to make the web more like native, driving an extra minute of engagement.
♥ 3
Huge braintrusts and what is there to show for it? Clippy that puts your email in tabs? Ooooh...
♥ 2
Replying to @DrJenFuller and @soundbitelife, @histoftech
I find looking for a pun much easier with a punopticon.
♥ 1
Replying to @donohoe and @swodinsky
Yes and no. Pre-internet, the cost of print was very heavily subsidised by advertising. Yes, people had to pay for a copy or sub, but at prices much lower than what it would otherwise command. (IIRC @superwuster has some notes on this.)
♥ 1
We could have the same model on the Internet, and it might actually not be bad (especially with a three-level paywall ramp: free, micropayment, sub). Except digital advertising is not working well and the market is on the verge of collapsing.
♥ 2
Replying to @null and @nitot, @LeonYin, @themarkup
I saw — I even got a pull quote from one of my favourite publications ;)
♥ 2
Replying to @psd
Is it good? I looked the blurb up and I can’t tell if it might be super interesting, extremely bland, or a tantalising blend of both.
The famous Bok Choy ballet in action. So moving.
♥ 11
Replying to @julien51
Nothing but the best jokes for their education.
Replying to @null and @HNissenbaum, @hoofnagle
That's not the only study, I am sure that I have seen others, but a comprehensive review of the literature and research to plug the gaps would be great. A lot of large advertisers have social responsibility programmes that could be influenced by this argument.
The green angle on adtech is a potentially powerful one. Any suggestions for good research beyond the obvious? t.co/SnOOkrEiko
♥ 4
Replying to @BrendanNyhan
I don't think anyone gets any credit this late. So much of the damage has been done.
♥ 1
Replying to @swodinsky
She's still stuck on The Master and Margarita. Though The Anatomy of Melancholy has been adamantly recommended for her general state of mind. (It in fact has!)
♥ 1
Replying to @swodinsky
"Fierce, true, hilarious, and creepy" - you clearly have excellent taste in cats.
♥ 1
Replying to @alextcone
Thanks! I am indeed interested, but I think this would be better if it happened as published research rather than just to sate my curiosity :) I wonder if maybe @HNissenbaum has a colleague who’d want to take this on?
♥ 1
Replying to @psd
I’m relatively deep into ethics these days, but I’m not even sure I know what planning is (in the sense covered there) so it drew my attention :)
Replying to @alextcone and @HNissenbaum
I don’t think that "setting another group of actors against” would be the goal; I would be more interested in accountability (as I think you are :). I’m interested in people competing on quality, or on users’ ability to block selectively with some criteria.
I think that pricing in energy costs would help. A lot of data is collected “just in case” without really being able to affect outcomes. That would get trimmed. Creatives that use React and jQuery too.
Replying to @alextcone and @HNissenbaum
There are indeed large opportunities for optimisation in creatives. At the more “footnote" end of the scale, I wonder how much energy goes into serving the AdChoices icon :)
Replying to @alextcone and @HNissenbaum
I believe your motives Alex! I think we should all agree that it’s dead, deprecate the whole thing, and move on.
It might! Encryption vs minimisation pull in different directions — it would be great to work from actual numbers. I think the big buyers would be interested.
Replying to @psd
You make it sound like this is the right thing to read after _Seeing Like A State_ :)
♥ 1
Replying to @bayesianboy and @dsinghaie, @davidthewid, @Aaroth, @shoshanazuboff
I have not but I think I really should, I've been trying to get a bit into mechanism design that aims for social good and that would fit right in.
♥ 1
Waaait is this for a test? Did I fail something?
♥ 1
Replying to @rasmus_kleis and @themarkup
Except where it'll still be required, and except if the CWV keeps using ranking instead of threshold 😁 But it's progress!
♥ 2
One more example of the marketing industry shooting itself in the foot (or, given that this comes from the cringeworthy “Marketers for an Open Web”, MOWing itself down).
If Google prevents others from tracking while tracking itself, there are two options: 🧵
Quoting a tweet by @ProfFionasm ↗
♥ 12↺ 2
1. Allow everyone to track. This is what MOW are proposing. This makes the name of this small group quite ironic since third-party tracking is inherently a violation of trust and trust is the most important value of the open web.
So that one’s off the table.
♥ 7↺ 1
2. Prevent Google from tracking. This is a strongly pro-consumer option that is also aligned with the values of an open web and that prevents problems of structural anticompetitiveness in the data market. (MOW’s proposal helps the biggest.)
♥ 10↺ 1
Just as with the similar complaint in France against Apple, I certainly hope that @CMAgovUK will see that this complaint goes against fundamental rights and does nothing to help competition — but GDPR enforcement could solve both.
♥ 5↺ 1
Replying to @mario_gug and @CMAgovUK, @UGambini, @wavesblog
Much agreed! The only aspect that worries me is that the “GDPR gold standard” as you aptly call is taking a long time to materialise — and that is definitely having anticompetitive effects.
♥ 3↺ 1
Replying to @Chronotope and @metaviv, @brooklynmarie, @Google
Oh lovely — that’s not a lot of detail to track the creative down but I’ll pass it along.
♥ 3
Replying to @null and @SimonDeDeo
That all seems true for any presentation length!
Everything on this list is good and there's much more that didn't make the cut! If you don't yet read @themarkup now is a great time to dive in. And if you do read it, don't forget that you can support them with a donation or subscription! ♥
Quoting a tweet by @themarkup ↗
♥ 12↺ 2
Replying to @alextcone
I’m suggesting that the GDPR should be enforced! Specifically, on two points.
1: a strong standard of consent for controller to controller data exchange. It should be rare, nearly impossible, to have a third-party data controller. G (and others) can collect, but for the 1P only.
2: by enforcing purpose limitation. G (and others) can still collect on their own property, but cannot use across products in their offering. Eg. Android’s constant location monitoring can’t be used for ads. Gmail data can’t be used to recommend flights in Search, etc.
This is actually nothing outrageous — it’s pretty much how it’s supposed to work!
♥ 1
Replying to @Moonalice and @MarietjeSchaake, @nytimes
This is programmatic, definitely not us — I know the team is looking into it. It could be Google, it could be another programmatic partners, it’s hard to tell from just a screenshot.
Replying to @alextcone
I think it changes quite a lot for privacy — use in context and in purpose can be reasoned about and is far more likely to be appropriate.
Regarding impact on ad spend: if they win in a fair fight instead of by cheating, it’s harder to complain about?
Replying to @alextcone
For the OS: it’s an agent, not the first party, but even if the first party then purpose limitation would prevent that data from being used elsewhere.
I am often told that digital advertising that’s respectful of users’ privacy is something “only The Times” could pull off. This thread shows that’s not true.
Quoting a tweet by @AlanToner4 ↗
♥ 7
The argument is often that only The Times has the scale to make this work. I don’t know what @politiken’s audience is, but there are fewer people who speak Danish than there are Times subscribers — so I think it’s fairly safe to assume they’re substantially smaller. Yet it works!
♥ 3
The NYT will continue to offer free unlimited access to US high school students all the way through to September 2021.
nytimes.com/initiative/hig…
♥ 5↺ 3
Replying to @rigow
Not sure @TimHerrera is familiar with the *specific* implications of that reference.
youtu.be/HkVhN64dyd8
♥ 1
Replying to @TimHerrera and @rigow
One of the greatest French songs of all time *and* almost how the Web was called due to the Wikipedia-like vision that stood at its early days.
♥ 1
This is a great thread about the misuse of quantitative methods in the workplace, especially for complex work.
The fun part is that we nevertheless determine how to organise the world's information using the same failed methods.
Quoting a tweet by @nickstenning ↗
♥ 8
Replying to @laurie_winkless
Thing is, by 2020 standards that's not even clickbait enough 😁
"Et on se prend la main
Et on se prend la main
Un garçon au féminin
Une fille au masculin"
♥ Indochine+Christine & the queens
youtu.be/PJ7H-INlclg
♥ 1
Replying to @null and @koalie
"C'est pas vrai qu'ils sont conquistadors, asexués une fois dévêtus."
Replying to @ElieSl and @koalie
Je te trouve dur! Contrairement à Cabrel (et au pire du pire, Michel Berger) au moins ils savent faire rimer autre chose que des é.
♥ 1
We can stop doing art, folks, the Form of Beauty has been located.
Quoting a tweet by @cadenzasmith ↗
♥ 3
Pre-order here! bookshop.org/books/your-com…
♥ 7↺ 1
Replying to @mnot and @Bookshop_Org
only does US and UK for now, but I'm sure the book will be available elsewhere!
♥ 1
Replying to @lilianedwards
A picture is personal data, which can be considered special category since it can reveal race and possibly other special cat info. What is the legal basis here?
♥ 1
Replying to @lilianedwards
That's why I said "can" 😁 But I would argue that the ML black box can't generate nudes without some form of processing of race-adjacent data.
Replying to @lilianedwards
In general I don’t disagree, but in this specific case it seems difficult to generate nudes without touching on skin colour, which, while it does not define self-identification with a race, is very very adjacent to it. I don’t think this extends it to all ML!
♥ 1
I’ve written many scrapers over the years, including in my current position. I’d like to think that doesn’t make me a criminal.
(Thanks for the package @nabihasyed & @themarkup!)
♥ 31↺ 2
Replying to @lilianedwards
I agree it’s unlawful anyway, I was just looking to make the fine heftier :) But yeah, that’s probably not the most effective option, I was just looking for solutions.
♥ 1
Replying to @lilianedwards
I’ll admit I don’t know those well for the UK. But I agree that at least at the conceptual level this matches the definition of non-consensual pornography.
♥ 1
Just sayin’ store.nytimes.com/collections/ti…
♥ 1
Replying to @heydonworks
This could be a series, the topic changing between things that tech folks are wrongheaded about and the description being the same.
Replying to @daniellecitron and @JuliaAngwin, @rcalo, @andy_sellars
You look *way* cool Danielle!
♥ 2