July 2018
272 tweets
At the beach in Coney Island, every other t-shirt is an @Ocasio2018 shirt. 👕 Hope is back in fashion!
♥ 1↺ 1
Good luck Mexican friends! 🤞 🇲🇽
♥ 4↺ 1
There's a free rider problem in that I have little patience for debating but at times great interest in following the debates of others.
♥ 2
Multicellularity appears much more easily, at least based off of modern organisms, than I knew was possible.
sciencemag.org/news/2018/06/m…
↺ 1
Best captain too 🤗
The jury is still out on beer-inspired gelato, though.
♥ 1
Do you have anything in email that you wouldn’t want random strangers to read?
«Google reportedly allows outside app developers to read people's Gmails»
Quoting a tweet by @tsmullaney ↗
♥ 2↺ 2
There is a huge difference between allowing a local client access and allowing a third party access. No developer of an email client has ever been able to read my email.
Bring your kid?
Because Little Snitch would tell me!
Asking that "internet companies must place humanity first and profits second" is sweet but wishful. The problem won't just fix itself.
Quoting a tweet by @nytopinion ↗
♥ 3↺ 1
Entreprise avec une présence en France et un lien avec le Web? Ça vaut vraiment le coup de sponsoriser @ParisWeb, une des rares conférences françaises de qualité internationale. paris-web.fr/telechargement…
♥ 2↺ 10
Until such a time as we figure out how to make cosmopolitanism belonging work, I agree with @asymmetricinfo. When you build a State without some form of belonging, you get the European Union. Much as I love the project, it's not an unmitigated success.
Quoting a tweet by @asymmetricinfo ↗
There’s a bit of it in this thread: tweets.berjon.com/1009611570274619393. I should probably take the time to write it up properly.
♥ 1
There is no multiverse problem that a good theodicy can't fix.
♥ 1↺ 1
Back to the Future has a futuristic kitchen.
♥ 1
I haven’t tested every last page, but it happens in most locations I use. It’s not always slow enough to trigger the dialog, but it’s always watch-the-text-catch-up slow.
Est-ce que vous auriez des recommandations de colos près de Paris vers la fin août? Merci!
♥ 1
Whoever Putin tells him to!
It might! Sadly we'll see.
Any experience building a search engine (mostly for HTML and JSON, in the million-docs range) in Node?
I’ve tried a bunch of stuff but not finding anything great. search-index fails on Node 10, Elastic has unclear full-text support, etc.
♥ 1
Hmmmm, I’d missed that Pouch now supports Mango. That might be an option. There is no world in which I am ever writing a Couch-style map-reduce again.
Hahaha
I'm afraid that might be worthy of some double-checking 🤔
Both my dictionary and Wikipedia say it's of unknown origin, with citations to match. Etymology is a world of lies 😅
Does it work on no budget? 😇
♥ 1
"This internal self-consistency of the quantum formalism is general, and it is perhaps its most remarkable aspect."
And, so... hasn't anyone given it a name? From plato.stanford.edu/entries/qm-rel…
♥ 3↺ 1
The Vorrh, followed by The Erstwhile. Volume 3 is out in a few days.
Additionally, we might need to do something about SVG so that people screw it up less!
♥ 1
Haha, that's a good start! But is it really self-dependence given that self-measurement is impossible?
Annotating books can work as well.
That's fair. I'll admit I'm still working on my diracish mannerisms.
♥ 2
If I get hit by a bus please tell the world that I’ve brought the measurement problem to a satisfying conclusion in one of those margins…
♥ 1
«Mémés»?
♥ 1
Ha, same, right knee. It's still the only time I've ever seen one of my bones, so #NoRegrets.
♥ 1
I’m in no hurry to see my bones again, but I still distinctly recall sitting on the playground looking at this hard white thing surrounded by flesh darkened by dirt. Probably my favourite scar!
♥ 1
I remember visiting Huawei’s public surveillance showroom in Shenzhen a few years ago. It was exactly like any high-tech snazzy, shiny, glossy showroom, except about dystopia. We were guided through it by a cheerful, enthusiastic youngster. nytimes.com/2018/07/08/bus…
♥ 2↺ 4
Then again, for all the «OMG China!» handwringing this is no different from what happens to you when you browse the Web without some form of tracking protection. You just have to hope that Google and ad tech are benevolent, instead of the Party.
That’s not an obvious bet.
♥ 2↺ 1
Could you use a reliably interoperable ridge to get the colours, then use those to generate a test?
I meant interoperable in shape. Rasterise, get browser-specific colours from that?
Don't they use the old system colours?
I just found out about @thedoband’s live album, and it’s pretty awesome.
open.spotify.com/album/5TGAFIJ6…
OH: “Is it still bikeshedding if _they_ think it's not trivial?”
♥ 2
That's machine fucking learning for you.
The more I see how the ad industry is approaching #GDPR the clearer it is that consent as a legal basis needs to be dragged into the back alley and shot. If ePrivacy doesn't change tack this might have been for nothing.
Quoting a tweet by @baekdal ↗
♥ 2↺ 1
Fair enough, next time I will look for something that doesn't actually happen to people here.
♥ 1
That might spill over. Guillotine?
♥ 1
How are we supposed to progress towards social justice when there isn't even a guillotine emoji?
♥ 18↺ 5
If you think whales are lazier than koalas I have news for you. 🐳 🐨
♥ 2
It’s a lovely gif, but I kinda like the unobtrusiveness of emoji :)
Well, Koalas evolved from Drop Bears and essentially inherited their hunting style of just hanging from a branch, waiting for a prey to walk by, and dropping on their heads to eat their brains. Except they don’t even drop anymore.
If there’s lazier than that I don’t know.
♥ 2
Let’s put it on the roof!
♥ 1
“Subjective and strung together with little narrative” is how I roll, Amy!
♥ 1
😽
Yeah, I’ll admit that communicating through dead plant genitals has never been my vase of tea either.
♥ 1
“The angel octopus robot has had its heart darkened by fear (it’s a Jedi) — WTF!”
♥ 1
imgur.com/gallery/vTCW4Cy :) (Can’t find the original @SMBCComics, comics SEO is a pain.)
It’s “only” a regional court, but if this ends up setting the bar for data minimisation under the #GDPR things are going to get pretty strict! natlawreview.com/article/german…
♥ 1↺ 1
Please root for Belgium 🇧🇪!
I’m not sure canned rosé is an approved drink for this circumstance.
♥ 1
A bit more about René Carmille, the French Résistance's hacker who screwed with the IBM punchcards of the Vichy census to help keep many Jews from deportation.
blog.cybertraining365.com/2017/02/19/hac…
♥ 9↺ 5
The battle lines are barely drawn. It's going to take another twenty years!
Hahaha, on that I can't disagree, but then again I probably stopped being Continental after Leibniz 😏
So the latter. We are barely understanding our first experiments. We are only beginning to grasp what modern freedoms are about.
♥ 1
Oh, Leibniz *inclusive*. Can't do much without Leibniz!
Love the graphics work from @Digiday here.
digiday.com/media/new-york…
When your leading metric is engagement, fake news is a boon; it certainly represents a large chunk of Facebook's bottom line (and Google’s, particularly through YouTube).
Yes, it'll kill them in the longer run. Why do it anyway? See: The Scorpion and the Frog.
Quoting a tweet by @mcnees ↗
♥ 3↺ 2
The ability to converse on the phone. I'm getting ever so slightly better over time, with extensive concentration.
♥ 1
Really? Mostly I just mentally drift somewhere else :)
Et c’est bien meilleur sec!
En général il vaut mieux éviter de me croire sur parole trop longtemps, essaie d'en boire rapidement.
♥ 1
There should be a name for this international alliance of Putin's stooges.
"Trump Inserts Himself Into Brexit Debate, Criticizing Theresa May" nyti.ms/2mbKcM5
I find that statement needlessly... polarising.
The iciness of your reception to my joke is... polar.
But can you take the pole out of the slav, because, you know, ouch.
♥ 1
Hang in there!
♥ 2
Faced with severe shortage Britain requests that the EU keep open the common market and open borders for Conservative Ministers.
Quoting a tweet by @Floridi ↗
♥ 3↺ 2
Wait, you can notice only now that Trump has been depleting the Strategic Snake Oil Reserves?
Quoting a tweet by @axios ↗
♥ 4
300 Ashland, penthouse. Il y a Victoire et Alban, je crois que vous vous connaissez ☺️
♥ 2
Your forest just has higher trees, so the buildings can't be seen so much.
♥ 1
I'm hardly an expert but 20 to 1 seems way off. When did they close?
I've never encountered the shampoo one!
If anyone asks you what the role governance is in technology (and why so few seem to care) point them to this @zeynep thread.
Quoting a tweet by @zeynep ↗
♥ 3
Oh yeah, security is a form of governance. They're our cousins in dark hoodies 🤗
He who has the last slav slavs loudest.
♥ 1
You know it.
This seems to require logging in to the list server? Is there another link?
♥ 2
I miss the macarena.
No, it's not. Leavers are settling things up so that when #Brexit is the failure it can't avoid being they can blame it on not being hard enough. Then they can get back to their usual fear-mongering.
Quoting a tweet by @lawfareblog ↗
♥ 3
Is it possible that you answered your own question with: "if a gene's name has a hyphen in it"?
♥ 14↺ 1
Melatonin, force yourself to see sunlight regularly in the day (especially the morning, in London this can require some dedication), eat proper meals at the right hours, plenty of water, no screens after 8pm, read if you can't sleep even a dumb book because tired (no screen!).
♥ 1
Until such a time as we can get rid of them entirely we should work as an industry to force all third-party code into supporting SRI. Anyone up for making that happen? theregister.co.uk/AMP/2018/07/12…
♥ 1↺ 1
3 and 4, but I would double-check the data protection implications.
♥ 1
You could live only half that *and* be a geoduck. So, silver lining right there and all.
Il me manque une étape. Le rot vient nécessairement d'une interaction entre systèmes digestif et respiratoire et les leurs sont déconnectés?
I'm obviously all for that but is there any interest in privacy from browser vendors other than WebKit?
♥ 1
Apart from @brave none of those is on by default, I'd be shocked if telemetry showed any major uptake? There is a lot of talk about privacy features indeed but it feels like a lot of maybes.
♥ 2
Ouch, oh dear, people are stillness doing that? Sometimes it feels like I spent half my time at science.ai trying to push back on just that 😑
♥ 2
Yes I mentioned that! ☺️
♥ 1
Oh, we were so young! I even thought that user consent could help with privacy, that's so adorable!
♥ 1
The workshop mentions relating to regulations, do you expect attendance from regulators?
Yeah but Focus forgets everything, that's a bit much!
Careful what you wish for. Citizens United essentially made bribery legal in the US, I don’t think the Supreme Court would necessarily stop at considering fraud free speech.
Yeah — HTML all wrapped in React instead of the other way around, CSS-in-JS, Redux for trivial stuff… all killed Web dev for me; I’m now doing something else entirely.
Wake me up when the Web gets over its “let’s make all the same mistakes Java made!” phase.
♥ 19↺ 4
Don’t complain Dave, if we do the same thing every decade eventually we’re bound to get it right.
Yes, I think it should. No one should get to update the code that’s running on my site but me!
♥ 1
In seriousness, I completely agree. Our threat model was largely wrong. The regulatory environment has changed.
♥ 1
I would love to see a sociological study of who, in 2018,is still using Facebook.
♥ 2↺ 1
Yeah I meant actually use, not just keep around for some distant relative who still writes on your wall once in a while.
I suspect the only reason that Facebook stock isn't junk is Instagram.
Oh yeah, I've never met a teen using Facebook. They don't understand the Christmas tree UI nor why you'd share random uncontrolled stuff with people.
I gave up on Instagram years ago, so I wouldn't know. My vice is Twitter 🤗
♥ 1
Oh yeah, I'm not saying it's bad. I'm just not into pics that much.
The EU isn't moving "to rein in the clout of American tech companies" @satariano, it's following its mission to uphold free markets and democracy.
This does regularly hit American companies. Huh, I really wonder why. 🤔
nyti.ms/2NXlU5f
Sigh, Koltès 💔
♥ 1
Yes, there have been economic nationalists who want to return to the postwar model of the national champions for a while (and it's how China operates). But I've seen little such nonsense from @vestager or Competition generally.
Ah, fond memories of seeing some of his productions!
That sort of nativist nonsense I've only seen from politicians, there's plenty of innovation on the ground and no one cares where you're from.
But Google is abusing its dominance left, right, and centre. That I've seen it first hand working for the media, in science publishing, for mobile operators and manufacturers, and with browser vendors is IMHO quite telling in itself.
The @nytimes's security team is great, I'm proud to be working with such talented people.
open.nytimes.com/introducing-lo…
♥ 2
Dadaism?
♥ 1
I think it might be a matter of our brains not being built to process what that language describes, either.
♥ 1
One thing that surprises me: it would seem that (at least in the US) teens not only still use SMS but apparently sometimes in preference to most other channels.
♥ 2
Remember when Nokia boasted that they had the world's largest social network, being address books on phones (and then did nothing with it)?
I wonder if there's still room for that to happen.
♥ 2
They should be the ones striving 😏
♥ 1↺ 1
Awww, ça m'a fait super plaisir de vous voir 😘
Seriously tomorrow I am storming into Laura's office and asking that we rename things "cyberdigital". It's where the future's at.
♥ 1
In her cyberdigital office, silly! 🦑
♥ 1
That’s one way. There are some small libs that do that, I generally much prefer them to Redux et al. A recent one I saw (but haven’t tried yet) is valoo: npmjs.com/package/valoo
♥ 3↺ 1
I haven’t tried lit yet (but want to!); it looks like the two would work really well together!
♥ 1
I’m really glad that someone wrote this about JS const. It’s bloody useless. It needs to die in the fire of its own cargo-culting stupidity.
Quoting a tweet by @jwajsberg ↗
It’s not huge in itself but the sheer amount of repetitive boilerplate will kill you.
If you’d paid the enslaved for their work but they’d remained owned and branded, you wouldn’t have ended slavery.
In the same manner, paying people for their personal data does not bring back privacy.
Quoting a tweet by @ProSyn ↗
♥ 2
Promising the end of “digital feudalism” sure sounds neat, but I have read the full original @GLthinktank/@GaspardKoenig report and it is full of elementary errors about privacy, the data economy, or the actual goals and effects of privacy regulation.
♥ 1
Property rights for personal data might make sense if personal data were sold as address books, if there were no externalities from refusing to sell, and if the processing involved in the data economy were as simplistic as the report believes.
Not one of these conditions holds.
♥ 2
A general rule for personal data property rights proposals is to check if they even answer questions like:
• Does lying about yourself become theft/fraud?
• What happens to that one kid in class who refuses to sell his sexual orientation information?
No answer? ➡️ 🗑
♥ 1
I almost forgot! The data of affluent people are also more valuable (for obvious reasons). Privacy property rights then make the rich richer overnight without any change in the actual production of value.
I could go on… The fact is: it’s about as illiberal an idea as it gets.
Je ne suis pas sûr de ce que tu en attendrais? Si ePrivacy renforce le consentement (comme prévu) ça va juste empirer la situation. Cc @alex_lenoir.
Tu veux dire DNT? C’est un peu nulle part pour l’instant…
Le site de la @SNCF:
• Ne marche pas sous Firefox (recherche de train tronquée).
• Pense qu’on est déja demain.
Mais franchement comment vous faites pour être aussi mauvais?
♥ 1↺ 2
Je connais, je voulais juste vérifier un horaire qui me semblait louche.
Accessoirement, @trainline_fr ne marche presque jamais avec une CB US, et leur intégration PayPay mouline un peu dans la semoule.
I'm not sure what you mean? It's not like you'll be shipping npm to the browser?
Bring natural philosophy back!
♥ 1
Some are, some aren't. Something like valoo is 122B and has no dependencies, so yours pretty much all set!
I'm glad there's someone out there who got the joke!
You can't install node?
So where do the audio recordings go when the ML fails to understand the prompt? Are those humans all Google employees? Is the audio notification transferred?
♥ 1
On dirait que quelqu'un fait des calculs en local time... Ou en mélangeant les deux.
Ouais, ou des opérations qui mélangent ça plusieurs fois sans faire gaffe. Il était off de beaucoup...
D'autant plus aux US! C'est vraiment WTF.
I find it somewhat alarming that you would not consider privacy a fundamental right. It is just as essential to democracy as free speech and attacking a journalist for supporting it does not strike me as constructive for all that it is fashionable.
♥ 3↺ 1
Which part of the below tweet is about the report and not @BrusselsGeek's own opinions?
Quoting a tweet by @mmatthiesen ↗
♥ 1
I will let her say whether that is a faithful representation of her position rather than a strawman, but I have to point out that "data-driven" is not a parenthetical in impact, nor is it an unavoidable part of the ecosystem.
♥ 4
I have never heard anyone argue that we should offer services against our will.
I have, however, spoken with many publishers who would like not to have to sell their users downriver to adtech against their will.
♥ 1
My experience with smaller publishers is that they care about their readers just as much. They have options for scale (eg. co-op variants).
♥ 2
The current race to the bottom for impressions is forcing all publishers that can't sustain massive social media impact into content farms anyway. It's unsustainable, let's not pretend that adtech is on their side.
♥ 4↺ 1
Wait. You're simultaneously saying that it's the publishers' choice but at the same time they can't change their business model. Which is it?
♥ 1↺ 1
I have zero worry that advertising won't find a business model when it can't use people's data. It always has. The data bubble will burst, things will be rocky for a couple years, and advertising will find a way. Always.
♥ 3
Lawmakers work to protect democracy, it's sort of a lobby group that way. We pay them for it.
Please cite one service that credibly has no alternative.
Calling it "data-driven" is like saying that cotton picking is "muscle-driven". It sort of misses the point.
Hey, I get it. This will wipe out half of adtech. But if you want to enlist support from publishers at least say which.
♥ 2
It's hard to discuss evidence based on off a report with murky sources. It would also need to be weighed against damages to privacy and the decrease in article quality when impressions are at the helm.
♥ 1
Ah, I was just wondering when the "but the market should solve this" argument was going to show up! Financial services are heavily regulated because they are so complex that the knowledge asymmetry is heavily detrimental to users. 1/n
♥ 2
Adtech and what is generally done to personal data is exactly the same. Consent to that is a travesty, regulation is needed.
Like smoking, the injury from privacy loss is separated in time from its cause. Humans suck at that, regulation is needed. (2/n)
♥ 1
(Sorry, interrupted by the French countryside.)
Like pollution, individual preference, industrial choice, and eventual harm areas misaligned. Again, not one for the market. (3/3)
I do not speak in any manner of official capacity but my personal perspective is that I would be surprised if our views aligned with the platforms on ePR. Please don't use publishers as human shields for adtech when regulation is badly needed.
You can get a very good proxy for that by looking at media that still have distinct paper vs digital newsrooms and seeing how many articles per day are expected from writers on either side.
Some what happens in that world when browsers stop enabling tracking?
That is an astounding statement. I'm not sure that there is a time in history when publishers have had *less* control over their supply chain.
What is economically viable depends on the regulatory environment. It won't go away but it can be changed.
♥ 6↺ 1
I think the jury is still very much out on whether the framework works. Are you publishing interoperability and testing reports to establish that?
♥ 1
Hahaha 😂 Sadly, I can't discuss any details from those discussions. But (in a completely unrelated statement) I find the word "choice" very, very funny.
Wait, there's no testing? Seriously Matthias, I was extending an olive branch around the framework but how can you say it works without the means to establish than empirically?
♥ 1
Are you folks still open for feedback BTW? There's problematic stuff in there...
Used != works. Who is in charge of checking that's it's not just operating under cover-your-arse assumptions? Because if no one checks, everyone remains fully liable...
♥ 4
That's why specs aren't usually called standards before they include the means to demonstrate interoperability.
♥ 2
Between control preferences (publisher or CMP) and what actually happens. That it can work is obvious; that there is an empirical test for it is what makes it a standard. Otherwise it's just a theory!
♥ 2
Wait. How does the Framework address this type of malvertising?
That when a client requests something the server does it is, like, Protocol Interoperability 101. You keep using the word "compliance". Compliance without testing is like science without experiment: devoid of content.
♥ 1
So you would ban all JS in creatives unless they are accepted for a given vendor/purpose reason? If not then top.location = "http://evil" remains just as easy.
So what you're saying is that the level of maturity of the @IABEurope Framework is that you've shown people can exchange JSON and bitstrings. That's a pretty low bar to call it working.
♥ 3
Ok, so none of that prevents a redirect. Which is fine, the Framework was never billed as a solution to that, I'm not sure why you said it would!
I've been doing standards with many different organisations for twenty years: I'm not at all surprised that it's hard. It's *because* it's hard that we generally don't consider a specification meaningfully standard unless it comes with tests sufficient to prove it!
♥ 3
Subtweet much? Speaking personally, I would fully expect imperfection, that would never be a deterrent to participation (in fact, if it were perfect why bother participating?).
However if I were to recommend deploying resources I would have to see a genuine commitment to making it a real standard rather than a CYA/PR op. Your attitude to compliance and testing does not reassure me.
Things that want to be standards fail all the time. I get requests for resources for these things every week. Without minimal commitment to basic fundamentals of quality on the part of the org in charge, it's hard to see the point in joining tbh.
It's not about "could", it's about proof, about establishing quality. Trust me, USB can prove interoperability with an extensive test suite. That you don't understand the issue is in itself problematic!
I did not call it a CYA/PR op, I'm trying to *help* you prevent it from turning into one. If it doesn't have a built-in reality check, that's what it'll turn into.
Immediately giving you bona fide feedback right now and all you're doing is saying the problem does not exist and being hostile about it. How can you simultaneously complain that we're not participating and reject constructive feedback about basic issues with the Framework?
An open source project without solid testing *is* a major red flag. Honestly, I don't know how to put this more clearly. It's just a very elementary engineering consideration.
♥ 3
Baudelaire seems like a strange comparison point to me, I would almost have rated them antithetical. Pessoa is so much more subtle.
♥ 2
Ditch digging frozen ground by hand?
Can’t disagree!
♥ 1
Can we simply file issues at github.com/InteractiveAdv… ? That should provide more context for clarity. Is the plan to handle them faster than for OpenRTB?
♥ 1
I’m a bit puzzled… @ChrisAldrich says “your personal identity on the internet can end up fragmented like a Horcrux” like that’s a _bad_ thing. #IWannaLiveForever
alistapart.com/article/webmen…
♥ 1
TFW a co-worker translates "hot tip" as "leçon chaude". 😁🔥
♥ 5↺ 1
I'm going to stick with leçon chaude 😏
♥ 1
Social media mob dynamics are a ritualised form of prëexisting mob dynamics; noisier but less harmful. Ritualising them further is not necessarily a dystopian future.
♥ 1
Oh it's definitely not as snappy, do not use it unless you really mean the innuendo.
If you want to consider it an attractor (which might indeed make sense) then there must be forces preventing it from devolving into pure mob violence. Social media is complex, it could _also_ reinforce those.
Outsourcing accountability? Building an SEP field?
♥ 1
Littlehampton know fact: the American diet contains 700 times the daily recommended amount of salt precisely to avoid the icicle issue.
@runasand is one of the reasons why it's great to work at The Times, I'm really happy to see her profiled! nytimes.com/2018/07/24/ins…
♥ 11
N'oublions pas l'agence SpinYak et ses consultants velus.
♥ 2
Tout de suite les menaces...
It might be worth reaching out, any solution will have to work with the law. Also, I sense a belief from regulators that DNT solved the all the problems. It would be useful to have that conversation.
♥ 1
Do people really go to business school to become entrepreneurs, though? It does make the rap sheet for toxo look pretty cool, with increased risk of:
• car accidents
• mental illness
• neuroticism
• drug abuse
• suicide
• entrepreneurship
This might be the best description of the newspaper business in the 20th century.
Quoting a tweet by @asymmetricinfo ↗
♥ 5↺ 3
Does someone maintain a compendium of worldwide privacy regulation?
Are there extensions that switch DNT on for people (maybe even without asking) rather than require them to go monkey around in preferences?
I ask because people seem to care a lot more about DNT than I had possibly expected. Well, well beyond a few privacy geeks.
I bring it to school and then I drop the kids off and keep the lunch.
♥ 1
So @IABEurope’s report on the value of behavioural targeting deliberately included Google and Facebook as «publishers» so as to falsely (and massively!) inflate the economic impact on, well, actual publishers.
That’s not exactly building an ecosystem of trust.
Quoting a tweet by @johnnyryan ↗
♥ 12↺ 8
I wonder if it’s the same group that writes the GCP docs, those are terrible.
Yeah, there’s a lot that should be illegal in docs-writing. It’s probably easier to list what legal looks like.
♥ 1
There are several Markdown or similar options, notably Mermaid. Typora exposes a bunch of them.
♥ 1
With maybe the occasional exception, eg. Stand Out Of Our Light is explicitly Diogenian (Diogenesian?) if admittedly lighter on the coruscation. That's what Twitter is for.
♥ 1
Euh, regardez à la réouverture du marché...
NASDAQ reopens in a few minutes. Any bets as to how much Facebook will be tanking by? Are we looking at a privacy-driven minicrash of internet stocks?
↺ 1
Certes, mais l’audition n’a eu aucun effect, alors qu’ici c’est à la suite d’un call avec de mauvais résultats. Je pense que ça a des chances de coller plus longtemps.
A wonderful programme, so many outstanding questions. It has all the promise of a breath of fresh air, I much look forward to it!
medium.com/ixy-labs/is-th…
♥ 1
Firstly they ignore you. Then they laugh at you. Then the @w3ctag obsoletes you. And then you win. 😂
♥ 4
It's several things. Their EU audience has shrunk (for the first time). Their numbers are bad on several fronts. People bolting and shareholders bolting are two different dynamics.
If you think that Facebook's audience shrinking for the first time in Europe is shocking, wait until they start actually removing the bots. 📉
♥ 11↺ 4
Ixy feature?
Despentes 2022.
↺ 2
No one used to care about DNT because no one thought users would turn an obscure, half-hidden feature on. Looking at actual numbers might change your mind.
Quoting a tweet by @drogersuk ↗
♥ 1↺ 1
We didn't care because users seemed not to care (and browsers made it hard, you can't even find the setting in Chrome by searching for "DNT" FFS), but now that I see they do care it's changing my mind. Folks should check their stats.
♥ 1
That's a good question and it depends notably on whether DNT is legally binding and on whether there is a clear path to exceptions. Browser tools are not always very nuanced and sites don't have a standard way to communicate exceptions.
♥ 1↺ 1
Some apps I allow telemetry, others not. If browsers just brutally block, that doesn't work, whereas with an exception mechanism there's a way. Also, most browsers blocking is about third parties (it's hard to usefully block first parties without breaking).
↺ 1
But DNT has useful first-party semantics. I'm still thinking it over but I may be switching over from doubting. We're seeing DNT above 15%, even in Chrome which hides it carries and doesn't even use it in incognito. That's very high given the conditions!
Wait, what, to the AB? I chair nothing on vacation 😘
♥ 1
Beautiful work on the contrast.
Putain, il y faut faire quelque chose pour le café en France les enfants. C'est pas mortellement compliqué, il y a l'Italie juste à côté, je suis sûr qu'ils accepteraient de faire des formations à but humanitaire. ☕
♥ 6↺ 2
Non, je crois que certains ne savent pas, mais preuve: Nespresso ça ne marche que dans les pays à mauvais café, et ça cartonne en France.
♥ 1
Ah, merde, j'espérais avoir une bonne théorie... Bon, en tout cas il est temps d'agir, on ne peut pas laisser les gens se faire mal à la bouche comme ça!
♥ 1
Certes, mais c'est moins drôle
Mois j'en ai pas bu pendant longtemps parce que c'était toujours dégueulasse 😊
You're an old soul.
Privacy policies are a dreadful tool with which to produce transparency. We need not just to go beyond transparency, but also to do it better.
Quoting a tweet by @privacyint ↗
♥ 3↺ 1
Some would say polls are better when there's more than one answer...
♥ 1
Prescriptivist?
I don't think verbing coffee looks like a typo, but that's likely contextual!
Ouch!
♥ 1
That, too, is contextual I guess! 😏
♥ 1
9yo: If we all had a the same middle name AND we had to use it, then when we fought for a revolution the bad people were could never identify and find us.
I think she still needs a little more training. #ImprobablePloy
♥ 3
Automating doesn't have to be AI-based, though. Algorithmic automation (which need not be implemented in silicon, eg. French election validation is algorithmic but with people and paper) can go far. Gerrymandering is a problem that's begging for this type of solution.
♥ 1
And also that revolution is a thing you do 😊
I certainly agree, and see why. It's just that it's been a running joke that I've been in the employ notably of MIT and Keio without a degree, so it made me smile that you'd got one. (Congrats BTW!)















