I honestly cannot assess to what degree this is genius or madness.
Quoting a tweet by @fizziksgabe ↗
♥ 4↺ 1
June 2018
201 tweets
Replying to @shimown
I've gone down that road, I don't think it will works, not in Word at any rate. You can get very decent basics (certainly on par with LaTeX's limited document structure) but anything a little complex (eg. like structured captions) gets too convoluted.
Replying to @ned and @shimown
It's been a very long time since I looked, I like the idea but the devil is in the details.
Replying to @blackfoundry and @bfluzin
I'm not sure what the issue is. What personal data do you need to assess pageviews?
Replying to @blackfoundry and @bfluzin
An IP address is indeed personal data, but it sounds to me like you have a perfectly legitimate interest in using it. Also, you probably don't need to keep it around very long. I don't see why the GDPR would block you.
♥ 3
Replying to @thephreck and @xphreckx, @snipeyhead, @OaklandElle
Ooooh, please yes! I need one!
Replying to @davelab6 and @blackfoundry, @bfluzin
I am not sure I would recommend reading that. Google are declaring themselves a controller for their Fonts product because they are not just serving fonts: they’re tracking people and profiling them.
If you’re just serving fonts you’re selling you’re in a different business.
♥ 1
Replying to @davelab6 and @blackfoundry, @bfluzin
Sorry, I’m not sure what you mean?
Replying to @davelab6 and @blackfoundry, @bfluzin
So how would you justify taking a “Controller” position if the service is doing nothing other than serving files at the express demand of someone else? That’s almost as far away from being a controller as it can get.
Replying to @davelab6 and @blackfoundry, @bfluzin
If logging for debugging and service improvement purposes makes one a controller then everyone's a controller. That's an absurdly extreme view.
Replying to @coda_hq and @dblohm7
It would be better than nothing, for sure. But a remaining concern is openness: if the team isn't coding to the open Web from the get go what are the odds in can get the data I put in there back out in a usable format later? Data is harder than code!
♥ 1
Font get me wrong, I've done startups, going after big changes in entrenched worlds, I know it's hard. But still! Does this mean you have a Chrome monoculture internally? Aren't you concerned this separates you from real users?
Marmite is science!
Quoting a tweet by @jesswade ↗
Marmite is food!
Quoting a tweet by @ChrisBloomerPhD ↗
Replying to @null and @dtemp_09f9
Thanks David, but now I need the recipe to your marinade!
Replying to @alexmilowski
Victory!
"What do we pay when we 'pay' attention? (...) We pay attention with the lives we might have lived."
I'm not even halfway through @WilliamsJames_'s "Stand Out Of Our Light" and I already feel like dusting off my molotovs to storm Silicon Valley with.
♥ 9↺ 2
Replying to @null and @amyvdh, @nytimes
Vegemite then?
I always wondered if men who are feminists because they have a daughter are the same who have that one black friend.
Quoting a tweet by @TrulyTafakari ↗
♥ 2↺ 1
Replying to @_abialas and @femalegazebot, @PicturesFoIder
I wonder if you can reply all on this one.
♥ 1
If you need a tablet just for the kids to watch/play a bit in travels and the such, is there any reason to get something more powerful than a Fire HD 8?
↺ 1
Replying to @runnerpod
Thanks, that's super useful feedback!
Replying to @xchaotic
I don't know man, they're four times the price and then you have to use iTunes.
Replying to @hexalys
Agreed, that's been a problem with any device as far as I'm concerned! I've spotted a refurb that has 32go for $72, I reckon that's hard to beat!
♥ 1
App platforms are way worse than the Web for privacy. Throw Facebook into the mix and you've got a party going 🕺
"Facebook Gave Device Makers Deep Access to Data on Users and Friends" nyti.ms/2Hh5sbn
♥ 1
"Stand Out Of Our Light", by @WilliamsJames_ (winner of @NineDotsPrize). A thoughtful, well-written, necessary call for an Attentional Rebellion. If you make digital things, read this. If you just use them, read it too.
♥ 5↺ 2
Replying to @davelab6 and @blackfoundry, @bfluzin
I am not a lawyer either! But I happen to have reviewed a lot of Google terms recently, and making Fonts a controller really smells tbh. As a comparison, Analytics is a processor, it becomes a controller when you enable ad-related features.
Replying to @davelab6 and @blackfoundry, @bfluzin
It’s not personal data that makes one a controller, processors also have personal data — otherwise there is no GDPR.
Logging IPs for the purposes of typically associated with logging does not seem like a reason to switch to controller status.
Replying to @blackfoundry and @davelab6, @bfluzin
The law does not make it harder to know what the traffic is, it requires you to think about whether the data you are collecting for that purpose is proportional with your needs and those are well balanced with the rights and freedoms of data subjects.
♥ 2
Replying to @blackfoundry and @davelab6, @bfluzin
Obviously I am not a lawyer but I would say: if you collect just what you need to identify uniques, if you minimise the data properly (eg. keep the referring domain but not URL and keep the data only for the month you need it in before aggregating) then you’re really fine.
I told Marie-Antoinette she needed to update her privacy policy but she just replied: "Let them eat cookies!"
♥ 14↺ 4
Replying to @mnot
This is a rare occurence but I think @mnot has it backwards. We have more APIs and formats for a decentralised self-hosted whatever than one can humanly read. What we’re missing is a gritty app that extends the adjacent possible.
Quoting a tweet by @mnot ↗
I could even see this starting from a Git client! First you handle some code together, you get a way to spin up a repo in common with others. Then you chat in there, make some bits public. Git becomes the means for sharing and publishing instead of being focused around code.
Before you know it you have a bunch of client and a crufty standard that makes all authors of previous APIs and formats cringe and that someone will have to spend five years cleaning up. But it works!
Judea Pearl sometimes talks of causal diagrams (or interventions thereupon) creating information flows (in both causal and noncausal directions). Is there a good source to read on the mapping from causal to information networks?
♥ 3
Replying to @mnot
It's been too long Mark, and does anyone ever know one really? But candidates such as? I've seen stuff but it tends to fall on the "Linux on the desktop" side rather than explore a new niche.
♥ 1
Replying to @blackfoundry and @volatil_paris, @davelab6, @bfluzin
Yes, as soon as it's convenient. Before storing the IP, just apply your favourite hashing algorithm and salt!
♥ 1
Replying to @danbri
Yes, there is some of it there, but less less graphical than what I have in mind.
Replying to @jkbren and @erikphoel
I’m (somewhat) familiar with EI from Erik’s work, but the mapping to network science is pretty much what I’m curious about. I look forward to your paper!
Replying to @TheAnnaGat and @IxyHelps, @NotionHQ
I’m looking at that space too, let me know if you spot any useful info!
♥ 2
Replying to @fabricedesre and @hugozap, @pfrazee
Agreed, Beaker does a lot more than dat. It would be cool to see that level of innovation in more browsers.
♥ 2
Replying to @gchampeau
La nationalité est sans importance, mais ça coupe la défense de Facebook selon laquelle ils n'auraient partagé qu'à des partenaires de confiance triés sur le volet. J'ai visité le showroom "Surveillance de masse" de Huawei, c'est très poétique 😁
♥ 1
Replying to @trekonomics
Don't be a lonely French person! Hop over to this coast for some wine 🍷
♥ 1
Replying to @mikewest
I've declared email bankruptcy many times over, yet I always end up failing. The first day off or two...
♥ 2
Replying to @null and @dtemp_09f9, @kelseyj1123
Thanks a lot David!
Replying to @johnwilander
I've never taken nine weeks in a row, but I've certainly taken that many per year, in France. The @nytimes is wonderful but it's the first job I have where I need to plan my vacation time if I don't want to run out. It's pretty brutal.
♥ 1
Replying to @sideofhail
Wow, where can I get that suit?
Replying to @null and @clry2
I agree that some degree of targeting is acceptable, but we're very far from a trustworthy ecosystem. I have close to zero trust in the ability of most of adtech to support sufficient data protection proportional with what they know.
♥ 1
Replying to @null and @clry2
Fingerprinting is a big no. If the user said they don't want to be tracked no one gets to do it. The attributes available for targeting are also way too precise, typically with zero consideration for sensitivity.
♥ 1
"The Book of Why", by Judea Pearl and Dana Mackenzie. A good laypersonish introduction to causality. (Having some knowledge of statistics helps though.)
♥ 1
Replying to @Chri_vo and @loouisfernandes
J'imagine qu'il doit faire froid/chaud/les deux. En tout cas en Australie c'est utilisé pour l'isolation.
Replying to @null and @SimonDeDeo
“I work in data governance” is pretty effective as well. And who hasn’t moved some files around to organise things a bit better?
Mind you though, this one time the agent asked me if the XPath API I was to present at a conference could be used to implement an XSLT processor…
♥ 2
Replying to @null and @DBarriosONeill, @Roo_Kelly
Markdown is a relatively modern format, you can throw in any Unicode and it should just work, including emoji.
Replying to @amyngyn
It is a powerful tool, use it wisely.
This one time, as a young fool, I distractedly opened mine while it was still spinning full speed. I heard a loud SPLAT! When I wiped the salad from my face I saw that every wall was covered in salad.
♥ 9
For months after that we would find small bits of dried salad hiding in the remotest corners of the kitchen.
May you spin safely. 🥗
♥ 4
"Notes 11", de @Bouletcorp. Intemporel ouvrage de référence sur "le cerveau est un connard". Parfait pour un cours niveau post-graduate en proprioception de l'absurde.
I’m looking for a good introduction to the command line to help train some people here. Any recommendations?
♥ 1
And because binary XML is special, it can also binarify JSON and a bunch of other stuff, straight from the JavaScripts! exificient.github.io/javascript/
♥ 1
Replying to @xchaotic
Yeah, but I have a history with EXI :)
Replying to @vorahsa
I know, right? I was 25 when I started chairing XBC… TWENTY. FIVE.
Replying to @danbri
You’re a bad bad person Dan :)
Every time I research a question related to data governance, data management, etc. I come away with my eyes bleeding. Why? It's a space with many interesting issues that can make a real difference and is really about good design and toolsmithing. Instead we get suitbro.
Almost everything on the topic spends half of its time providing tips to convince your boss to support you. Pro tip: if your boss does not understand that data has to be usable to be useful, you're going to need a better boss.
Also, I apparently suck at my job because my team never puts any IoT in its AI. In fact, NYT DG has no IoT and no AI. I should probably resign in shame.
And since you're asking: yes, you have to blockchain all the things!
But human topics that are core to the job like "How to build tools and standards people actually want to use?" or "How to get people to actually like governance so they don't work around you all the time?" are nowhere to be seen.
Anyway, I should stop ranting. We're figuring out how to do it our own human way at The Times; if you're interested in governance and enterprise that doesn't suck let's chat!
I think the core issue is that the difficulty in producing good UI grows faster than linearly compared to the underlying complexity of what it represents.
Quoting a tweet by @dalmaer ↗
♥ 2
Just look at the best in breed WYSIWYG editors (Word, @googledocs, whatever). There is a wealth of powerful things you could do with documents, yet they are basically the same old linked-list semantics-free dumb documents that we had in the 80s.
Building a good, new, powerful UI for documents is really hard. That’s why we still have lossy formats like Word, PDF, or LaTeX. And that’s just documents.
As @jamespearce and @brianleroux debate in the comments, there was a lot of great things that VB did. I agree! I’d love a VB for Web stuff, for those simple projects. But like WYSIWYGs it would likely remain stuck at the level of power of its time.
J'adore: les lettres du Département Éducation de New York sont en anglais, espagnol, français, chinois, arabe, russe, farsi, créole haïtien, et coréen.
Je n'arrive pas à imaginer une ville française faire ça sans qu'une déferlante de connards ne hurle à la mort. J'ai tort?
♥ 9↺ 5
Replying to @fabricedesre
Genre des fachos de télévision et tout ça!
I reckon I'm definitely changing how I address people in email. Parting with "Ever yours" also a plus.
Quoting a tweet by @DerekSkillings ↗
♥ 3
Come on @Microsoft, you know you want to be about privacy!
Quoting a tweet by @caprivacyorg ↗
Replying to @karen_meep and @nukemberg, @amitayh, @martinkl
If you can't access it it no longer is personal data. In fact it's barely data. Cryptodeletion works, but it's only good for events that only map to one person. It's not always cheap, too.
Replying to @karen_meep and @nukemberg, @amitayh, @martinkl
There are two cases: you have the key somewhere, and you were able to get it back all along; or the key was destroyed. In that case, the data is as close to random as is reasonable.
♥ 1
Replying to @cepcam
Je crains que ce ne soit le cas depuis un bon bout de temps...
Replying to @cepcam
Au moins! Je me souviens d'avoir eu des discussions "Libé c'est plus un journal, c'est juste un magazine quotidien" avant même twitter.
My 8yo has raised ~$40 through lemonade sales in support of her own Kids Against Pollution Society (KAPS). They want to put it towards acquiring equipment to pick litter up.
I worry that’s not the best use they could put it to. Any alternative recommendations?
♥ 2↺ 1
Replying to @TheAnnaGat
Too late 😊
La mondialisation c'est aussi une classe de CE2 à Brooklyn qui chante Enrico Macias pour sa fête de fin d'année.
♥ 1
Replying to @TheAnnaGat
So... You prefer Vegemite?
Replying to @TheAnnaGat
Oh. So the Guinness special edition Marmite then? Or maybe the Vegemite Cadbury?
Replying to @TheAnnaGat
What’s your unpopular preference?
Replying to @TheAnnaGat
Hahah, that’s okay, those are pretty mild :)
I have in the past been supported in my work by @Microsoft. Seeing how cheerfully they will jail children for a few million dollars I can say I will never take money from them again. I encourage you to do the same, if you can.
Quoting a tweet by @wilto ↗
♥ 1
Replying to @Hodapp87 and @whyevernotso
It's also a core part of the Unix philosophy. A tiny number of abstractions do all the work through small tools.
♥ 1
Replying to @shimown and @Microsoft
They like it enough to have blogged about it as if it were an achievement. That’s cheerfully enough.
Replying to @shimown and @Microsoft
Oh yes, that’s always the issue with «tough stances» taken under relatively liberal regimes: their checks and balances are supposed to be that the people running them aren’t evil. That’s not a sane form of government.
Replying to @shimown and @Microsoft
It’s still ICE, under Trump. We’ll see, if they change their position and make amends, I’ll change mine. They’ve had a few weeks to react already, they haven't.
Replying to @shimown and @Microsoft
IBM didn’t suffer because the USA did not retaliate much against collaborating companies. It’s different in Europe where they got nationalised.
Replying to @frivoal and @Fly_Norwegian
Their payment system is very capricious. One time it charged and refunded me dozens of times in a row before failing every time I tried to pay. It's weird.
♥ 1↺ 1
"Lost in Math", by @skdh. In turns insightful, damning, funny, and personal, it also depicts a lively community (for all its sins) and is consistently sharp and clear throughout. A great read.
♥ 23↺ 2
Replying to @le_hasard
En même temps, si vous tu saignes de la main tu verrais les traces de sang. #speculation
“The White House looks to coordinate online privacy plan”
Am I the only one thinking this is the tracking industry partnering with the Trump administration to curtail the progress of privacy regulation?
axios.com/scoop-the-whit…
♥ 3↺ 1
Replying to @can and @MikeIsaac
It's not easy for news organisations to act collectively and go after a specific company. I'm not a lawyer but I'm pretty sure it would run afoul of the pretty strict antitrust regulations that govern us. Yes, I appreciate the irony in antitrust law acting to protect Facebook.
♥ 1
An excellent @EFF article on browser fingerprinting under the #GDPR. I disagree on only two points:
eff.org/deeplinks/2018…
♥ 1
First, I don't think that fingerprinting can ever hold water under legitimate interest as it deliberately circumvents the "reasonable expectations of data subjects". It is, as defined by @w3ctag, unsanctioned tracking.
♥ 2
Second, legitimate interest is not some sort of inferior compromise between privacy and business. It is much better for privacy than consent, which abdicates responsibility to the user for a system so complex they can't possibly understand.
♥ 2↺ 1
Consent also massively incentivises dark UX patterns. Those will be very hard to litigate. Legitimate interest forces those who know what is going on to act responsibly, and be accountable for it.
♥ 1↺ 1
The idea that consent is good at scale for privacy is a fairy tale that needs to die. There's a reason why Google and the rest of ad tech are pushing so hard for consent. Hint: it's not out of the kindness of their hearts.
↺ 1
Replying to @domenic and @briankardell, @annevk, @Hixie
I really hate having to be the old guy reminding you two to know your history, but intertwingularity is a foundational concept of hypertext from Ted Nelson. Everything is intertwingled.
en.m.wikipedia.org/wiki/Intertwin…
♥ 4
So, uh, yeah: rack those points up for Slytherin will you?
♥ 2
Replying to @briankardell and @domenic, @annevk, @Hixie
Ah, as usual any excuse is good for Gryffindor to steal some points.
♥ 1
Replying to @firefox and @random_walker
Easier to turn on hardly counts as privacy by default...
As a Firefox user this has been bothering me. @mozilla's principles purport that: "Individuals’ security and privacy on the internet are fundamental and must not be treated as optional." Yet their protection is optional. That makes no sense whatsoever.
Quoting a tweet by @random_walker ↗
♥ 9
I'm completely with @random_walker on this one. I have tremendous love and respect for many friends on the Chrome team, but Chrome is basically selling its users to ad tech. It's hard not to think that the relationship tout Google is causal here.
Quoting a tweet by @random_walker ↗
♥ 1
In fact, the outstanding security work of the Chrome team (which they so happily trumpet in the article) makes it all the more clear that it's such a bad browser for privacy on purpose.
Does this not qualify as consumer harm?
Replying to @bxchen
I like the article but you state: "The nonprofit has no direct relationship with advertisers; it gets a small sum from search providers like Google and Bing when a search is conducted through their sites using a Firefox product." Last I checked that was the bulk of their income!
Both Google and Bing are ad-driven, so the overwhelming majority of Firefox's income is just one step removed from ads. They do have incentives not to upset those two customers. (I'm still a happy user, just doubtful of that argument.)
Replying to @stshank and @mozilla, @firefox
Yeah, I've heard similar things but they've been working on TP for what, 2.5 years? I know there Web is hard but what's the hold up?
My 8yo: "My mum is French, so I love baguette, couscous, and pastries."
♥ 12
Replying to @andreasgal and @stshank, @mozilla
How do you mean?
Given its advertising marketshare, its browser marketshare, and the revenue difference between tracking and non-tracking ads I wonder how much money Google has riding on selling the personal data of Chrome users.
♥ 2
Replying to @jevakallio and @mweststrate, @sail_ci
In fairness, YAML always feels like a return to Makefile whitespace hell, so it reads pretty appropriate to me.
♥ 6
Replying to @jevakallio and @ndw
I reckon it's time for you to make big bucks reimplementing XProc in React 😁
♥ 4
@meredith_levien shares a few details about #GDPR and ad tech with @Digiday: digiday.com/media/nyts-mer…
The @YouTube radicalisation machine is seriously out of control. I practically never use it, yet they just sent me a push recommendation for this batshit conspiracy stuff.
♥ 2↺ 4
This isn't hard. A data science intern with two weeks of experience could write code that reliably detects this degree of crackpotness. But they're not trying. They've decided not to try.
Google is just making money off poisoning society. They need to be regulated like the polluter they are.
♥ 3↺ 1
Replying to @espadrine
I don't think they own it, they certainly didn't originate it, but I think they ⚙ feel the winds turning against Big Tech and it works well with their antiglobalist narrative.
♥ 1
That "⚙" was meant to be "too". For some reason @SwiftKey wastes tons of space on the right which makes it easy to hit the wrong completion.
Replying to @gchampeau
Donc tu as... essayé?
Replying to @gchampeau
J'y pense, parfois, mais je n'ai jamais franchi le pas.
Replying to @EurIdentitaires and @BobLeCentriste
Je ne comprends pas, elle est chouette cette image. Ça change un peu de l'image vieux con guindé style OSS 117 qui représente souvent la France.
Replying to @espadrine
Yeah, they don't really care that they're doing bad things, just that they're from outside the tribe.
♥ 1
The French have a saying (dating back to 1982), which is that «Football is a game involving 22 players, and at the end Germany wins.» #GERSWE
♥ 4
Small things can help: you can now use HTML in schema.org values as `HTML` instead of `rdf:HTML`.
Replying to @Timeo_Danaos and @GaryLineker
I had no idea! It’s something that people say in bars :)
Replying to @sr5450
Indeed, see thread!
If you make devices meant to be in people's lives and you don't design to help prevent abuse then you don't understand technology and you don't belong in tech.
"Thermostats, Locks and Lights: Digital Tools of Domestic Abuse" nyti.ms/2KdGgVC
♥ 4↺ 3
Replying to @tobie
Sure, I won't dispute that, but people will get those devices no matter what so they might as well be safe.
New York friends (who can vote): go vote in the primaries, your voice matters. When turnout is 3% just showing up is everything. And bring your friends!
Quoting a tweet by @Ocasio2018 ↗
♥ 9↺ 8
"Developers I speak with would rather dance naked in public than admit to posting a site built with hand-coded, progressively enhanced HTML, CSS, and JavaScript they understand and wrote themselves." Maybe @zeldman, but isn't the former more fun?
alistapart.com/preview/cult-o…
♥ 9↺ 2
In all seriousness though, there isn't a word in that article I disagree with.
♥ 1
Replying to @null and @Gallorum, @temptoetiam, @XetraSherpa
Dur. Je suis parti mais j'y ai grandi. Solidarité.
♥ 1
I wonder how much trouble I’d get in for putting a print of this up at work.
Quoting a tweet by @ZachWeiner ↗
Replying to @tnajournal and @LMSacasas, @adamjwhitedc
A “common baseline of facts and information” of the same kind that they actively promote through YouTube?
Replying to @seanmcarroll
Cream is unforgivable, but it is an incontrovertible fact that garlic improves virtually any foodstuff.
♥ 2
Replying to @seanmcarroll
I would agree for a spice or condiment, but garlic is a staple. Staples are not for charm, they’re for sustenance (in this case of one’s joie de vivre).
♥ 1
It looks like the CA Senate just amended the proposed bill to replace the #CCPA to limit transparency requirement to “categories of third parties” instead of listing them.
Could that kind of move lead @caprivacyorg to keep its initiative on the ballot? leginfo.legislature.ca.gov/faces/billVers…
Replying to @jaffathecake
So many characters! ev/err cut it for me.
Outstanding piece from @michelleinbklyn!
mobile.nytimes.com/2018/06/25/opi…
Replying to @LogicalAnalysis
What does it taste like? Cheese? Natto?
Yes! Hell yes @Ocasio2018! Hope is on the way!
Rep. Joseph Crowley, Longtime City Powerbroker, Loses Democratic Primary to Insurgent nyti.ms/2lCUpkg
♥ 1
Replying to @LogicalAnalysis
I know! I just haven’t had a chance to visit yet.
Replying to @drsekula and @skdh
We have a Pulitzer based on science reporting too, from the 1920s I think. I don't know if that's happened since.
Replying to @MikeIsaac
Ha! I visited the IRS this morning and the welcome desk had this article prominently framed: nytimes.com/2016/02/28/you…
♥ 1
Replying to @metaprinter and @Ocasio2018
I'm... not sure what that means in this context.
As @WilliamsJames_ said, "the freedoms of the future are the freedoms of the mind." We gladly welcome California to the future with its new privacy law, may it be kept safe from lobbyists!
wired.com/story/californ…
♥ 2↺ 1
You'd sort of hope that in 2018 Wired wouldn't repeat such blatantly false statements about the #GDPR requiring consent, though. Cc @WIRED @issielapowsky
Replying to @shimown and @WilliamsJames_
If you like the quote read the book 😉
That media should maintain the truth but search engines may need to remove results turned irrelevant is the best option for the right to be forgotten.
Quoting a tweet by @nbitiukova ↗
Replying to @shimown and @WilliamsJames_
Well, it’s a book about services that get you hooked and milk your attention, so it seemed like the right thing to do.
Replying to @thekitze and @dan_abramov
It's also spectacularly ignorant of history. We had CSS in JS in the late 90s. We've had variants on JSX every few years for a long time. If I had to do it again I would still go with the principle of least power.
♥ 1
Replying to @baekdal
My concern is that this will only get worse if upcoming privacy regulations increase reliance on consent over other legal bases. Consent abdicates responsibility to the user and incentivises dark UX patterns. And fatigue would be a problem even without that.
Replying to @baekdal
The publication/transparency requirements on legitimate interest could be strengthened, which puts the onus of responsible behaviour back on the data controller, where it belongs. I believe that would have a genuine positive impact.
Replying to @jensimmons
This just baffles me. How could they ever get to be "amazing experienced" with that degree of stupid behaviour?
Replying to @baekdal
There is certainly a great variety of compliance levels, but I don't think that one can state that everyone is breaking the law without much more careful analysis than that article provides.
Replying to @baekdal and @Digiday
That's why there's an option in DFP (and others) to disable personalised advertising. But you can't guess that from a cursory glance at cookie banners.
That problem however can and should be fixed at its root: browsers.
Replying to @baekdal
Beyond that, I agree with you about the bad guys problem. The ad world is in a needless data frenzy and ad-based publishers have little option other than to comply.
Replying to @baekdal and @Digiday
It does not under legitimate interest! You have to afford the right to object to non-essential processing.
Replying to @baekdal and @Digiday
Oh, I'm not saying we have no choice. We do, and we must act (and some of us are). But the incentives are set up in a really nasty way that means being responsible is a risk position. This does not help privacy.
Replying to @baekdal and @Digiday
I *completely* agree with your sentiment, it's your manner of assessing GDPR compliance that I think is too broad-brush.
Replying to @baekdal and @Digiday
Part of the problem is that when we it comes to ad tech this is not a controller/processor relationship that would mirror the fashion world. Google for instance claim joint controllership (while rejecting responsibility for obtaining consent).
Replying to @baekdal and @Digiday
It might be notamment show evidently but a lot of publishers have massively limited data sharing to the ad world. Is it perfect? Not at all, but I've seen a lot of progress.
Replying to @baekdal and @Digiday
Again, that is only true with consent as the legal basis! I think what's going murky is that one has to guess what a publisher's reasoning is from very little visible information. That's why I'd like to see stronger transparency for legitimate interest.
Replying to @baekdal and @Digiday
I think we're going around in circles between GDPR consent versus ePrivacy Directive consent, and the requirements of other legal bases. This might be better discussed over the drinks 😊
Replying to @baekdal and @Digiday
But that's precisely why people are disabling personalised ads!
Replying to @baekdal and @Digiday
Obviously! I didn’t mean people as in users but as in publishers disabling personalisation in DFP et al.
Replying to @baekdal and @Digiday
And, to conclude: as a Web guy I think the Web world needs to take responsibility for the negative incentives it creates. You can’t just sell weapons and claim innocence.
Have a great weekend!
Replying to @trekonomics
Nah, killing a game with defence when you’re ahead is what the teams we don’t like do. Keep attacking!
♥ 1