Replying to @Rich_Harris and @nomadtechie, @AMPhtml, @openjsf
Open governance requires actual effect, too, not just an advisory body.
♥ 2
September 2020
291 tweets
What's this "September" of which you speak?
♥ 10
Replying to @Sniperovitch
It *looks* like April!
Replying to @jkohlmann and @roseveleth, @swodinsky, @Dr_C_Taylor, @nappy_techie, @shoshanazuboff, @Chronotope
Thanks! Though I would hardly qualify as either academic or journalist!
♥ 2
Replying to @ProfFerguson and @roseveleth, @hypervisible, @ruha9, @EvanSelinger, @alvarombedoya, @daniellecitron
Also @hartzog on that list.
♥ 9
Replying to @jkohlmann and @roseveleth, @swodinsky, @Dr_C_Taylor, @nappy_techie, @shoshanazuboff, @Chronotope
In house, on this topic, we have @Aaron_Krolik, @natashanyt, @kashhill, @jenvalentino, @fmanjoo, @stuartathompson... I know I'm forgetting some right now!
♥ 2
One thing worth noticing in this "chickenisation" thread from @doctorow is that it's not just how the gig economy operates: it's also the situation of much of the media today. I don't think that the parallel is coincidental.
Quoting a tweet by @JuliaAngwin ↗
♥ 26↺ 11
Much like poultry farmers, publishers carry the investment risk but have to abide by standards set by platforms that will decide their success. This week it's AMP, Apple News Format, and FBIA; next week it's WebBundles. Now news are all important. Oh wait no, people connecting.
♥ 3↺ 1
When the rules change some will go under, the platforms stay. Remember "pivot to video"? Who took the fall?
It's not a sustainable system for producers, but that's not what is being optimised for.
♥ 3↺ 1
Replying to @null and @SimonDeDeo
I have the same question but with A/B testing, with the added complication that I think it also harms human learning. If they produce something from that question, I'd love to read.
Replying to @null and @mccamj
Much agreed, one of the best guides to understanding tech today, too!
"The entire internet is now a walled garden." — Barry Lynn, speaking to @TheMarkup twitch.tv/teammarkup
♥ 5↺ 1
One of the weird side-effects of having spent so much of the past six months talking to faces inside a video rectangle is that when I now see faces inside a video rectangle it's very hard not to talk to them.
♥ 6
"If you have a monopoly, you have the power to govern. But we don't want the people who have the monopoly to govern the people who depend on it." — Barry Lynn, @openmarkets, speaking to @TheMarkup.
♥ 13↺ 8
Replying to @brannonj and @openmarkets, @themarkup, @CJL4News
Absolutely. Independence of the press depends crucially on tech antitrust.
♥ 4
Never doubt that partisan short-sightedness can stand in the way of protecting democracy.
Quoting a tweet by @nxthompson ↗
♥ 7↺ 1
Replying to @BrendanEich and @cramforce, @jyasskin, @slightlylate, @aboodman
I think that a good question to ask of new web technologies is whether they help or halt the continued centralisation of the web. As defined, this one solidifies the massive centralising landgrab made by AMP. That's not a great option.
♥ 11↺ 2
What I'm still somewhat puzzled about is that what seem to be the most important use cases don't require that. Single origin, unsigned bundles would seem like the logical experiment to start with, and judge issues and the need for increments from there.
♥ 1
A native app store is two things from a single entity: discovery and monetisation.
When the web becomes such that, outside hobbyists, all discovery and monetisation is from a single entity then it is indistinguishable from native.
That's not a Web worth wanting.
♥ 23↺ 5
Replying to @jyasskin and @BrendanEich, @cramforce, @slightlylate, @aboodman
Maybe there's some kind of consensus process that we could all follow to get somewhere.
♥ 2
"The entitlement of these fuckers is just off the charts. They have zero right, none, to the tracking they’ve been getting away with."
Completely right.
Quoting a tweet by @daringfireball ↗
♥ 18↺ 6
Replying to @daringfireball
Tying privacy issues to real-world situations is how I've been approaching it (except I tend to involve wine as well). See open.nytimes.com/how-the-new-yo…
♥ 1
Home pages, you’ve discovered home pages, @techreview. technologyreview.com/2020/09/03/100…
♥ 8
Replying to @oliviasolon
I just don't think I've ever attended one that wouldn't have been better as a much shorter read, but maybe I'm going to the wrong ones?
♥ 1
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
How do you expect to ever get legally valid consent? It's not happening today with CMPs and the TCF. Putting invalid consent on the blockchain just gives you traceable invalid consent.
♥ 1
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
Actually, you can give a different email per origin.
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
How is this even an equivalence? Typical CDNs take strong pro-privacy protections. Yes, they could cheat, but they don't. In OpenRTB violating privacy is the norm. Actual threats are the problem, not hypothetical ones.
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
No. Adtech doesn't have to be bad, in the same way that CDNs don't have to be - and aren't. It's the browsers' job to protect users against what they don't want to see happen, as much as they can. Adtech could stop that simply by ceasing to work against users.
♥ 5
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
Again, what do you mean by consented vendor? Do you have an example of any vendor getting legally valid consent for RTB? I've never seen that.
♥ 1
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
As a first party? Which purpose are we talking about? A lot of their first party processing I would expect is legitimate interest. What they do in profiling probably requires consent, I doubt that they could get it.
♥ 1
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
Of course the party is key. Talking to someone directly and talking to someone while another person hides outside the room listening through the door have completely different characteristics and must be treated accordingly.
♥ 1
Third parties can only be legitimate when they are trustworthy agents of the first party and cannot reuse data in a separate context.
♥ 5↺ 2
It's possible to get consent for profiling, but it has to render the implications clear for any profiling variable that may be detrimental to the user. Facebook has a pretty bad track record there.
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
Consent for what? You're signing a contract.
Replying to @null and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @johnnyryan, @brave, @random_walker
Again, you keep focusing on consent - why? That's exactly how we're pushing to build nytimes.com. We're not there at all across the board yet but there's nothing technically problematic about it, the only barriers are Google and adtech companies that refuse privacy.
♥ 1
Replying to @null and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
That's not enough information to make an informed decision about the usage of my data. By using privacy preserving conversion measurement and not retaining identifying information for delivery you don't need consent.
♥ 3
Requiring consent is often a sign that you're doing something the user wouldn't want you to, and you have to ask them to be sure.
I say "often" because regs sometimes require consent for nonsensical reasons.
♥ 5↺ 2
Replying to @johnnyryan and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
Also, if you don't show an ad but have still seen the data you're in scope.
♥ 2
Replying to @null and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
I'm not sure how to explain it more times. Johnny gave a lot of detail already. The GDPR bar for consent is high and I've never seen a consent dialog that meets it. This isn't some opinion, just read any DPA guidance. Even IAB Europe admitted as much.
♥ 1
Free content comes from an ad model. There are plenty of good ad models, they don't all have to violate privacy and harm publishers. The energy industry does a lot of good, but it's not being tin foil to say it should move away from oil.
♥ 10↺ 4
Replying to @null and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
That completely depends on what is done with the identifier, and under the GDPR it can never be in exchange for free content, otherwise it's not valid consent. This isn't "adtech bad", it's just the law.
Replying to @null and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
There is no problem, except that only a small percentage of users will consent if it's done according to guidelines. I think the highest I heard of in testing legal dialogs was 14%, most were half that.
♥ 1
So if your ad strategy hinges on gaining legally valid consent, following official guidelines, to what is done in RTB today then it's just dead on arrival.
♥ 1
Replying to @null and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
14% was only achieved once and by a highly resourced publisher. And most of the other tests were closer to 5%.
It's a better bet to try to monetise those other 95%.
Replying to @null and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
It's possible, it's just not a good business option for publishers who stick to the law.
♥ 2
Replying to @null and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker
Well, if anyone in the adtech stack can reuse that identifier and what they learnt from the bid independently from the publisher then that's not helping the publisher stay in business, either.
♥ 2
Replying to @gordonbrander and @michael_nielsen
Indeed! I'm not sure what @michael_nielsen is looking for exactly, but if I think of smart matter at small scale carrying out relatively complex tasks without being controlled by a clunky big computer, that kind of stigmergic-ish approach comes to mind.
Replying to @johnnyryan and @therevoltingx, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker, @seanblanchfield, @Pagefair
What Johnny said. Possible and practical are two different things. Adtech without scale isn't very valuable.
♥ 3
I'd love to know what kind of targeting would give me these two ads next to each other on the very same page.
♥ 8
Replying to @danbarker
Ah, I didn't look, I gave up on that a while ago as never all that helpful?
Replying to @null and @_james_fuller
Lol, I really doubt it's the latter. But the way audience segments are sold it's 100% true that you can be in contradictory segments all the time.
My local bodega cannot procure Marmite anymore.
2020 will stop at nothing.
♥ 5
Replying to @JBSemerdjian
All the guns you can take! Huge "concealed" carry.
Replying to @_alastair
That's what they told me at the shop.
Civilisation is such a brittle thing.
♥ 1
Replying to @null and @prushforth
I might as well — life is tasteless without Marmite for breakfast.
Replying to @shails
I'm willing to bet that it's much more likely that I'm in a nonsensical overlap of segments 😁
Replying to @fatemehx2
They never carried that, it's an Indian/British bodega of sorts, so none of that Aussie nonsense.
I'll be in Princeton soon, I hope they have reserves there.
♥ 1
Replying to @cepcam
Au Times on a un peu mieux: on a un modèle qui nous donne des motivations ou des sentiments probables générés par un article (mais pas personnalisé). Mais sur l'ensemble c'est vrai que le ciblage par topics est très limité.
La motivation économique n'y est pas: les fournisseurs génèrent des segments sans queue ni tête, ils cherchent juste à ce qu'ils soient assez gros. Après, sur un malentendu, ça peut marcher.
Si tu as des idées d'améliorations, surtout qui peuvent marcher dans un meilleur respect de la vie privée, je suis preneur!
Replying to @RobinvanBeauDow and @RobinCRLee
Haha, I'm not sure!
Replying to @duncanwc
A highly prized segment!
Replying to @cepcam
Tu as élégamment deviné le fonctionnement de l’industrie :) La plupart du temps, ce qui est optimisé est effectivement la conversion, donc l’utilisateur est traqué jusqu’à la page “thank you”, auquel moment il passe au contraire sur une liste à qui ne pas montrer la pub.
Replying to @cepcam
Ce qui est compliqué, c’est que pour des raisons de privacy très justes, on ne pourra bientôt plus tracer les utilisateurs. Il y a aussi très peu de recherche sur l’optimisation post-conversion (eg. si la conversion est un abonnement, l’utilisateur reste-t-il abonné longtemps?)
Replying to @hexalys
I know, and it’s not just Google. Every time someone asks to see what BlueKai segments they’re in, especially over time, they think there has been a mistake. It’s extremely common to be in completely nonsensical audience overlaps.
♥ 1
This ONE WEIRD TRICK lawyers don't want you to hear about! If you break the law, but in a specific and limited way, then it's OK!
Eg. for burglary, bring a list (very specific) and don't take everything (limited), then it's good!
Quoting a tweet by @AdamJSchwarz ↗
♥ 10
Replying to @TomicahTD and @MarietjeSchaake
Open Source is almost certainly insufficient for self-sovereignty, I'm afraid. Given the level of complexity this has to operate at we need open protocols with multiple implementations catering to different user types. Self-sovereignty requires usable designs.
Replying to @shoniko and @privaci_way, @iamtrask
The book is a must read.
Also, shameless plug: open.nytimes.com/how-the-new-yo…
Replying to @alextcone and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker, @Google, @amazon
Possibly, though the idea that audiences can only be defined through pervasive tracking has no clear foundation.
♥ 1
Replying to @alextcone and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker, @Google, @amazon
Oh, sure. I expect they'll run into the problem that sooner or later they just won't have the identifiers to match from.
Replying to @alextcone and @therevoltingx, @johnnyryan, @BrendanEich, @samtingleff, @acfou, @kickstand, @mrr619, @WolfieChristl, @johnwilander, @brave, @random_walker, @Google, @amazon
I am keenly aware.
Replying to @TomicahTD and @MarietjeSchaake
I certainly agree that it’s helpful, but I worry about considering it even a central requirement. Don’t get me wrong, I love OS & have practiced it a lot. But at the level of complexity we are talking about, I’m not convinced you can have rigorous oversight of the implementation.
Something like WeChat embeds a complete app ecosystem that’s built on HTML or HTML-like tech. That alone verges on being impossible to audit sufficiently thoroughly and regularly (if you think of something like Chromium for instance).
And Chromium is probably a good case in point: being open source doesn’t prevent Google from using it in ways that remove self-sovereignty.
I think we can get the right properties with better governance and protocols, and open source as a nice to have, no?
Replying to @adambroach
I think there are a few good people, but they’re probably not in PRAM :)
♥ 1
If you're in any way surprised at the intensity ethe West Coast fires, I highly (if sadly) recommend @dwallacewells's "The Uninhabitable Earth".
He describes just what we're seeing. It's only one chapter.
bookshop.org/books/the-unin…
♥ 4↺ 1
I'm seeing a few people saying that Apple's IDFA change is somehow a problem under the GDPR. Often the argument is that it conflicts with the IAB TCF.
I actually think it's the exact opposite: under the GDPR, they have no other option than to make this opt in.
♥ 19↺ 6
(As always: you take legal advice from a random person on Twitter at your own risk. If in doubt, ask a lawyer, or even just anyone with some kind of education.)
♥ 1
Managing the IDFA is not something that Apple does at the user's direction, it is in no way necessary in order to support anything that the user is asking for.
Put differently, Apple determines (on its) the means and purposes of IDFA processing: it is a data controller for that.
♥ 4
Replying to @montezumachavez and @EU_EDPB
Yup, but I don't think they ever intended to discuss the role of user agents in that document, no?
♥ 1
Replying to @montezumachavez and @EU_EDPB
Yes, that's a good one too, but I don't think it covers the relevant issues. Notably, the in-app browser situation would be relevant for monitoring (and could usefully be compared with the Fasion ID case).
♥ 1
Replying to @_alastair
And it's no better on Android.
In-app browsers are a privacy shit show no one is talking about and that regulators seem to completely ignore.
♥ 2
Replying to @montezumachavez and @EU_EDPB
Agreed, but I believe that user agents warrant special care. There are cases in which they act as controllers, but no one is looking at their legal bases. They are particularly deceptive because they are supposed to represent the user, but sometimes don't.
♥ 1
Also, user agents are the gateway through which most behavioural data is captured (in digital services), and there are only a few. Clear guidance on their behaviour would be extremely effective. And experts from @W3C and all are there to help.
♥ 1
Replying to @montezumachavez
Yes, that's a very good example, I agree. I think that the same kind of care should be applied looking into agents that mediate all our interactions but far more invisibly than in that case (browsers, operating systems, identity services...).
♥ 1
Replying to @montezumachavez
I know! But you can still have better German 😁
I love how this also demonstrates just how bad Gmail is at structuring and displaying useful information about others involved in a conversation. It really is optimised for talking at people.
Quoting a tweet by @zachportman ↗
♥ 4
Replying to @WilliamHughes4
It's possible that it may not be enough, that's true. But I like the clarity of the text. Straight and to the point!
♥ 1
Replying to @_RDemol
I see no value in them using the TCF there. They would have to present exactly the same dialog *plus* all the other vendors, defaulting to off and with an easy option to opt out of all of them. If anything you would get fewer opt-ins (which reflects user preference).
The only purpose of the TCF is to convey signals of invalid consent. I’m not sure how it would improve the situation?
Replying to @airavn and @montezumachavez, @johnnyryan, @EU_EDPB, @w3c, @PiracyByDesign, @WolfieChristl, @NOYBeu
Yes, absolutely! The reason I’m thinking about broader guidance is because user agents are technical and have their own specificities — but the NOYB complaint is directly relevant.
♥ 3
Replying to @BuggeErik
Couldn't agree more! This needs to be surfaced because the lobbying that politicians are hearing isn't that, sadly.
♥ 1
Replying to @_RDemol
Apple's sharing of the IDFA to the app's controller is just one purpose. There is no granularity to add there. They are doing consent by the book.
Replying to @jessicamalonso and @Rich_Harris
"Why are so many brilliant well-meaning people working together to destroy democracy?" is a question that has been on my mind a lot this past year or two. With my limited capacities I've been trying to dig into it some.
I agree with your lack-of-stem assessment, but…
…I think the problem might start\ earlier? For instance, Veblen in the 1920 is talking about handing the economy over to engineers in ways that feel very contemporary (even if no one would call it a "Soviet of Technicians" today).
Replying to an earlier tweet of mine
#1 "The Engineers and the Price System", by Thorstein Veblen. Mostly a rant (from 1921) about how industry is being mismanaged (he says "sabotaged") by people who run businesses to make money instead of being run by engineers who would do it optimally, because they know how.♥ 6
♥ 1
Also, reading "Seeing Like A State" and "Red Plenty" made me think that the solution needs more than just putting some sociologists in the room. It would help and we should, but I still don't know how to reach out to these technologists.
I believe we need a much crisper articulation of the problems and of the virtuous ethics of tech that's short enough to draw the attention of well-meaning people and actionable enough that it won't forever be in someone's backlog.
I say all this with no background in anything, just some personal reading and (sic) almost 25 years working in tech.
♥ 1
Replying to @jessicamalonso and @twifkak, @Rich_Harris
It's certainly true that the people who seem to be the best at scaling things up often appear to be the worst at understanding the effects of scale!
Replying to @jessicamalonso and @Rich_Harris
We should certainly refund the humanities, we should also stop treating them as something weird and separate. My experience is that tech-literate humanities people make for much better technologists. Understanding people is hard, everyone can paste from Stack Overflow.
♥ 1
Replying to @jessicamalonso and @Rich_Harris
There's a lot I like in there. (I am less sure of the "do no harm" bit because I think ethics is more complex, but that feels like a side-note.
I have a similar idea (which I recently stole from Popper) which is that to solve problems you want institutions, not leaders.
Most startups start by finding leaders. This is THE person (or few people) who will solve Big Problem. That's a lot like looking for the Philosopher King who will rule over the perfect state because they're such a great person.
In practice, it sucks.
You're much better off designing an institution: something that will work(-ish) even with imperfect leaders, in fact ideally even with really bad ones. Democratic institutions are a good example.
I think your idea of building from the moral culture is similar. I like it!
♥ 1
Replying to @jessicamalonso and @Rich_Harris
I see what you mean but I'm still not sure it's sufficient. Maybe we just disagree on this :) I see what you describe as the virtue of care, which is important, but isn't the only one. (Disclaimer: I like virtue ethics, eg. Shannor Vallor's "Technology and the Virtues".)
♥ 1
Replying to @jessicamalonso and @Rich_Harris
Sadly, I'm not surprised. But conversely, I've also heard a great person in tech, who's been doing a lot of good philantropy, say not long ago "All you need to know about ethics is 'Do no harm'."
I'm like
♥ 1
Yet another quality @temptoetiam thread. You don't need to read much French to read it 😁
Quoting a tweet by @temptoetiam ↗
♥ 3↺ 1
Goodbye Fort Greene, for five years you were the best of homes. 😢❤️
♥ 32
Replying to @macsym
Awwwwwww!
Replying to @null and @amyvdh
Thank you! 🤗
Replying to @null and @arainert
Thanks!
Replying to @gordonbrander
Can you share? I think I'm about to reboot the web, could use some shortcuts.
♥ 1
Replying to @gordonbrander
I'll be curious to see it when you do!
Replying to @1312Sam and @arainert
Great schools, nice town that's actually lively, greenery, lots of nerds, feasible commute, and a few cool people I already know. Move here, that'll make more!
♥ 3
Replying to @tobie and @danbri, @amyvdh
This pressure is insane. I don't know that I can resist.
♥ 3
Replying to @mnot and @amyvdh
Why?
Replying to @RobinvanBeauDow and @RobinCRLee, @amyvdh
Hi! I haven't had time to tell people, but we need to catch up!
Replying to @null and @amyvdh, @tobie, @danbri
So, uhm, no, not changing jobs, you can count on my remaining that cranky guy complaining about the state of the web and democracy 🤗 It's not a huge commute if that's something that has to happen some day, but that day seems far off.
♥ 3
Replying to @mnot and @amyvdh
That's true! But my place here is nice too, you need to come visit when the end of the world takes a break.
♥ 2
I guess this isn't Brooklyn anymore.
♥ 24
I wonder how TikTok users would react to Oracle ownership, knowing that they're basically a data broker.
nyti.ms/3mi56a6
♥ 13↺ 6
Replying to @hackylawyER
I'm old enough to remember a time when they sold database software. You give us old folks time to adapt.
♥ 2
Replying to @null and @benlikestocode
🔥 🔥!
Replying to @alexanderdanilo
People always complain that Oracle's 3P data lacks accuracy, that would of course help!
♥ 1
Replying to @RobinvanBeauDow and @RobinCRLee, @amyvdh
I think this week is still a chaos of boxes, but I'd love to meet up starting next week!
I've been wondering if the Great Filter couldn't be culture. A runaway evolutionary process inherently cut off from the checks & balances of biological evolution. Sort of like a monopoly, but on life.
Eventually everything dies. Or every intelligent enough to have fast culture.
Quoting a tweet by @erikphoel ↗
♥ 3↺ 1
Replying to @erikphoel and @IntuitMachine
One thing that could be the case is that life emerges rarely, however in-system panspermia is common. So once you have it on one planet in a system, with meteorites or whatnot it relatively quickly spreads to other viable planets in the system.
♥ 4
Replying to @annbass
You can come visit in Princeton, it’s not that much of a detour :)
The North Fork is very nice, but it’s also a 3h train ride to work if I need to commute some days. It’s much much more removed. Also the schools aren’t quite so good.
♥ 1
“There is no conflict there.”
twitter.com/matthewstoller…
♥ 5
That feeling when you're totally making a change to meet your users' needs and not at all to leverage dominance, and your designers end up having to add notifications to explain this choice that makes total intuitive sense.
♥ 25
Replying to @slayser8 and @jarroddicker, @Chronotope
That's a good point, but I don't think it's a reason to give up on a promising path. We can push hard on the browser front to force Chrome into being honest. There are a few good policy options there.
♥ 3↺ 1
Replying to @slayser8 and @Chronotope, @jarroddicker
That's why I think that better consumer choice is a nice side effect but is not the right lever to get there.
♥ 2
Replying to @FSGOriginals and @moiragweigel, @logic_magazine
I wish you’d made a bundle instead of having me order all four separately :-)
"Amérique années Trump" de @jcartillier et @Gil_Paris, disponible dès aujourd'hui dans les bacs! Commandé illico, j'ai hâte de le lire!
amazon.fr/gp/aw/d/207285…
♥ 3
Replying to @cydharrell
It's a French perspective but it's as seen covering the White House directly, so I don't know how "outside" you'd rate that. I'm excited because I know one of the authors, and I know this is going to be quality analysis, not sensationalist.
♥ 1
My 13yo cat, who's always lived in an apartment, saw me through the window of the new house this morning and she totally freaked out.
♥ 13
She kept opening her mouth, staying stuck, shaking her head, then looking from another angle.
You could tell her brain was like "What the fuck! What kind of sick magic is this?"
♥ 10
Replying to @ASankin
I don’t know if that was the case for TikTok, but accessing the clipboard is often used as a hack when you’re driving native installs from Web so that you can reopen the app on the same page you had in browser. I wouldn’t be shocked if they just did that.
♥ 1
Replying to @_abialas and @femalegazebot
Not yet, we need to make sure she has the right immunisation to go outside. Soon!
♥ 1
Replying to @publictorsten and @nytimes, @w3c
Anything in particular you would like to know?
Replying to @simonrjones and @nytimes, @w3c
Thanks! Looking forward to helping out with some cool stuff.
♥ 4
Replying to @publictorsten and @nytimes, @w3c
We need to have a voice in how the Web evolves. The Web is in one of the darkest places it has ever been, at the same time some of the most fundamental components of how business works on the Web are being revisited. There is opportunity to build something better.
♥ 48↺ 8
Replying to @ASankin
That’s the use case, yes. I can’t tell what TikTok specifically is using it for. It could be both! What I’m saying is that it’s a common hack and most apps that look at the clipboard probably don’t even know they’re looking, they’re just using an SDK that works that way.
♥ 1
♥ 1
Replying to @manuelmeister and @w3c, @publictorsten, @nytimes
That’s a longer discussion for another day!
♥ 3
Replying to @juanof9 and @manuelmeister, @w3c, @publictorsten, @nytimes
Is there a standard for this?
Replying to @juanof9 and @manuelmeister, @w3c, @publictorsten, @nytimes
It’s just that I don’t know how native works all that well :)
♥ 1
Replying to @tilgovi and @ImageSnippets, @publictorsten, @nytimes, @w3c
Absolutely. Publishers have a history of not showing up and then complaining that things don’t work for them. I’ve been on the non-publisher side complaining about this; I hope that more others will join too!
♥ 2
Victor Hugo is with us tonight.
"Si l’on n’est plus que mille, eh bien, j’en suis! Si même
Ils ne sont plus que cent, je brave encor Sylla;
S’il en demeure dix, je serai le dixième;
Et s’il n’en reste qu’un, je serai celui-là!"
- Ultima Verba
♥ 6↺ 1
Replying to @realmaplesyrup
I wish it translated better.
Replying to @DPWF0
A lot of countries just change their constitution when it becomes clear that it needs an update (often in the wake of crisis). The American cultish devotion to this old thing is puzzling to most abroad.
Replying to @realmaplesyrup
I think the high school French probably comes out better!
Replying to @DPWF0
True, and it's not like there's a spirit of consensus blowing across the land.
♥ 1
Replying to @realmaplesyrup
I don't think it's bragging so much as somewhat archaic? "I braved the elements to come to you!" It's somewhat archaic in French too.
Replying to @catthekin
No, I think it's an older spelling of Scylla en.m.wikipedia.org/wiki/Scylla
Or it might be a typo. I believe he was on a boat when he wrote this, so it works even better. On the Internet I see both spellings in the poem, but I can't get to my paper copy, it's in a box somewhere.
Replying to @odinho and @nytimes, @w3c
I'm hoping that I can help my colleagues get into it and won't have to do too much myself 😉 In theory, I'm retired!
Replying to @jesslynnrose
There's a moment when I started preparing my presentations ahead of time instead of the night before and... it's nice actually? I'm still a little bit in shock.
♥ 1
Replying to @jesslynnrose
I think it's like the people who kept saying all along that exercise is good. It's painful to admit they were right.
♥ 1
Replying to @catthekin
Yes! Well, he was in exile as an opponent to Napoléon III who had overthrown democracy, so you could say he was up against some pretty serious opponents!
I dug some more (it's been bothering me!) and it looks like your reading with Sulla is the right one. I found a BNF version which I trust with Sylla not Scylla, and that's a better source. Cool reference, thanks!
♥ 1
Replying to @catthekin
Oh, you went straight to a credible source instead of poking around French literary criticism homepages from 1997?
♥ 1
Replying to @catthekin
I think I kind of miss ~ the Internet ~ from back when you could get lost following links between ugly but personal sites on a random topic, so on the rare occasions it still happens I get pulled in 😄
Yes, I learnt it the first time Le Pen made it to second place. It stuck.
Replying to @MichelleLynneB and @clancynewyork
I've been wondering if someone could put together a consultancy of ethical pen-testing of sorts. It would be so useful.
♥ 1↺ 1
Replying to @clancynewyork and @MichelleLynneB
Thank you 🤗 To the extent I do, it's only thanks to reading smart people like you!
♥ 1
Replying to @gsnedders
Why would you expect TERFs to ever be happy so long as trans people exist?
♥ 3
Replying to @mariafarrell
Wow! A few of us have been thinking about something very similar for the Web, with (perhaps unsurprisingly) pretty much the same grounding in Seeing Like a State.
♥ 2
Is it possible for us peasants outside universities to buy paper copies of Synthese, or is it a SciHub-only journal? I live in Princeton, in case that opens up a possibility.
♥ 1
Replying to @mariafarrell and @kkomaitis
If she’s up for sharing I would definitely be interested!
♥ 1
Replying to @danbri and @netflix
Yes, equally frustrating but for different reasons. There might be a typology to tease out when it comes to loss of agency.
Replying to @dialtone_ and @johnwilander
It means that the spend shifted when there were other options, not that it's needed for CPMs.
♥ 1
Real talk from always-amazing NYT boss of ad innovation Allison Murphy.
Quoting a tweet by @johnwilander ↗
♥ 15↺ 1
Replying to @dialtone_ and @johnwilander
It's not a thesis, it's a description of the inferences we can draw from the data.
Reminder that “Privacy is Power”, by @carissa, it out tomorrow! penguin.co.uk/books/1120394/…
♥ 5
“Story ate the world. I’m biting back.” An excellent and excellently-written piece from @mariafarrell. I’m quoting just one paragraph that describes a sentiment I’ve often had, but there’s much more to read in the whole thing:
crookedtimber.org/2020/09/23/sto…
♥ 11↺ 3
And no, I haven’t figured out my life’s work. Or at least, I can’t seem to bring it down to just the one. I’ll just try to do a bit more than homeopathic solutionism in the meantime!
♥ 4
Replying to @hober and @marcosc
Ask it! I… am very rusty so Marcos is probably a better bet, but I can try!
♥ 1
Replying to @hober and @marcosc
So basically you want to have the refs at the end but none of all that unsightly [ALL-CAPS-CRAP]? Perhaps the simplest option is to write this as if it were meant to be daunting, with the refs where you’d include them, but then add `.bibref { display: none; }` to your style?
♥ 2
Crap, actually that won’t work, it doesn’t catch the []. I guess something similar to what you did might be the simplest:
<p style=“display:none”>[[CRAM]], [[IT]],[[ALL]],[[HERE]]</p>.
♥ 1
Replying to @jyasskin and @hober, @marcosc
I think that’s a little different? Maybe I misunderstood the issue but that looks like a daunting list of dfns, whereas I think the question is to have refs listed in the refs but without the ugly [REF] bits.
♥ 1
Replying to @jyasskin and @hober, @marcosc
I don’t know, all that newfangled stuff was added after my time — you’ll have to ask Marcos! That said, doing a localXref should be trivial if there’s a need for that.
♥ 2
Why can't more tech orgs be like @mozilla? Technology is supposed to make the world better, and this thread does. If you work for a tech company with many times Mozilla's resources, are you making the world better, or are you just making money from technology?
Quoting a tweet by @mozilla ↗
♥ 37↺ 8
J'ai soutenu @unboutdumonde_, un projet de @CageJulia pour que nous, citoyens, nous organisions pour développer l'indépendance des médias français.
Rejoignez-nous!
kisskissbankbank.com/fr/projects/un…
♥ 9↺ 6
Replying to @null and @nitot
Je me pose des questions similaires. Il y a encore quelques trucs que je veux faire évoluer en privacy et pour la bonne santé du Web, mais après? Comme toi, je pense à l’écologie mais je me méfie de la tendance des codeurs comme moi à penser être utile partout!
Replying to @null and @nitot
C’est ma façon de me méfier de mes élans :) Mais je pense que nous nous posons des questions similaires. Si tu trouves des réponses, ça m’intéresse beaucoup!
Replying to @ThainBBdL and @nitot
Je ne suis pas d’accord. Je pense que nous avons la capacité de rendre le monde meilleur, mais que c’est devenu difficile parce que la technologie s’est égarée.
Ce dont je me méfie, c’est cette sorte d’hégémonie techno-solutioniste qui fait que certains qui ont résolu quelques problèmes techniques complexes pensent pouvoir tout résoudre.
♥ 1
Quand je suis passé de la tech à la gouvernance des données (pourtant pas un domaine très lointain), je me suis ramassé des mois à ne rien comprendre et à devoir tout lire et apprendre. Au final, mes connaissances tech aident beaucoup, mais j’ai dû apprendre un autre métier.
Je pense qu’il est possible de faire la même chose pour l’écologie, avec la même nécessaire humilité — ce qui est difficile c’est de trouver la porte d’entrée par où se rendre utile le temps de devenir compétent. C’est pas évident du tout!
Replying to @null and @nitot
Oui, mais le problème qui va nous tomber sur la gueule c’est que ce qui est efficace, c’est de centraliser… Si sauver le monde nécessite de tout donner aux plateformes, on n’est pas rendus. nytimes.com/2020/02/27/tec…
Replying to @ceciliakang and @Facebook
This is insane? WTF. They can’t handle adults safely and they expect to do well with kids?
Replying to @null and @nhoizey, @nitot
Je sais bien, mais on voit bien comment ça glisse de l’un à l’autre. Tu commences par te mettre sur une infra commune. Et puis il y a cette API qui est juste là devant toi qui t’aide un peu. Et puis…
Ce n’est pas perdu d’avance, mais ça complique la chose.
Replying to @null and @nhoizey, @nitot
Oui oui, je crois que nous sommes en fait d’accord :)
Ever heard about net guns? I want a net gun. We should all have net guns. netgun.com/netgun-info/ul…
Quoting a tweet by @tomwarren ↗
♥ 5
Replying to @rigow
Right — but for this one we need indoor options!
Replying to @cwilso
I mean, it *is* tempting.
Worst-case scenario no drone army shows up and I can have fun with my cat.
♥ 1
Replying to @BotJunkie and @rcalo
I’m not sure that’s the best debate to be had here? It’s a bit of the “would you rather die in a fire or a vat of acid” variety.
♥ 1
Replying to @jamesvasile and @nabihasyed, @Dymaxion
Interesting! I’d love to hear more. I agree that ethical certification wouldn’t mean much.
The reason I’m thinking about pen-testing is because I’m interested in the “we’re trying to break this” mindset. “We tried to use this for evil, and here are the five ways we found.”
The people I work with tend to be quite ethical, so I sort of want to bring in a criminal mastermind who can see how our tools could be used for evil — so that we can then plug holes in the process.
I try to be that criminal mastermind (my team sometimes introduces itself as “we think about adversarial dystopias so you don’t have to”) but I know that someone who does that professionally would do it much better!
Replying to @jamesvasile and @nabihasyed, @Dymaxion
That’s the idea. Trying to think of issues not when the product is used by your team of adorable kittens but instead a few years later when everything’s gone wrong and you’re being sold for parts in an LBO raid.
♥ 1
Replying to @Dymaxion and @jamesvasile, @nabihasyed
I’d love to chat, my DMs are open :)
Replying to @null and @SimonDeDeo
Holland on Complexity, too.
Any recommendations for phone keyboards (given a Note 9)? There seems to be very little, I'm almost tempted to get a Blackberry...
♥ 1
Replying to @null and @icopilots
Hardware. The software one I have is decent.
Replying to @DoraCrisan and @mozilla
This is interesting, but I’m not sure what you have in mind from the NYT side?
Replying to @null and @icopilots
I’ve seen that one, it looks great but you can’t hold the phone with it, right? You need something separate.
Replying to @DoraCrisan and @mozilla
Oh, I get it — sorry, I thought you were suggesting that we contribute content, which I didn’t see how we’d have!
I’ll let @kevinroose decide if that’s his thing or not.
♥ 1
Replying to @null and @icopilots
Ooooh, will check that thanks — hadn’t seen it.
Replying to @null and @icopilots
I was *just* about to tell you that I went with that one. Thanks a lot!
Replying to @erikphoel
Is that human nature or is it locally cultural? If you don’t feel qualified in aesthetic matters because you’ve always been told that was someone else’s talent, you’re likely to find originality stressful. You’ll want a standard house with West Elm & the same fake Chinese vase.
♥ 1
Replying to @hare_brain
Note, though, that the coronavirus contacts of my coronavirus contacts are my friends.
♥ 2↺ 1
Replying to @arnaudpc_ and @arghnoo, @w3c
À noter cependant que ces deux positions ne sont pas coordonnées, ce sont des parties complètement séparées de la maison.
♥ 1
Replying to @arnaudpc_ and @arghnoo, @w3c
Ce n’est pas quelque chose qui se produit d’un bloc. Différentes parties du Times ont des intérêts divers, depuis des trucs assez techniques dans CSS jusqu’à des éléments plus stratégiques autour de la pub et de la privacy.
Replying to @arnaudpc_ and @arghnoo, @w3c
Merci!
A little peace for your evening.
♥ 11
Quality thread about the choices in front of us to govern tech.
Quoting a tweet by @glichfield ↗
♥ 8↺ 2
WTF is wrong with you, Twitter? That's the exact opposite of the direction I want to take this.
♥ 4