Replying to @cynddl
It looks great! And thanks, that's super kind, but indeed even in normal times things get through barely half the time!
♥ 1
May 2020
401 tweets
Replying to @laurendw
Nice but I want something where I give someone money and they give me a mask 🙂
Replying to @svgeesus
Thanks, it looks like it would fit! I'd rather avoid a valve model though because they protect the wearer from others but not vice versa (is my understanding). Then again, I have a mold allergy so it's tempting!
♥ 1
Replying to @temptoetiam
La surmortalité US est estimée au-dessus de 120k, donc je pense qu'il n'est pas déraisonnable de doubler ce chiffre.
♥ 1
Replying to @temptoetiam
Il semble y avoir un gros écart de cas entre régions en France (j'aimerais être plus précis mais les données sont un peu pourries). Ça pourrait expliquer?
Replying to @Trinart
That would be great but I think it would take forever, and we don’t know where we’ll be staying on any given week. But they look great so maybe when things quieten down!
♥ 1
Happy Labour Day everyone!
♥ 13↺ 1
Replying to @aljopainter
Indeed!
Quality journalism doesn't just write itself, I'm proud to support The Guardian and really happy to see that they are growing and in decent shape to weather the storm.
Quoting a tweet by @pilhofer ↗
♥ 6↺ 1
Replying to @lvedkhanna and @lvedd
Gauging how sober he is to assess how many margaritas he needs.
Replying to @null and @KendraWrites, @erichanauer, @pfairley
♥ 2
Replying to @jathansadowski
Aha, this additional background makes a lot of sense. It was hard to put my finger on it but that passage "felt weird". The words seemed to be heading towards "just nationalise data" but that went against the grain of the rest. The extra context helps, and that quote's great!
♥ 1
Replying to @jathansadowski
I don't have a fully-baked solution there either, but I've been thinking that part of the solution might be to make this behavioural data rivalrous (by law obviously, that can't be done technically). It would introduce friction to the scale of the big data hoarders.
♥ 1
Personal data should always be owned by the data subject, but we can influence who is a data controller of it (in the GDPR "determines the means and purposes" sense).
Today you can have multiple controllers of the same data, and the cost of sharing is close to zero.
♥ 1
But the value extracted from the data is not intrinsic to it: it benefits from network effects with other data a given controller has. At the point of collection your behaviour might be shared with a dozen controllers, but the biggest reap the most value — in a vicious cycle.
♥ 1
Limiting controllership would heavily change the economics of data sharing, and would cause stringent limits to scale for major collectors — not a solution on its own, but I think it could be part of one.
♥ 1
This looks like an amazing opportunity, I wish I could apply!
If agriculture solves unfair power asymmetries and data practices between small producers and mega-intermediaries before we do I’m going to be pissed but that shouldn’t stop them. 🔥
Quoting a tweet by @drkirsteen ↗
♥ 5↺ 2
I’ve been asked to give a course on the technical history of the Web, which sounds exciting.
I’m looking for material so I don’t have to do it all from scratch — anything you fine people would recommend?
♥ 74↺ 14
Replying to @masooga
It’s invariably going to be that to some degree, but it would still help to have some existing material on which to weave.
Otherwise, it’s going to be all memes.
w3cmemes.tumblr.com/post/127643099…
♥ 1
Replying to @masooga
Some are easy to understand from the outside: w3cmemes.tumblr.com/post/822197405…
♥ 1↺ 1
Replying to @masooga
Some feel familiar: w3cmemes.tumblr.com/post/821375867…
Replying to @danbri
Oooh lots of good material on the CG’s blog, thanks! I remember you project to run WWW! That sounds like super cool stuff.
♥ 2
Replying to @fantasai
Cool, thanks! Yes, I think one lesson on layout could absolutely be good!
Replying to @scottbw
Yeah, I thought of that, but I have no idea where my copy is!
Replying to @kev097
Ha! That’s an interesting angle :)
Replying to @null and @amyvdh
Yes, definitely that!
Open question: is relying on dark patterns an admission that you know that what you're doing is wrong?
♥ 19↺ 3
Replying to @snowjake and @Chronotope
Haha, thanks! The initial idea was internal to Data&Insights at The Times, but I don't think that it would have any confidential information 🤭
That said, it's not a done thing, another option might be an Intro to Data Politics in tandem with @masooga.
Will keep you posted! 🤗
♥ 5
Replying to @willikampmann
Ah, but that’s my point though! If you think those are unequivocally good you select them by default with a neat presentation and a friendly sentence, but let people just uncheck them.
Here this is full on dark pattern — it’s what you do when you *know* you’re screwing people.
Replying to @null and @sivavaid
These are (very) smart & curious adults, technoliterate but not generally engineers. Everything else is for me to decide! The brief is basically “hey Robin, you know that Web stuff, in fact you were there for some of it, so maybe a course?”
♥ 1
Replying to @null and @AmberBenson
These are very smart and technoliterate people, but not necessarily engineers (though some are). I’d love to see your list!
Replying to @null and @sivavaid, @JohnNaughton
Ah yes indeed, I thought of @timberners_lee's, and it's definitely a good idea for the earlier parts, but I'm not sure where my copy is 😬
Replying to @tigt_ and @AmeliasBrain
That's brilliant, thanks!
Replying to @alexanderdanilo
Ah! I was looking for that and couldn't recall what it was called! Thanks!
♥ 1
Replying to @jason_kint
I think Gopher and Minitel will have to be featured big in there! Also forgetting to switch your ftp client to binary.
♥ 2
Replying to @AmeliasBrain and @tigt_
Oh my, thanks for digging those up. Cc @danbri in case there's stuff you haven't seen?
Maybe @dckc knows what <dlc> was meant to be?
♥ 2
Replying to @honestduane and @ericlaw
Yeah, gotta love: "Four days later, Pei Wei told www-talk that he had made a browser."
♥ 1
Replying to @honestduane and @ericlaw
Oh, I have memories too. But… I don’t think you’d get a modern browser out the door in four days no matter what you binged on.
Sure, there are plenty of difficult questions in both science and philosophy, but at the end of the day: does anyone actually understand Microsoft Teams?
♥ 15↺ 1
Replying to @ericlaw and @honestduane
I mean, the initial git clone certainly might feel like it takes four whole days there ;)
♥ 1
Replying to @jorydotcom and @jay_hoffmann, @zcorpan, @tinysubversions
Thanks, these are great! That’s a lot of cool stuff to read.
♥ 1
Replying to @null and @tinysubversions, @jorydotcom
Can’t wait to go through these. I’m very much resisting “only ~8000 to go” jokes :)
♥ 1
Replying to @judyshapiro
That’s absolutely true!
Replying to @plinss
Oh yeah — though for the hardcore Web folks I think they were never more than an oddity, I’m surprised they made it into Mosaic!
♥ 1
Replying to @ogrange and @glazou
Ah! I haven’t seen that one, looking forward to it :)
Replying to @gloptimist
Thanks! I should probably read that no matter what :)
Replying to @htmlparserbook and @jorydotcom, @jay_hoffmann, @zcorpan, @tinysubversions
The HTML Parser is one of the strangest objects produced by human civilisation — it was going to have to be in there no matter what!
♥ 2
Replying to @zcorpan
Thanks Simon!
Replying to @thinman and @marccanter4real
Real and Shockwave! I wasn’t aware there were people old enough to recall those too!
♥ 1
Replying to @veroandi12 and @drchuck
I’ll check it out — thanks!
Replying to @oatesmeal
Thanks! I need to find time to start watching this.
Replying to @oatesmeal
Thanks — adding Abbate’s to my list, I happen to already have “How Not To Network” on the pile next to my desk!
Replying to @oatesmeal
Those sound very interesting — do you happen to have material you would share?
Replying to @MidasNouwens
I love the angle :)
Replying to @G4S90
Thanks!
Replying to @jay_hoffmann and @jorydotcom, @zcorpan, @tinysubversions
Super helpful — thanks a lot. Lots to dig into!
Replying to @staffanm
Thanks!
Replying to @TzviyaSiegman
I’d been meaning to read that, thanks for the reminder!
If it does happen, I’ll see about sharing! We’re still talking about doing Data Politics instead :)
Replying to @r0xanaradu
Thank you!
Replying to @KathrynTomasek
You mean the Web History CG?
Replying to @davrous
I’m tempted to say everything? I only use it for one org I’m in but I understand very little. I seem to get notifications in the local language of the person triggering them, but then I’m not sure because when I click on them I don’t understand what I was notified about.
Replying to @davrous
It’s also often unclear why some things are folded up (documents? threads?), when I’m chatting or when I might be commenting on something?
It doesn’t help that the org doesn’t seem to get it either and they’ve reorganised how they use it already because folks were confused.
Replying to @davrous
It’s also very slow, at least on Mac. And a colleague had a problem in the Web version where an Excel file wouldn’t load until he switched off ETP.
I guess it needs some iterations :)
Replying to @rektide
Yes, I’m already thinking of going more for principles and their history than a plain history. I think I would suck as a historian :)
♥ 1
Replying to @AmeliasBrain and @simevidas, @tigt_
If I recall correctly, @fantasai has opinions about this not being just a CSS problem :) (See discussions around <di>.)
♥ 1
Replying to @TzviyaSiegman and @danbri, @AmeliasBrain, @tigt_, @dckc, @timberners_lee
And we still have to contend with PDF.
Replying to @philarcher1
Whoa, cool — thanks!
Replying to @thinman
I know. In turn, Shockwave borrowed some of its eventing and painting from Win3.1 IIRC.
♥ 1
Replying to @TzviyaSiegman
Ooh, I need to check that.
Replying to @johnwilander
Man, we knew how to have fun back then.
Replying to @oatesmeal
No worries, I was just trying to be lazy. I have a lot of sympathy for slide minimalism :)
♥ 1
If you missed this yesterday — the wealth of material and ideas that people have provided has been amazing and overwhelming. I’m still processing what to do with all this (that also doesn’t repeat what has been done). So many deep thanks!
♥ 5↺ 1
Replying to @BrendanEich and @awbjs
Man, it can’t possibly be twenty years, I still remember when it was a baby!
Thanks Brendan!
Replying to @stuartweibel
Thanks!
Replying to @tigt_ and @AmeliasBrain
That confirms the alternative to <dl compact> theory.
Replying to @BrendanEich and @awbjs
I know, I know — it just seems impossibly long ago :)
Replying to @CuterThanBacon and @fgeorges
If you get me started on Perl there’s a risk I won’t stop.
Replying to @tigt_ and @AmeliasBrain
You say that like it would prevent it from happening in HTML.
♥ 2
Replying to @AmeliasBrain and @tigt_
And they were angry at the idea you could inline them!
Replying to @willikampmann
No, those options are on, you can check a checkbox to review them on a later screen — and the review is a trainwreck.
♥ 1
Replying to @willikampmann
This is when setting up a Chromebook account (for a child, but they don't care about that because they make you sign away your kids' data).
♥ 1
Replying to @awbjs and @BrendanEich
Thanks!
Replying to @tim_libert
Yeah, “how things went wrong with privacy” is a good thing to cover in general. I can also grab a lot of good things from @shoshanazuboff there, her history of Google is good stuff!
A lot of great resources on obsolete bits of HTML in here! I love how Lynx supported <FN> and <LH> — both of which I’m still waiting for — as well as less obvious ones like <BANNER>.
Quoting a tweet by @tigt_ ↗
♥ 3↺ 1
Replying to @jkohlmann
I’ve been wondering to what degree to go into issues — that sort of ties more into the other potential course, on Data Politics.
♥ 1
Replying to @gordonbrander
Thanks! Adding it to the list :)
Replying to @tigt_
Sharon Adler might be able to say more! youtu.be/E7JINlEt8Gc?li…
Replying to @jkohlmann
Right, if this were for kids or college stuff, or in the Valley, I would definitely include this sort of issue. But it's for the woke people of DIG so if I go into that kind of issue it's probably not as a small component.
♥ 1
Replying to @tim_libert and @shoshanazuboff
I think those two things are facets of the same process. I see privacy as a key angle to remove incentives to scale up acquisition of users as raw material. But I think overall we're in violent agreement 🤭
Replying to @rektide
OMG that's so perfect 🥰
We don't ever stop enough to wonder at how marvelous an invention the URL is. To take it away is to steal the heart of openness. (This superb extract heard through @rektide.)
Quoting a tweet by @karlicoss ↗
♥ 6↺ 1
Replying to @null and @marcosc
Based on a very superficial investigation it seems that it breaks when there's some form of tracking protection. It's so much incompatible with other engines as requiring a surveillance browser.
Wait till you see the UI though!
♥ 2
Replying to @tim_libert and @shoshanazuboff
Privacy is only partly an engineering problem 😀
Replying to @krishkhubchand
Thanks!
Replying to @dwhly and @solidstateux, @TzviyaSiegman
Yes, these are all good ones!
Replying to @nmg
Thanks a lot!
Replying to @pietercolpaert and @RubenVerborgh
Somehow I have never had the luck to see Ruben lecture! I need to catch that some day.
Replying to @samuelgoto and @cwilso
Thanks! Chris might have been around for a few of the interesting moments ;-)
♥ 1
Replying to @lauren_kelley
I ordered a kitchen knife thinking it was described in centimetres when it was, in fact, in inches. I now have this very sharp implement I could probably use to storm a castle or something.
♥ 3
Replying to @null and @maritza_johnson
I know, right?!
There were times in there when I thought “do I really need to learn this much about agrarian policy in the 1930s?” but then realised that no one had ever made power dynamics of the internet economy make more sense :)
Replying to @jyasskin and @WolfieChristl
I’m not sure that “what about other industries?” is the right question, and even less “what about corporations?” (cf Coase) or “what about brick&mortar?” :)
Questions I would ask might be:
• In what other industry would it be acceptable that 15% of the money in unaccounted for?
♥ 2
• How can it be that an alliance of some of the best investigative companies and some of the richest brands could barely pull this together, on a tiny fragment of the ecosystem?
• If we consider a free press essential to democracy, is this how we should run it?
♥ 1
• Is a system that treats every source of inventory as a “publisher”, whether it’s the only thing holding local government to account or the 7,000th torchlight app, the right system?
♥ 2
• If all of brands, publishers, and users are deeply unhappy with the system, what is wrong with the power structure in advertising that this system does not change?
♥ 2
• And of course, the all-time-favourite 😇: there’s one company that has the data to provide the transparency this study is (yet again) showing to be missing, why isn’t it working to organise that part of the world's information to make it universally accessible and useful? 🤗
♥ 1
Replying to @jyasskin and @WolfieChristl
Can you cite industries in which none of the primary participants know who the middlemen are or where their money is going?
This isn’t so much “cut out the middlemen” as it is “who the fuck are the middlemen and where the fuck is my money?”
♥ 1
Replying to @jyasskin and @WolfieChristl, @thezedwards
Sure, but I’m not sure what point you’re trying to make (sorry!)? You seem to be saying that it’s normal that people would cheat and pillage if there are no rules and/or no enforcement? But then shouldn’t we have rules and enforcement?
Replying to @jyasskin and @WolfieChristl, @thezedwards
I don’t think that deep structural issues in the ad market are “a distraction”, but given that they are organised by your employer and pay your salary, I guess you’re entitled to feel otherwise.
Replying to @jyasskin and @WolfieChristl, @thezedwards
Having said that, if you want an answer specific to that: I think it’s a requirement of democratic societies to ensure the existence of the type of accountability systems that the press contributes to.
♥ 2
Replying to @jyasskin and @WolfieChristl, @thezedwards
So if these mechanisms are dying, and the resources are going to monopolistic megacorporations and intermediaries that produce questionable value, then we should certainly look at how to organise our economy with different rules.
♥ 1
Replying to @thezedwards and @jyasskin, @WolfieChristl
I’d love for that to be true Zach, but in the best case scenario only the biggest of us can invest in that fight, and then not even enough. Small pubs simply can’t deal with this.
♥ 1
Replying to @dmarti and @jyasskin, @WolfieChristl
Don — that analogy only works if the shoplifters also run the protection racket and will beat you up if you don’t let them shoplift :)
Replying to @thezedwards and @jyasskin, @WolfieChristl
It’s something I’ve been wondering about. I think people are watching that and wondering what to do depending on how it pans out.
I get a sense a lot of people don’t feel they can meet sufficient standard of proof for (accessory to) fraud.
♥ 2
Replying to @jyasskin and @WolfieChristl, @thezedwards
I don’t think you can reach that conclusion. We’ve also removed open programmatic in apps in January; we’ve also very substantially decreased tracking. We’re always working on more improvements, we just can’t “Big Bang” get rid of everything.
Replying to @jyasskin and @WolfieChristl, @thezedwards
That’s not what that article says. It says that in one instance, at a very specific moment of a very specific market, that happened.
For most premium publishers, OM is for unfilled inventory. So it’s revenue.
Replying to @jyasskin and @WolfieChristl, @thezedwards
You are drawing completely the wrong conclusion with what you know is completely anecdotal information.
Replying to @thezedwards and @jyasskin, @WolfieChristl
I agree no one has cracked that nut yet. @UnlockProtocol definitely has an interesting offering (I like what they did for Forbes) but overall it’s not necessarily an easy product to put together well.
♥ 2
Replying to @dckc
A lot of the time these days I’ve given up and use RG, if only because a few times I’ve had some great exchanges with scholars from requesting papers (assuming that’s what you’re doing).
Alternatively, I’ve emailed authors directly — it often works (though not real time).
Replying to @dckc
Of course, there’s also Sci-Hub but of course you shouldn’t break the law.
Replying to @acfou and @thezedwards, @jyasskin, @WolfieChristl
It’s very sad.
Replying to @dckc
Oh, yeah, that’s very frustrating. You might want to look at unpaywall.org. They have an API (and a DB), if you have a DOI (which your ref manager should help with) then it can give you a link to the open access paper.
Replying to @zeynep
Both also have relatively high climate variety, too, so it seems unlikely it could be that.
Replying to @hackylawyER
Do you know how Google Photos is avoiding this?
Replying to @hackylawyER
Damn. 😕
Finally some advertising I can relate to. And since you’re asking: yes, yes they exist. store.brave.com/#/products/bra…
♥ 8↺ 1
Replying to @EliotSeattle
Would be a cool cameo :)
Replying to @briankardell
E4X was so close to being the right thing, it could have obviated JSX and made the DOM much nicer, more like lit-html. It was the unfortunate victim of XML bashing.
We needed custom elements twenty years ago, I mean we had drafts! Implementations!
♥ 3
Replying to @briankardell
Intents could still make the web better.
I agree with @AmeliasBrain that generic transformation tech (for JSON, XML, HTML) would make a huge difference but I'm unsure that XSLT in its current syntax would help.
♥ 3
Replying to @briankardell
It was not the best approach but fuck do we still need something like XForms. HTML forms are dumb.
It would double as a good data binding layer, too!
♥ 2
Replying to @null and @prushforth, @briankardell, @brucel
We could've made namespaces useful too! There should have been a default set always defined, limits on where to declare them, and ties to behaviour eg. through RDDL. Use cases should have come from limitations of the web, not from limitations of SGML.
♥ 3
In France, the tradition is for hairdresser salons to be called with abysmal puns. What should I call mine?
♥ 13
Replying to @AmeliasBrain and @briankardell
Actually, I'm pretty sure you could implement DOM updates from templates in the same way that lit-html or Svelte do, *and* the functional nature means when data changes you can rerun just the templates you need. Apart from syntax it's a great match!
Replying to @briankardell and @prushforth, @brucel
I like it but not sure we still need it? That said JSX is getting there, might just do it.
Replying to @edasfr
Nice!
Replying to @fberjon and @edasfr
Si si, il le faut!
Replying to @julien51
You're on fire 🔥!
Many good suggestions in here but I think this one is the most on brand 😺
Quoting a tweet by @_aurelienfabre ↗
♥ 3
Replying to @glazou and @briankardell
Yes, there were annoying incompatibilities. I think that was fixable, though.
♥ 1
Replying to @fatemehx2
Nice! 😁
Replying to @willikampmann
I didn't know! Yeah in France there's often the word "hair" used instead of what would be pronounced roughly like that in French. I haven't noticed that in English-speaking countries.
♥ 1
I'd be curious to hear which countries have traditions of terrible puns for hair salons, and where it comes from. Any beyond France and Germany? 🇫🇷 🇩🇪
Quoting a tweet by @willikampmann ↗
♥ 4
Replying to @jx
Haha 😂
Replying to @null and @benlikestocode
Haha those are good!
Replying to @plinss
All the Brooklyn ones near my place are unfunny 😭
Replying to @docum3nt and @willikampmann
Where did this tradition come from I wonder? There's a great @Bouletcorp comic I can't find in which he imagines an alternative world in which all shops do that, it's equal parts hilarious and terrifying.
Replying to @docum3nt
Wow, nice!
Replying to @null and @amyvdh
Hahaha brilliant!
Replying to @hober
Is it a West Coast thing? I have seen none in Brooklyn. I don't just mean hipsters, my low key salon is just called "Chaplin".
♥ 1
Replying to @justinfagnani and @briankardell, @kennethrohde, @marjakh, @manucorporat, @bmeurer
No need to dig up previous discussions to make it happen now. Almost any good Web tech idea has been discussed on a mailing list already, often in the 90s. The trick isn't op discuss it, it's to make it happen 😉
♥ 2
Replying to @briankardell and @justinfagnani, @kennethrohde, @marjakh, @manucorporat, @bmeurer
Oh, sure, I didn't mean to deter that interest! I just don't want to see it standing in the way of having something like lit-html part of the platform.
♥ 2
Le délire anti-voile était déjà le tréfonds de la connerie française, mais sous ce nouveau jour on en mesure encore mieux la bêtise.
Quoting a tweet by @lorientxpress ↗
♥ 5
Replying to @chaals
Ha!
Replying to @danbri and @JibberJim
That's what I'm suggesting: if the camp thing is cross-cultural, which seems to be the case, the punny names could come from the exact same root. I mean, if you're camp why not pun your shop name?
♥ 1
Replying to @michaelhkay
The lockdown changes *everything* 😁
Replying to @null and @prushforth, @AmeliasBrain, @briankardell
Yup! It's a lot of JS, though.
Replying to @bobdc and @hober
I assumed that's what you'd meant 😁 I'll check when I'm next there. But there are a few decidedly uncool salons near me, including the one I go to, and they all have boring names.
Replying to @null and @marionpdaly
They overlap but are not the same. You can't have privacy without meaningful security, you can breach security without breaching privacy. Privacy has a bunch of aspects completely unrelated to infosec, though.
♥ 7
Replying to @null and @marionpdaly
Just consider Google: amazing at security, one of the worst at privacy.
♥ 3
The notices will continue until privacy improves.
♥ 12↺ 1
Replying to @null and @funnymonkey
I mean, have you seen the CCPA regs? That's pretty much the idea 😲
Replying to @null and @funnymonkey
It's only going to get worse. With the stuff in the regs I'm not even sure how many disclosures are needed where anymore.
Is there a documented way to target bits of the Gmail UI for extensions, instead of relying on their autogenerated class names? A new version is out and so all the useless clutter is back.
♥ 1
Replying to @plinss
Man, if I could stop using it I would, but it’s company policy. Privacy issues aside, that thing is a usability disgrace. It’s basically a bunch of engineering gimmicks in a trenchcoat and it’s clearly optimising to extend time spent on site.
So workarounds it is.
Replying to @fantasai and @plinss
I. Can. Not.
Otherwise it’s evident that I would have left this steaming joke of a product in the dust a long time ago.
Replying to @plinss and @fantasai
IMAP is blocked, otherwise I’d be using it :)
Would really love to read more about this. I think it's particularly true of content recommendation algorithms, especially when you think you see why they're wrong.
Quoting a tweet by @badnetworker ↗
♥ 12↺ 4
Replying to @swodinsky
Ha! That would be good! Maybe you could explain why with years of my purchase history, particularly in books, TV, and movies, Amazon would keep plastering me with NFL ads.
♥ 2
Replying to @swodinsky
Lol. I just bought some new running shoes. The previous ones dated from I don't know when but definitely before I met my wife. We have a kid soon to turn 11. So I don't think it's that!
♥ 1
Replying to @masooga and @swodinsky
The context here is ads that Prime shows for for-pay channels & VOD (top slot in the EPG) so this is more like Facebook, they know everything.
♥ 1
Having said that, Facebook knew my gender and still very clearly put me in the women segment.
♥ 2
Replying to @swodinsky and @masooga
I don't know... I really don't see what would unless there's a weird connection between buying too many science/philosophy/tech policy books and watching NFL games!
♥ 1
Replying to @swodinsky and @masooga
I had given them my actual gender, and you'd hope they can model for given names outside the US? I suspect it's possibly behavioural.
♥ 1
OH: “If we didn’t scare people with our standards, they wouldn’t be good standards.”
♥ 10
@restofworld looks like an amazing and much needed project. I'm super excited to see where they go!
Quoting a tweet by @restofworld ↗
Replying to @null and @KendraWrites
I did oysters a couple of days ago and my only regret is not doing it sooner!
Replying to @null and @KendraWrites
A recent discovery! You wedge the oyster into it and use one of the slightly curved oyster knives to lever in and it pops it open much more easily than with the knife alone. I'm never going back!
TIL if you don’t have milk for your coffee, don’t use buttermilk.
♥ 22
Replying to @StefaniCrabtree
All I have is buttermilk and sour cream, and I know from experience that sour cream is a bad idea. Oh, and Marmite. Maybe I should have used the Marmite.
♥ 1
Replying to @null and @maritza_johnson
That’s an accurate depiction of what I looked like, except since I want the caffeine I’m actually still drinking it 🙄
Replying to @StefaniCrabtree
There is no point, Marmite always wins.
Replying to @StefaniCrabtree
I think the ice is key there. Otherwise it clots and has a weird texture.
Replying to @fatemehx2
I’ve actually never tried butter in coffee but that sounds like just using the half of buttermilk that’s bad in coffee anyway!
But anyway — the smaller humans in the house used up all the butter yesterday to bake a cake :)
♥ 1
Replying to @fatemehx2
I'm picturing something that turns into little blobs of oily fat floating around, but should I find myself in that predicament I will give it an open minded shot!
♥ 1
Replying to @null and @marcosc, @StefaniCrabtree
Aren't you Australian or something though?
Replying to @swodinsky
At least I had the good sense to put the glass down before reading more tweets 😁
♥ 1
Replying to @SBernardLermite and @NicolasTollena1, @FerryDanini
Il n'est pas rare dans l'introduction d'une nouvelle édition d'un texte (ce qui est le cas ici) de forcer un peu le trait sur l'importance de l'oeuvre. Cette citation manque complètement de contexte!
Replying to @null and @SimonDeDeo
Lol, just a wee bit.
Hell, I still remember when the worst thing that happened was how mad people got that Flickr decided to support video.
Quoting a tweet by @imcatnoone ↗
♥ 5
"New analysis offers roadmap to limit Google’s anticompetitive behavior in digital advertising", from the @OmidyarNetwork, by @ProfFionasm and @DavidDinielli. medium.com/positive-retur… (via @masooga)
♥ 3
“New York Times phasing out all 3rd-party advertising data”, you heard it first from @sarafischer!
axios.com/new-york-times…
♥ 154↺ 56
Replying to @hartzog and @sarafischer
We’re taking it in steps of course but we’re actually trying to fix this mess :)
♥ 6
Replying to @null and @theantitheory, @hartzog, @sarafischer
Very much agreed. I think that part of the problem is that there has been little marketer interest for it, it's not necessarily that publishers didn't want to build it but more that they couldn't sell it.
Scale is part of the issue. Marketers want to buy once, not once per pub.
♥ 3
If there were good adtech standards it would be possible to set up a campaign, distribute it across decentralised actors, and have it work in predictable ways that can be verified. But instead you have to go through intermediaries. This has hampered 1P efforts.
♥ 5↺ 1
Replying to @swodinsky and @AdMonsterGavin
This is about our ad business, not our marketing. When buying inventory from us we won’t be selling 3P targeting anymore.
♥ 2
Replying to @matthewstoller and @TimCameron, @ericwilson
The whole point is to not use 3P data, so that's definitely not involved.
I'm not sure which details are public at this point, but this is done entirely within the confines of the publisher-reader relationship.
Experian, ew.
♥ 4
Also, for the record, I know that people have a hard time hearing it but privacy and monetisation aren't opposites. You can actually seek to optimise for both.
♥ 2
Replying to @ka_iwanska
I think you’re reading more into the announcement than it is. This is a big step in that direction, but it doesn’t completely rid us of the kind of adtech you have in mind. We’ll still have open market programmatic on Web (it’s off apps and in Europe).
Replying to @ka_iwanska
But it’s progress I think! We’ve already removed a lot, it’s starting to look like something I like is on the horizon.
Regarding access & control the priority has been “safe by default” and we do DSRs of course, but yes that’s an area of progress. Input welcome!
♥ 2
Replying to @null and @theantitheory, @hartzog, @sarafischer
Yes, that’s where I hope we can get eventually — but it’s a whole industry to flip over and they’ve been ignoring problems for so long it can be hard to cut through.
♥ 2
Replying to @TimCameron and @matthewstoller, @ericwilson
Ha! If you reach out to the ad sales team I’m *sure* they’ll be delighted to bend your ear about it.
♥ 1
Replying to @gchampeau
Je crois que les deux choses sont distinctes. La privacy est l’avenir de la presse. Le consentement pour les cookies n’aide en rien la privacy.
♥ 1
Replying to @null and @borisschapira, @gchampeau
Renforcer la barrière à l’entrée n’est pas le but mais il est certain que le modèle tel quel ne marche qu’avec suffisamment d’échelle. Il faut trouver des solutions pour tout le monde.
Replying to @ka_iwanska and @idonevenknotppl
I’m not sure that makes NYT a walled garden. Or, rather, where privacy is concerned *everything* should be a walled garden. But it doesn’t have the walled garden aspects of being based on user content that can’t go elsewhere.
Replying to @ka_iwanska and @idonevenknotppl
And that tweet is, I don’t know, some kind of galaxy brain libertarian conspiracy theory?
♥ 1
Replying to @ka_iwanska
Sorry — those are *different* things :)
This article talks about stopping selling access to 3P data for direct sold ads. That’s _everywhere_. But it doesn’t address issues with programmatic, which includes other shady parts of adtech.
♥ 1
Replying to @ka_iwanska
For open market programmatic, we’ve already announced removing in apps and Europe. We haven’t announced anything for Web.
Replying to @ka_iwanska
So, to summarise:
• Direct-sold 3P targeting: off in 2021.
• Open-market programmatic: off in apps & EEA.
• 3P marketing data controllers: off everywhere except marketing offer pages.
A lot of adtech gone, but not all of it yet!
♥ 6↺ 1
Replying to @ka_iwanska
Yes, I agree. I think the explainable part matters, and that requires some thinking to get right.
Replying to @ka_iwanska and @idonevenknotppl
IMHO the single-controller case solves more than just security because you can infer a lot less if you fragment the sources (of course this assumes you’re not a huge company users use all the time). But in general agreed, we need more than that. I guess I have some work left ;)
♥ 1
Replying to @ka_iwanska
Yeah, you’re not the only one reading it that way. I think it assumes a bit too much understanding of the inner workings of the ad market!
Replying to @ka_iwanska and @idonevenknotppl
*blush* Thank you! :-D
Replying to @heathborders
Thank you!
Replying to @RobinvanBeauDow and @RobinCRLee, @sarafischer
I'm not sure what details are public yet, but I think you could guess 😁
♥ 1
Replying to @AureliePols and @theantitheory, @hartzog, @sarafischer
All good questions! I won't answer them here as I'm not sure what details are public yet but as you know much of this has to be public eventually anyway.
♥ 2
Replying to @null and @icopilots
Can it be perfect? No. But reducing the number of parties reduces risk.
Replying to @LAM_Barrett
Is it that you have a Word file but Word can no longer open it without crashing?
Replying to @Joshua_Pollack and @LAM_Barrett
Exactly. Start by trying to open it in Libre or GDoc. You might even be able to save it back to Word and see if that works. Check things carefully though because it’s likely you’ll lose some stuff (notably footnotes, I assume you have mostly footnotes).
♥ 1
Replying to @Joshua_Pollack and @LAM_Barrett
If neither of those options work, or if they destroy your doc too much, I’m happy to poke at the doc to see if I can fix it from the inside. Don’t… don’t tell too many people but I actually have some weird expertise with the DOCX format.
♥ 2
Replying to @lethargilistic and @LAM_Barrett
If this is a legal article of the kind I’m thinking of I would recommend somewhat against that (at least not without knowing what you’re doing). For one, the footnotes won’t be there, neither will any embedded reference management. But comments will be, in media res.
♥ 1
Replying to @lethargilistic and @LAM_Barrett
It’s entirely possible to fix this that way, but it’s a bit more involved (depending on document complexity)!
♥ 1
Replying to @LAM_Barrett
♥ 1
Replying to @lethargilistic and @LAM_Barrett
Absolutely! But if it comes to that we can save even more than just what is in document.xml :)
One weird thing in these pandemic times is watching yourself delivering a prerecorded presentation at a virtual conference so that you have the right context from which to answer questions at the end.
♥ 9↺ 1
Learning to use picture-in-picture in iMovie may have been a mistake…
♥ 4
Replying to @randypicker
Almost all I know about it is on the picture, you can find some variants on it in image search. There's a description of it in @superwuster's "Attention Merchants". I don't know much more!
♥ 1
Replying to @randypicker and @superwuster
Note 4 I believe is what you want, followed by a bit of context.
Replying to @randypicker and @superwuster
Replying to @randypicker and @superwuster
Replying to @TzviyaSiegman
I hope what was on screen looked better though.
Replying to @JulesPolonetsky and @JCE_PC
It would be great to do a review of the best ones!
Replying to @superwuster and @randypicker
Well thanks for that, I’ve used that picture while explaining privacy issues to advertising/marketing professionals and it never fails to grab their attention :)
Replying to @null and @sivavaid
Hahaha, there's only so much we can do!
March: I'm getting notifications whenever you install an app.
May: here's the root password just please let me pretend I still know how to read.
♥ 1
(You say you looked it up on the phone but what with, browser, OS, etc.)
♥ 4
Replying to @swodinsky and @zeynep
Yeah, I agree with @swodinsky, this looks like vanilla IP householding. You're using Safari which means you're protected from cookie identification but the trackers still load. So if whatever site you looked at the coffee maker on had the Pixel you're made!
♥ 9
I'm curious, did you look it up on a brand site and get an ad for another brand? Or on an ecommerce site and got an ad to buy elsewhere?
Because if so that's a neat smoking gun for the harms of tracking to publishers, brands, and businesses!
♥ 2
Replying to @zeynep
Do you mind sharing which site(s) you used? Because if you visited brand A and Facebook used that data to target you in favour of brand B, that's definitely something that marketers should be aware of and it would be a very strong reason for them to stop using Facebook trackers.
♥ 2
Of course it's anecdotal evidence, and many of us have suspected this for a long time, but just coming across a smoking gun is cool.
♥ 2
Replying to @can and @zeynep
It's hard to fully defend with an app inside the network, but in the case here it's possible to improve on Safari + guest chromebook. With stuff like Brave or a PiHole (or Tor, but there may be limits 😁) the link wouldn't have been made.
Replying to @can and @zeynep
Oh, in general absolutely, but that's apps. For the case at hand @zeynep used Safari on one device and sites can't enum SSIDs, and I assume FB in browser on the chromebook, which has the same restrictions. So that's not the vector, it's just IP IMHO. That's preventable!
♥ 1
Replying to @zeynep and @can
It's documented that they do householding and multidevice tracking. I'm not sure that how they do it is a matter of public record, they probably keep updating it to use the data they get.
Replying to @swodinsky and @zeynep
I think the answer to that depends a lot on what you consider to qualify as "disclosing" too. In clear language that users would understand? In detail as to how the black box works? Nope. But they do sell capabilities relying on that to businesses in the open.
♥ 1
Replying to @swodinsky and @aripap, @benthompson
The CCPA definition covers providing data to a third party such that they can reuse it for their own independent purposes (which is basically what selling means?). That doesn't happen here.
♥ 1
In fairness to the Chrome team this isn't a good year to boldly go anywhere much. I'm with them on this one.
Quoting a tweet by @marmalade ↗
♥ 2
I think that this is exactly right.
♥ 15↺ 2
Note that this matches what @timesopen also published very recently!
open.nytimes.com/to-apply-machi…
♥ 4
To build systems that extend the abilities of humans your leading skill has to be understanding humans.
That's why having engineers design technology is a mistake. And why companies dominated by engineering culture produce more oppression than liberation.
♥ 9
Replying to @randypicker
Yes, I’m aware of the MOOC, it’s just that I tend to fare better with books. I look forward to the related book too!
♥ 1
Me, not writing: Ooooh, writing is fun! I want to write! Yes, please please please I'll take this assignment it has a lot of writing!!!
Me, four hours and 1.5 sentences in:
♥ 11
Replying to @null and @benharnett
I know, man. And it's every. time.
Replying to @null and @benharnett
I’ve convinced myself that if writing isn’t painful you’re probably doing it wrong. So at least by that metric I must be a fucking genius.
Replying to @docum3nt
Hahaha that's good!
Replying to @null and @benharnett
I have tab, upon tab, upon tab of shitty outlines 😺
Replying to @StefaniCrabtree
That's the meeting at which they went "Fuck it, we're publishing Robin."
We're all in this together.
♥ 10↺ 2
Want to build great products that people can trust? Want to put users first like we used to before everything became about milking them for data? Come on over, we're hiring (for this and plenty of other positions)! 🌟
Quoting a tweet by @jkohlmann ↗
♥ 5↺ 1
I wish the ad industry wouldn't try to speak on behalf of publishers. As Jason clearly shows below we collectively went on the record in support of DNT as part of the CCPA regulations.
It's perfectly fine to disagree but please do that in your own name.
Quoting a tweet by @jason_kint ↗
♥ 20↺ 5
Replying to @RobinvanBeauDow and @RobinCRLee
Hahaha that's great!
Replying to @jevakallio and @paulg
Oooh that must hurt. I'm a VP too, but thankfully what little I studied was philosophy so I'm feeling pretty safe right there.
♥ 2
You've all seen this online, but in paper it's even more sobering.
And for once I agree with the US-centricity. Far more than 100k have died worldwide and they too deserve remembrance, but this crisis brings out US-specific maladies so sharply I find the focus justified.
♥ 95↺ 17
Replying to @tabatkins
They seem a lot more accessible too, no?
Replying to @gregwhitworth and @tabatkins
That’s really cool! I was more focused on the non-assistive parts: just looking at it it jumps out as having better contrast and legibility. I have 20/20 with just a hint of astigmatism and still find most digital products shite on this front. Good to see help from browsers!
♥ 1
Replying to @gregwhitworth and @tabatkins
Removing skeu is also a big plus! And colours aren’t used that much but whoa.
♥ 2
Much truth in this thread. TypeScript, React, CSS-in-JS, insane build setups… — JS feels like it’s been taken over by people who are sorry Java jobs aren’t fashionable anymore.
Quoting a tweet by @zkat__ ↗
♥ 11↺ 2
Replying to @null and @sivavaid, @cwarzel
I agree with Siva. You can safely ignore the cruft and just add notes to a notebook. I’ve been *very* impressed with the OCR, it actually seems to understand my handwriting, which is more than I can do myself at times.
Biggest downside is the API sucks, but you might not care.
♥ 1
Replying to @mariafarrell and @aral
I kind of like the thought that my face is really a small Republic of Bickering Spiders. Seems apt.
♥ 1
Replying to @justinph
Custom Elements come pretty close! I've been toying with a wrapper around LitElement that makes it usable without ramming TypeScript and imports down your throat and it's pretty nice!
♥ 1
In Australia, who laughs like a kookaburra laughs last, loudest.
Quoting a tweet by @RafaanDNgametua ↗
♥ 6↺ 2
When half the media thought that Apple and the iPad would save them I remember thinking that couldn't end well. Sadly, it turned out even worse.
Quoting a tweet by @stuartathompson ↗
♥ 5↺ 2
Replying to @mediated1
Yes, please!
What's the cool shorthand for "socially-distanced"? Socdis? Socdis drinks on my stoop tonight? That sounds horrible.
Maybe just "drinks apart"?
♥ 4
Replying to @temptoetiam
Je me permets d'émettre des doutes 🤭
Replying to @sroussey and @sarafischer
Not sure I understand, you mean let users import interests of theirs from third parties?
Replying to @sroussey and @sarafischer
I don't believe that's planned but it's early days, if there's demand why not?
♥ 1
Reading about the semantics of first-order logic and just went through several pages of how it's a problem that your signature might not have the right number of constants for your domain's elements... but then lol jk you can just add/remove constants until they match.
And I hate to be like this but I can't help thinking this might not have been that big a deal to start with? What am I missing?
CDA230 reform through executive order entirely out of trumpian spite isn't something I had on my list of upcoming tech policy topics but hey it's 2020 so ¯\_(ツ)_/¯
nytimes.com/2020/05/28/us/…
♥ 4
"Helped flourish" is a bit of an understatement. Under CDA230 Google has less liability for all of YouTube than your local bookstore does for the books it carries.
This lack of checks and balances is essentially a subsidy on runaway scale and monopoly.
CDA230 is good for the Internet in the same way that Citizens United is good for politics.
That said I would be shocked if that specific EO 1) held up and 2) got reform right. All it'll do is bring grist to the mill of people who still think it's 1993.
Replying to @johnwilander
That's very SoCal of you.
Replying to @null and @marcosc, @hober, @virginia
We've had ISO since 1947 and not once has it given us cocktails. I'd say this is an improvement.
♥ 1
Google announces they're removing the AMP requirement to become part of the carousel, without even a mention of Web Packages.
Is this the beginning of the end of proprietary portal formats like not just AMP but also Apple News & Facebook Instant Article?
webmasters.googleblog.com/2020/05/evalua…
♥ 27↺ 6
Also, is this the end of Web Packages? AMP was popular because Google forced people to use it, but without the "nice traffic you got there…" subtext it seems likely to be mostly niche?
♥ 4↺ 1
Replying to @briankardell
Sorry, unclear sentence structure. I meant to indicate that without Google's brawn pushing people to it, WP will be pretty niche.
AMP could stick around if people have to use it for other bits of the Google ecosystem.
♥ 2↺ 1
Replying to @briankardell
There's a difference between someone listing use cases and them actually materialising! What currently existing but imperfect solutions would it displace? How will producers be incentivised?
♥ 1
Replying to @briankardell
A lot of these work just as well with zip, the only difference is proof of provenance which many don't care about and can only be done by the owner anyway.
As expected, Big Tech is using the EO to shore up support for CDA230.
CDA230 hurts innovation by supporting runaway monopolies and harms free speech by providing safe haven to large-scale harassment. The EO is bad, but that doesn't make CDA230 good.
We expect better, @Policy.
Quoting a tweet by @GlobalAffairs ↗
♥ 6↺ 1
Replying to @histoftech and @ubiquity75, @mitpress
Indeed, indeed, though I suspect my computer will still be on fire.
♥ 2
Replying to @morar and @Policy
Healthy markets have balances that put costs on scaling up, without that you just get runaway winner-takes-all. Distributing content without accountability for its distribution is a great example of how that leads to structural damage.
This is not a EU/US thing, that's just playbook lobbyist talk. Under US law distributors eg. of an illegal book are accountable. Why not tech?
Replying to @teleclimber and @Policy
This thread is a start: twitter.com/cagoldberglaw/…. I also recommend reading Danielle Citron and Mary Ann Frank on the topic!
♥ 3↺ 1
Replying to @morar and @Policy
You only get a high volume of requests if you're too big to know your audience and if you favour centralised platforms over protocols.
Replying to @morar and @Policy
I did answer that question in my second tweet, and frankly that's a very caricatural dichotomy for someone trying to say others are divorced from reality.
Replying to @morar and @Policy
It's only an active choice if they have knowledge of the book's illegality. Under CDA230 that doesn't even apply.
Not sure what I'm dismissing?
Replying to @morar and @Policy
It's very hard to have a conversation when you're deliberately distorting everything I say. I did not refer to protocols as a legislative regime, I simply indicated that protocols lose to platforms because the legislative regime is playing against them.
Replying to @morar and @Policy
It's also hard to describe a regime when you have a completely caricatural view of the gamut of options.
So I reckon we stop here? If you want to have a real discussion about options maybe try not to shoehorn others into your own premade answers?
♥ 1
Replying to @chaals
Packaging would be great! I'd like to see (unsigned) support for bundles in browsers to see what people build with them. Maybe this requires building some primitives first?
SXG has a lot of great properties, it's the security model widgets wanted but never dared to define.
I hope it's not dead, I think it might actually stand a better chance if not in AMP context. Every chat about it I've been in started with "how is Google trying to screw us today?", not "is this thing good?"
As I told @jyasskin "there's a bunch of us who've failed there before so good luck!" It's 2020 and people still use PDF so clearly we need something!
Replying to @skdh and @WKCosmo
For-profit corporations have been going after people who deliberately spread lies for a while, unless we're dealing with things like libel or hate speech that's pretty common in Europe too!
No one knows how to port that model to social media, though. I'm not sure that Twitter is taking the right approach but I appreciate that they're trying something. It's certainly true that American constitutional fundamentalism isn't helping the conversation, though.
Replying to @erikphoel
And the books they were right to write are generally longer than they need to be.
♥ 1
Replying to @null and @borisschapira
Ewwwwww
Replying to @tabatkins
Can't argue with that.
Couldn't agree more, I have printed SEP articles piled up everywhere. I only wish it were a book, too.
Quoting a tweet by @susanthesquark ↗
♥ 1
Replying to @LAM_Barrett
Awwwwww! Thanks!
Wait — don’t those grow up to be, like, really really big?
♥ 1
Replying to @teleclimber and @Policy
It all started with good intentions, people I respect think that those good intentions are enough and/or magically still there despite evidence that it’s not working.
There’s plenty of room to get a sane reform of CDA230!
Replying to @swodinsky
Hahaha, what would REALLY crack me up is finding out that they have DoubleVerify targeting them away from coronavirus stories for that campaign.
♥ 5
That's a vibe to live by. t.co/kwnCt1LWkh
♥ 1
Replying to @BrentNYT
You had me worried for a second!
Replying to @gsnedders and @hober
What?! Stony beaches! It's like the beach but without all that sand! What's not to love?
Replying to @tabatkins
Maybe it's just in the process of bikeshedding the result?
No worries mate, always happy to give a hand.
"Our brand new Oversight Board can't work on what might be one of the most important cases they would ever get to discuss because they don't have laptops" has to count as one of the crappiest excuses Facebook has ever given.
Were they recruited over carrier pigeon?
Quoting a tweet by @JuddLegum ↗
♥ 13↺ 3
Replying to @Pinboard and @jkohlmann
You don't understand, it's because the Oversight Board, you know, they don't have laptops! It's a group of people specifically selected for having spent the past two decades in a cave. So no laptops. Also, because Covid-19.
♥ 4↺ 1
Replying to @cydharrell
They don't have laptops, Cyd, they just don't. It's because of the Covid-19 situation in These Troubled Times. Everyone understands that.
♥ 4
Replying to @jkohlmann and @Pinboard
The people are remote! They're so remote they don't have laptops over there.
♥ 1↺ 1
Replying to @Chronotope
Man, it's almost like there'd be some value in publishers joining a standards group and doing something about it.
♥ 2
Replying to @fgeorges
Not even planned!
Replying to @zeynep
But they don't have laptops, Zeynep! They picked a group of people without laptops. Surely we can all understand that would take until next year to set up.
♥ 12↺ 4
Replying to @dansaltzstein and @bokatlindell
Historically, that's not something they've been very good at.
♥ 2
Fort Greene.
♥ 14
Replying to @stuartathompson and @mariafarrell
does not do bad takes!
Replying to @can
Still handsome either way!
Replying to @can
I have yet to find one comfortable for my sizeable nose but apart from that I don't entirely dislike it.
Is anyone keeping track of how many journalists have been harmed covering these protests? Trump has been calling them enemies of the people for years now and it would seem some PDs have been listening.
Quoting a tweet by @MikeGeorgeCBS ↗
♥ 13↺ 1
Some of it here
Quoting a tweet by @studentactivism ↗
Replying to @jewcoffee
I guess guilty consciences don't want to be seen.