Replying to @histoftech
I actually got to use one a couple of months ago. Honestly, it took me a couple tries to remember how to actually do it. We had *time* back then.
♥ 2
April 2018
152 tweets
Replying to @TheAnnaGat
Mine too 😚 But sometimes I wonder about discovery because, you know, it's not like I already have too much to read or anything.
♥ 1
Replying to @null and @amyvdh
I'll give it a shot, thanks!
Replying to @OriPekelman
I think all I read there was the latest Voynich sensation. Is the rest better?
I reckon I'm officially retiring from politics and just letting @Noahpinion state what I think only a lot better.
Quoting a tweet by @Noahpinion ↗
♥ 1
TIL that teaching French was forbidden in Louisiana throughout much of the 20th century so as to stamp it out as the then-dominant language there.
♥ 3↺ 4
Replying to @AmeliasBrain
That was done in Australia too (and the language part was done throughout France as well).
Replying to @AmeliasBrain
Oh yeah, not just for language either. That said the big language killer was TV. Even accents have disappeared from entire areas.
Replying to @AmeliasBrain and @gsnedders
At least the BBC makes use of regional accents nowadays, in France they're still ridiculed.
♥ 1
Replying to @_abialas and @femalegazebot
Perfect name, too! Want to see!
Replying to @null and @liedra
I thought paragraph 3 of mobile.nytimes.com/2017/07/20/mov… pretty much gave the final word on it.
Replying to @coolharsh55 and @RubenVerborgh
Can you send me a screenshot? The contrast comes out fine in testing.
Replying to @coolharsh55 and @RubenVerborgh
You're right, the fonts seem to come out very thin. I suspect it's because resolution has increased a lot since I made it. It's overdue for a redesign anyway!
Disappointing, yes, but true.
Quoting a tweet by @EFF ↗
@ManyaJeanGitter Hi, I think I have your wallet, did you lose it on South Oxford St by any chance? My DMs are open, LMK the easiest way to get it back to you!
Replying to @Noahpinion
Funny how much that sounds exactly like the French point of view ask well. I wonder if it's driven by some commonalities.
♥ 1
Replying to @CNIL
Est-ce que vous avez une version accessible de l'article?
This is really cool: the @w3c is using an open GitHub board to share all that’s on its radar for upcoming work on the Web platform. github.com/w3c/strategy/p…
♥ 63↺ 37
Interactivity is coming to scholarly articles in the same way that 3D interfaces are coming to the desktop: predicted every few years, at best a niche.
Right now scientific papers can barely get text right. Let's start there.
Quoting a tweet by @skdh ↗
♥ 10↺ 1
Replying to @docum3nt and @jsomers
Yes, it basically ignores five decades of document engineering, it violates the principle of least power, makes what's wrong with LaTeX probably worse... And that's before getting into the details.
↺ 1
Replying to @TheAnnaGat
I'm glad if there are lines in London though. I assume they vote with more cosmopolitanism than the average Hungarian. 🇭🇺
Facebook to apply the #GDPR globally - except in the handling of particularly sensitive personal data for which it will keep treating non-Europeans like data cattle. 🤔
Quoting a tweet by @profcarroll ↗
♥ 1↺ 8
Replying to @null and @je5perl
Hah, that's a fun detail, I hadn't realised they had made that call. I suspect they'll change their ToS in short order, but I wouldn't be shocked if they were too chaotic to act on it.
Replying to @null and @je5perl
It absolutely is. I've been worried that the Irish DPA would be lax to dovetail nicely with their tax haven policies. I guess we'll see pretty soon.
Web standards: the early days.
♥ 14↺ 2
Replying to @LeonieWatson
In fairness though, "smiling pile of poo huffing with angry face" is a pretty accurate description of many Twitter users.
♥ 4
Americans are shocked that China is considering its own credit score system.
Quoting a tweet by @seanmcarroll ↗
♥ 8↺ 3
Replying to @mholzschlag and @stevefaulkner
I'll let you know if I see anything here that would work today (I'm at the Whitney) 😉
♥ 1
Are there any reliable Hungarian results yet (in languages I understand)?
♥ 1
Replying to @ppk
Things still uncertain but there's a bit more information in other responses.
Replying to @TheAnnaGat and @wvessna, @martapardavi
Depressing 😑
Replying to @TheAnnaGat and @eurocrat, @nickreynoldsatw, @raphdamico
Thanks for the pointer. I agree with the analysis, but I still find the result depressing on cosmopolitan grounds. It won't damage the EU much, but it will keep damaging Hungary.
♥ 2
La Hongrie reste Orbán de l'Europe.
#pardon
#HumourTriste
♥ 2
“We are Missing the Point about Facebook, Cambridge Analytica, etc.” by @baekdal baekdal.com/thoughts/we-ar…
Validation is wishing things into existence. One day you grow up and learn that doesn't work.
Quoting a tweet by @dauwhe ↗
♥ 3↺ 2
Replying to @noteon and @dauwhe
Exactly, and the gap between valid and functioning is the exact size of what you're wishing into existence!
♥ 1
Replying to @profcarroll
Taxation would be difficult and I’m not convinced it would not have adverse effects. Right off the bat, what degree of pseudonymisation would flip data from personal to non-personal? Under the GDPR you can make the case that it’s sufficient for specific threats to…
…the “rights and freedoms of data subjects.” For taxation, you either owe money or you don’t. Budgets would be cut for InfoSec-required data logging before they got cut for marketing retargeting, etc.
This also opens the door to data patrimoniality, which is a whole other hell!
In which @zeynep basically calls for an American #GDPR in all but name.
I disagree with the part on opt-in/consent but other than that it’s gold.
Quoting a tweet by @nytopinion ↗
♥ 1
Replying to @profcarroll
Heh, indeed, but my problem is much more with feasibility than taxation — I’m going through that for the GDPR.
To deliver the newspaper you’re paying for to your door, I need your address but that’s commercial. Is that included?
Replying to @profcarroll
Data is almost never kept for a single purpose. Is it per purpose? Is site-optimisation commercial? Is duplicated data taxed multiple times or once? What if I’m duplicating it to protect some PII (but both instances are use commercially — this is common)?
Replying to @profcarroll
I’ll stop before I spam your whole evening with questions, but it’s a long list :)
Replying to @profcarroll
Pseudonymous data is personal data under the GDPR, no disagreement there. HIPAA-level de-id would definitely not make my bar for anonymous data; the classic AOL or Netflix deanonymisation cases were both HIPAA-grade but that didn’t cut it.
Replying to @baekdal
Actually, that’s the only part on which I disagree. It’s easy to come up with plenty of cases in which information about you can and should be shared, from reporting your location to save your life to passing your address to USPS so I can mail you.
Replying to @baekdal
I think the summary in “Bit by Bit” is really good (though it’s worth reading the full argument from Nissenbaum): bitbybitbook.com/en/1st-ed/ethi…
Replying to @baekdal and @ezeq
If you use “hard to understand methods” then it’s not valid consent (see the WP29 guidelines), so that’s already off the table.
But consent is the wrong side to look at this, consent needs to remain rare. The question is what the sites will try to pass as legitimate interest.
Replying to @profcarroll and @dbarthjones
Oh I’m fine with HIPAA de-id for science research, I just don’t think it’s enough to de-id at a GDPR-compliant level. Most website data sets have a lot more than the Netflix or AOL controlled releases.
Friendly reminder: when you download your @Facebook data, it does *not* contain the tracking data they have from following you around the Web. They know a lot more. #Zuckerberg
Quoting a tweet by @karissabe ↗
♥ 7↺ 5
Replying to @mikewest
I don't believe that informed consent is possible at anywhere near the current scale of information collection and processing. I also think that consent should be restricted to cases that really warrant it, with less problematic cases handled by real transparency.
(This keeping in mind that transparency is also hard at the current scale.) At any rate, excessive reliance on consent is toxic to privacy.
Replying to @mikewest
I'd be curious to hear what details you disagree with. But yeah status quo delenda est.
Replying to @johnwilander and @mikewest
Right, and other companies are made — often against their will — to embed 3P tracking or face sometimes severe loss of revenue. That’s why I believe that tracking can only be stopped at the source.
Replying to @johnwilander and @mikewest
I’ve often heard that given how much tracking publishers do, they must want it. That’s not the sense I get from talking to colleagues in the industry — it’s a forced decision, and not a sustainable one.
♥ 2↺ 1
Okay, I think we don't have a better option than to launch a GoFundMe campaign to fix the C line. Anyone up for throwing a few dollars in? Maybe @BilldeBlasio? Or @NYGovCuomo?
♥ 1
Replying to @patrickkettner and @montogeek, @domenic, @thedomstandard
The goal of the initial fork was to explore ways of automatic sync that would bring WHATWG specs under IPP protection if possible without substantial change, to try to find a way to peace. I'm really not sure why changes would ever become justified in this spec.
My 4yo wants me to ask the Internet why pillows are so soft.
So, uhm, there you go Internet!
♥ 3
Replying to @ldodds and @danbri, @Gkellogg
I say examples and not recommendations because it's a bit unwieldy. There's room for improvement I would think.
♥ 1
Replying to @jkbren and @erikphoel
When you say roll out, you mean the talk or work? It sounds quite interesting :-)
♥ 1
Fashion tip: my 4yo prefers me when wearing the Uniqlo black "Ultra Light Down Jacket" because with it I "look almost like Black Panther".
uniqlo.com/us/en/product/…
♥ 5
Replying to @ChristosEllina1
I think @UniqloUSA should do a Wakanda edition.
You can get robots and chemistry sets all you want, nothing beats the original STEM toy: discarded cardboard packaging. 🤖 ⚗
♥ 10
Great interview of and book selection by @timcrane102 from @five_books; you might find out that you're a lot more interested in metaphysics than you thought.
Quoting a tweet by @five_books ↗
♥ 2↺ 1
I wonder how far to take similarities between the financial crisis of 2007 and today's privacy moment. An unsustainably complex setup that few if anyone understand in full, with strong interconnections and budding public perception.
The question is: Is there a tipping point?
♥ 7↺ 1
Replying to @jensimmons
If your timeline isn't full of people discussing Beyoncé all the time you're missing out ☺️
♥ 1
Replying to @jkbren and @erikphoel
Sweet, I'll keep an eye out for the announcement. Any of those conferences in the New York area?
Replying to @ReinH and @JaimieMurdock
That's actually almost exactly what governance is 😊
I really love this brutalist Lego (or is it Lego brutalism?). instagram.com/brutsinlego/ (via @kottke)
Quoting a tweet by @kottke ↗
♥ 1
When you use Facebook Login, it by default shares a ton of information about the user (like gender, age range) that you really have no business knowing.
Is there a way (short of disabling it) to get them to share less?
developers.facebook.com/docs/facebook-…
Replying to @gchampeau and @Senficon
I would say that you can exclude ad blocking users *if* you provide an option for ad serving that does not involve personal data processing (beyond the very basics like IP).
♥ 1
Replying to @null and @borisschapira, @slightlylate, @HenriHelvetica
All I’m hearing from this tweet is that you’ve never invited me over for a BBQ in your garden and that’s just sad.
♥ 1
Replying to @null and @borisschapira, @slightlylate, @HenriHelvetica
Wonderful, I've always wondered what the proper wine paring for squirrel would be.
Overall I’m a huge fan of the #GDPR but whoever had the idea to make it ship on a Friday needs a good talking to. (Cc @CNIL @ICOnews)
Quoting a tweet by @femalegazebot ↗
Replying to @null and @je5perl
I know, I get the mechanics, I still think no one should ever ship on a Friday 😁
Replying to @null and @je5perl
I know, I realise that — mostly I was joking, but if we could avoid doing that for the ePrivacy grace period it would be sweet.
Replying to @_abialas and @femalegazebot, @CNIL, @ICOnews, @kelseyj1123
OMG Kelsey is on the Twitters!!!
Yeah, we need to figure out some kind of party, I’m not sure what. Ideas welcome!
Replying to @_abialas and @femalegazebot, @CNIL, @ICOnews, @kelseyj1123
Thank you but let’s ship then call it a good job ;-)
You know what would be a very smart move right now? Buying an aging service with millions of users and turning it into a privacy-friendly social network, bypassing the bootstrap problem.
It looks to me like someone might be pretty smart at @SmugMug.
Quoting a tweet by @kottke ↗
♥ 47↺ 19
I mean, Flickr was seriously big. When they decided to add support for video it was the most controversial thing on the Internet that week. If they brought it back properly I'd start paying again, and I'm not even into photography.
♥ 4↺ 1
Replying to @null and @hollowaynz, @SmugMug
Yeah. Believe it or not but people actually used to pay for Flickr, and all it did was host your pics!
♥ 3
Replying to @null and @vlentz, @SmugMug
Putain mais tellement.
Replying to @lukOlejnik
Hahaha 😂
C'est une très bonne question. L'opération est purement sur la communication puisqu'elle ne peut servir à rien d'autre. Donc il y a une visée ultérieure. Quelle est-elle, pour qui? Mon tout est un plat traditionnel québécois.
Quoting a tweet by @versac ↗
♥ 5↺ 1
Replying to @__apf__ and @facebook
Yeah.
Replying to @rsip22
Yes, it's in the PDF of the Pixel docs, close to the start of the FAQ.
Replying to @null and @je5perl, @facebook
It can't, and the docs go on to explain it doesn't. They're using the old sense of "personal". That's not great, but less scary.
Replying to @sandhawke
Imagine GoT but you're one of the characters...
Replying to @S_Anderson1974 and @paulmwatson
Default-checked is only an issue if the legal basis is consent. This is looks like selectively opting out of legitimate interest non-essentials.
Replying to @muira_mccammon and @femalegazebot
Hi Muira! I don’t believe there is such a source, or if there is I haven’t seen it. I think a lot of the GDPR-related thinking is changing faster than anyone can write about it (certainly people involved enough to know what’s going on don’t have the time to write).
Replying to @muira_mccammon and @femalegazebot
I definitely think there will be some interesting post-morteming to do!
Replying to @johnallsopp
Is that and in the sense of grappling with both?
Replying to @webdirections and @johnallsopp
I'm not sure I follow?
Replying to @webdirections and @johnallsopp
No, that I know, but it doesn't answer the question I had for John 😊
Replying to @webdirections and @johnallsopp
As in: I've grappled (a lot) with GDPR, but I have yet to spend quality time with Australian privacy (though that could be in my future). I was just wondering if I could be of service.
♥ 1
Replying to @NealPoole and @facebook
Thanks, that's reassuring!
My daughters are auditioning for their school's talent show, so they put together a dance routine about women's rights to Beyoncé's "Run The World (Girls)". ❤️ #SJWFTW
♥ 16↺ 1
Replying to @annapieti
I couldn't video the audition, I hope they get selected so I can video the show!
I’m curious: has anyone attempted to set up some form of IRB-like structure in a corporate environment?
Replying to @WKCosmo and @mcnees
That’s a wonderful model Will, I think we could also fund journalism that way.
♥ 1
Replying to @deaneckles
I wouldn’t necessarily measure engagement as time spent; it seems from this article that they are optimising for interaction (share, comment, react) which is what I would expect alongside likelihood of seeing or clicking on an ad.
Replying to @deaneckles
The problem I’m hinting at with that kind of optimisation is that it really does feed into explosive models. If angry people are likely to share something that angers them (maybe for specific types of anger) then they are spreading content likely to anger others.
Replying to @deaneckles
To make what might be a very long discussion short, I would tend to say that it is incumbent upon this type of media to act on and limit this sort of epidemic when they have a reasonable possibility of guessing at it.
Replying to @deaneckles
I would in no way claim that this is easy — just that if your primary concern is spreading and you’re good at your job stuff’s gonna spread, and it might be a good idea to be careful about what it is that’s spreading.
Replying to @MichelleNMeyer and @deaneckles
Thanks to you both, these are super helpful. I will take them back internally and we’re going to spend some time thinking about how it could apply.
♥ 1
France, famous worldwide for its "Catmembert".
♥ 2
OH: “I would sooner rather do taxidermy than get involved in taxonomy.”
♥ 3↺ 1
Replying to @ElliotLepers
What do you mean by good analysis? It seems pretty straightforward, no? That said I wonder if there is a privacy shield aspect, I haven't been following that.
Replying to @ElliotLepers
This is not an aspect that I have looked at much because it's not a problem we have, but my reading is that this has no effect on EEA residents (and in fact they're not moving those). EEA residents have their GDPR rights irrespective of storage location.
What it does change is the situation of all the others. Data jurisdictionally in the EEA is under the GDPR independently of user residence. So *those* users' rights are being curtailed. Paging an actual lawyer to check: @rubin.
Replying to @rubin and @ElliotLepers
According to the article it's not even US users, it's everyone outside EEA, US, and Canada.
♥ 1
Replying to @rubin and @ElliotLepers
Oh yeah, the risk strategy makes no sense to me. It's like they're shouting "WE'VE BEEN DOING BAD STUFF" as loudly as possible.
I've always wondered what the whole set would cost.
♥ 7
The economics of journalism are more brutal than you know, and most of your comparison price points are heavily VC-subsidised. A wickedly good thread with the details 👇
Quoting a tweet by @asymmetricinfo ↗
♥ 3
Replying to @DrTormentas and @LogicalAnalysis
No offence but Europe is way, way more complicated than that. France alone has every regional option between 1 and 4 kisses, starting on either left or right, and including men or not.
Replying to @macsym
That does not sound hard to explain to users, and so falls under easily "above board" (we have some cases too). I would also dispute that interpretation. Joining a Meetup = joining a chat channel. It's a completely different purpose from selling addresses.
I dissent to a point with his view on dilettantism: I find the mind's attention to benefit from fallow periods.
But that would be be my only gripe with an otherwise great text.
firstthings.com/article/2018/0…
♥ 3
Replying to @ElliotLepers and @rubin
It's basic: those non-EEA people would have GDPR rights if they were jurisdictionally in the EEA. Moving them removes those rights.
Replying to @ElliotLepers and @rubin
US users were not in Ireland, other countries were (check the article). I would guess tax or regulatory reasons. Quite possibly taxes.
♥ 1
What would be your reading recommendation on prison abolitionism?
♥ 1
Replying to @0gust1
3 hours is hard, especially in a row. Call it aspirational! If you you give up TV it's usually almost enough 😉
♥ 1
Replying to @null and @amyvdh
This looks interesting, I'd love to hear what you think when you've read it. (Also, secret admirers should really send me more books.)
Replying to @fborgesius
Well, if we signed a contract in which I ask for advertising it might be 😁
♥ 1
Replying to @fborgesius
Doesn't that just mean that they expect third parties providing them with data to rely on consent?
♥ 1
Replying to @deaneckles
But again, this looks mostly like using human input to optimise engagement instead of behavioural input; but it does not reach into the broader topic of what the system-wide optimisation should be based on its effects, no?
Replying to @null and @tara_aaron
Would you mind pointing at which part of Article 6 informs your very idiosyncratic reading?
Replying to @null and @tara_aaron
I am well aware of the requirements of consent *when* it is the chosen legal basis. I would further refer you to the WP29 opinion on consent to get a more detailed sense of the expected bar for compliance.
Replying to @null and @tara_aaron
Subsections (d) and (e) are indeed rare in my experience, however it is probably worth pointing out that (f) — Legitimate Interest — is by very far the most common legal basis under the GDPR.
Replying to @null and @tara_aaron
I am not addressing you as a customer but as a fellow twitterer whose opinion is — and I say this with respect — very much atypical.
Consent is the legal basis to use when it is vital to bring the problem to the user’s unambiguous knowledge. Eg. when asking for sensitive data.
Replying to @null and @tara_aaron, @nytimes
Relying excessively on consent is bad for privacy as it leads to consent fatigue (also in WP29 on consent). That is emphatically not what the GDPR is about.
Replying to @null and @tara_aaron
People have been working on this for the past two years and planning for it since earlier drafts. A lot of us are talking together on a very regular, almost daily basis. I think it’s pretty clear by now.
Replying to @null and @tara_aaron
No, control is through user rights — not consent. Consent does not obviate user rights, and user rights exist in the absence of consent.
Replying to @null and @tara_aaron, @nytimes
It’s enforced in the US through your selected supervisory authority, same as in the EEA. Good luck!