Kids these days...
Quoting a tweet by @Crystal1Johnson ↗
May 2017
462 tweets
Replying to @null and @SimonDeDeo
That's every May 1st, there's a big heterogeneous march of the far right. Being close to the presidency, it's likely to be big this year.
Replying to @null and @oscargodson
All the same, it matches what I actually see when commuting (or from my kids).
Also worth reading:
• «Democratizing Education?» dash.harvard.edu/handle/1/23928…
• «No Rich Child Left Behind» computinged.wordpress.com/2015/11/25/no-…
Quoting a tweet by @ethanwhite ↗
↺ 1
I have several hypotheses. 1/ MOOCs tend to follow a format matching university classes. They won’t help those for whom that didn’t work out
2/ They often assume the ability to dedicate a solid chunk of time at once. This won’t work if your spare time is «interstitial».
3/ Video is a bad format for commuting (esp. by car). Who commutes most? Not the rich.
♥ 2
I know lots of you are doing good with MOOCs. I’m not pointing fingers, just saying we can do better than college-educated talks to same.
Replying to @thisNatasha
I haven't been an AC in ten years but you'd get my vote! LMK if I can help.
This is more or less what I’ve been waiting for mobiles to do for the past ten years or so. medium.com/samsung-intern…
♥ 10↺ 4
I’m just a wee disappointed that the mobile experience isn’t called MeX, and everything themed with «DeX-MeX» jokes. You can’t have it all.
♥ 2
Replying to @gregwhitworth
I always thought Continuum was just for Office?
Replying to @gregwhitworth
That’s very cool! Everything I read was all about Office, Office, Office so I’ll admit I just dozed off…
Replying to @gregwhitworth
Gotcha. When I see «Office» I think of all the work I do to turn Word documents into something sensible; I don’t get excited much ;)
♥ 1
Replying to @gregwhitworth
I’ve been meaning to give Win10 a shot since I’m really tired of OSX and last I tried was Win8.1 (sadly, didn’t go well).
♥ 1
Replying to @null and @hollykrieger
I think it is well know that they do it on purpose.
Replying to @rhiaro
Can one read it?
For all that I often complain about video in MOOCs, it does have the advantage for the self-taught of telling you how things are pronounced.
♥ 1
Breastfeeding policies are a barrier to trade? The U.S. trade office thinks so. (via @recifs) foodpolitics.com/2017/05/breast…
♥ 1
Replying to @ElieSl and @jwajsberg, @webetcaetera, @frivoal, @rdeltour
Mais des ateliers de visu? J'avoue, je ne suis pas certifié! 😱
Replying to @ElieSl and @jwajsberg, @webetcaetera, @frivoal, @rdeltour
Oui, mais je ne sais pas comment tu organises les ateliers :) En tout cas je suis curieux d’avoir ton avis.
Replying to @ElieSl and @jwajsberg, @webetcaetera, @frivoal, @rdeltour
De mon coté j’avais commencé à dév une plateforme collaborative, mais certaines parties sont bien complexes (& depuis il y a d’autres trucs)
Replying to @ElieSl and @jwajsberg, @webetcaetera, @frivoal, @rdeltour
Aussi, je reviens régulièrement au manque de fondements intellectuels (presque + que de projets). J’aimerais bosser là-dessus (ms pas que).
Replying to @keithfrankish
The problem though is that after reading whatever she says there is no picture of her that I don't find ugly.
Pour ceux d’entre vous qui sont encore assez cons pour s’abstenir ou voter blanc: challenges.fr/election-presi…
↺ 1
Talking about your opponent is a losing tactic (remember Tim Kaine?). This is great.
Quoting a tweet by @mathieuvonrohr ↗
♥ 1
"The Vorrh", B. Catlin. Madly inventive, great writing.
La jeunesse emmerde les abstentionnistes. dailymotion.com/video/x5kphnm_…
♥ 1
American, British friends: France is toying with its own Trump/Brexit. Would you post videos, messages of encouragement under #PleaseFrance?
♥ 1
Replying to @null and @borisschapira
Putain, venant de @Guaino_Henri c'est fort. La paille, la poutre, tout ça.
Replying to @ElliotLepers
Les gens ont pris l’habitude de cocher toute checkbox en bas d’un form sans lire, parce que CGV, etc. Avec un radio c’est mieux je pense.
♥ 1
Replying to @SciPubLab and @csarven, @CrossrefOrg
JATS quality is overall way too poor for long-term archival. DOI is little more than a way to "look real".
Replying to @SciPubLab and @csarven, @CrossrefOrg
Because the data model is severely underconstrained and extremely print/visual oriented (to the detriment of useful semantics).
Replying to @SciPubLab and @csarven, @CrossrefOrg
This is not the fault of the JATS people: their job was to unify all the proprietary XML publishers made, so they made a _target_ format.
♥ 2
Replying to @SciPubLab and @csarven, @CrossrefOrg
It’s super easy to get information into it; if you’ve ever tried to get more than superficial information out of it, you know how bad it is.
Replying to @SciPubLab and @csarven, @CrossrefOrg
A great example is PubMed supplemental data commonly having as its sole metadata «Click here to download» — because that’s what renders.
Replying to @SciPubLab and @csarven, @CrossrefOrg
A good solution is HTML with embedded RDF (we can debate the forms). In terms of long-term semantics preservation, it can’t be beat.
Replying to @csarven and @SciPubLab, @CrossrefOrg
Going into details over 140c is going to be hard :) But you have to assume a future in which a lot is lost. No need for the apocalypse…
♥ 1
Replying to @csarven and @SciPubLab, @CrossrefOrg
…In computer world you can just wait a couple decades. Try reading early word processor files on a 5in floppy, etc. So some properties help.
♥ 1
Replying to @csarven and @SciPubLab, @CrossrefOrg
Popularity (easier archeology). Clear spec (HTML5 >> JATS). Easy distribution (many copies). Multimodal. Semantics over presentation.
♥ 1
Replying to @csarven and @SciPubLab, @CrossrefOrg
100 years from now I would easily bet on HTML being read. JATS you can infer some. LaTeX & PDF largely lost.
Replying to @csarven and @SciPubLab, @CrossrefOrg
I could go on — but at some point that’s no longer Twitter :)
Replying to @csarven and @SciPubLab, @CrossrefOrg
It would be interesting to do this kind of solid characterisation for scholarly content: w3.org/TR/xbc-charact…
Coïncidence de timeline :) (feat @TeamIsaAttard & @auroreberge)
Replying to @RobBelics and @yoavweiss
Yes, in this case the radical view that a few nasty things did happen during WWII.
♥ 1
Replying to @AutomatedTester and @MerriamWebster
I don't pronounce them the same, but that might be the French influence.
Replying to @null and @Floppy
Yay!
Replying to @TzviyaSiegman and @dauwhe, @amyw3c, @hober, @bigbluehat
A playgroup?
Replying to @TzviyaSiegman and @dauwhe, @amyw3c, @hober, @bigbluehat
Oh _that_ kind of playgroup. Hmmm, a baby would be okay but at my daughters’ ages I would fear psychological scarring!
♥ 1
Replying to @TzviyaSiegman and @dauwhe, @amyw3c, @hober, @bigbluehat
Also consider psychological scarring _of_ others. Keep in mind my daughters have half my genetic material.
♥ 1
Replying to @jwajsberg
C'est un principe universel: les incompétents ne voient pas leur propre incompétence.
Replying to @jwajsberg
Incompétent != stupide. Il y a un état d’esprit particulier à l’aveuglement vis-à-vis de son incompétence propre.
Replying to @jwajsberg
Tu veux dire un sabordage volontaire?
Replying to @jwajsberg
C’est sûr, mais il n’y a pas de raison de foirer volontairement. Le fake est mauvais parce que qui l’a fait ne sait pas qu’il ne sait pas…
Replying to @jwajsberg
Ah oui, je ne doute pas qu’ils savent qu’ils mentent. Mais c’est différent de faire un fake volontairement mauvais.
Replying to @jwajsberg
Le même fake, tout aussi intentionnel, aurait pu être beaucoup plus crédible.
Microsoft Word uses the Private Use Area for already existing Unicode characters. Also, the Greek letters are out of order because fuck you.
♥ 15↺ 11
Replying to @tbroyer and @codepo8
I am well aware, but it's the default for this dialog and therefore what almost everyone uses. It's a terrible choice.
Replying to @jwajsberg
Intéressant. Donc tu penses que c’est du débunking volontaire parce que comme ça on en parle deux fois?
Théorie intéressante de Julien (thread avant). Le conseil à la presse serait d’ignorer plutôt que de détoxer les fake news.
Quoting a tweet by @jwajsberg ↗
En fait c'est devenu facile les élections, il suffit de voter pour le candidat attaqué par Poutine, pardon, par @wikileaks.
♥ 13↺ 16
Pourquoi @wikileaks a choisi de sortir trop près du vote pour que les journalistes puissent l'analyser?
Parce qu'il n'y a rien dedans.
♥ 17↺ 25
Replying to @null and @Ecthelyan, @wikileaks
Oui, la source est le FSB, comme à l'accoutumée. C'est donc une date choisie.
Replying to @0xt48 and @maur1th, @rubin, @wikileaks
WikiLeaks, FSB, quelle est la différence entre un pigeon?
Replying to @null and @thaiamoi
Essaie de réfléchir au timing. Prends ton temps, j’ai l’impression que c’est pas facile pour toi.
Replying to @null and @thaiamoi
Ah oui, effectivement tu as du mal. Même avec la ponctuation. C’est pas grave, prends soin de toi, bisou.
Replying to @RobBelics and @yoavweiss
No, in this context the Le Pen response. But I don't expect you have any idea what's going on outside your village.
Petit air de France 🇫🇷!
♥ 1
Replying to @null and @TeamIsaAttard, @edasfr, @tut_tuuut, @cGuille, @jwajsberg, @benoithamon
Quel meilleur moyen de lutter contre le régime présidentiel que d'élire une présidente qui n'aime pas le régime présidentiel?
↺ 1
Replying to @null and @borisschapira, @TeamIsaAttard, @edasfr, @tut_tuuut, @cGuille, @jwajsberg, @benoithamon
Oh comme tu trolles!
Replying to @null and @borisschapira, @TeamIsaAttard, @edasfr, @tut_tuuut, @cGuille, @jwajsberg, @benoithamon
Oui, il faut qu'on s'y mette. Il est temps.
Je préfère prévenir: si MLP passe grâce aux abstentionnistes, je vais changer d'avis sur les tribunaux populaires expéditifs à la libération
♥ 2
Replying to @edasfr
Tu es trop sérieux :-)
Replying to @risseraka and @OriPekelman, @edasfr
Hmm, non c'est faux. L'abstention est celle de quelqu'un. Dès lors qu'elle est différenciée elle nuit plus à un candidat qu'à l'autre.
Replying to @risseraka and @OriPekelman, @edasfr
Ce n'est pas un principe, c'est un fait avéré.
Replying to @molokoloco and @OriPekelman, @edasfr, @risseraka
C'est marrant, il y a 6 mois le même genre d'électeur narcissique me disait la même chose sur Clinton.
Replying to @molokoloco and @OriPekelman, @edasfr, @risseraka
Whatever. "Se compromettre". Tu votes pour toi, pas pour les autres.
Replying to @molokoloco and @OriPekelman, @edasfr, @risseraka
Ah oui. Je ne vois pas le rapport avec la politique ni avec une quelconque volonté d'aider, mais ça doit être un hobby sympa.
Replying to @molokoloco and @OriPekelman, @edasfr, @risseraka
Oui, c'est aussi faire pour changer. J'ai quitté la LCR à la fin des années 90 quand j'ai vu à quel point ces démarches ne vont nulle part.
Replying to @hexalys
C'est juste la poubelle, qui sait qui avait pris ce bulletin?
Replying to @jwajsberg and @molokoloco, @OriPekelman, @risseraka
Oui, c'est la même pensée houellebecquienne qui anime les deux.
♥ 1
Replying to @molokoloco and @titouangalopin, @OriPekelman, @risseraka
J'ai raté un truc, vous avez eu un impact sur la réalité? Amélioré une vie autre que celle du vendeur de merguez (qui bossait un férié)?
Le FN est le dernier vieux parti encore debout. Dommage qu'il y en ait qui n'ont pas envie de l'atomiser. lesechos.fr/idees-debats/c…
♥ 5↺ 5
Replying to @dascritch and @jwajsberg
Et des candidats de partis innovants comme Something New ont fait des scores honorables, au-dessus de UKIP.
↺ 1
Replying to @molokoloco and @titouangalopin, @OriPekelman, @risseraka
Ah, "tourner sa veste". Le tribalisme et la loyauté comme valeurs. Sans moi.
♥ 1
Replying to @Nathan_lenias and @Monomarchos, @Sniperovitch
Ça ne disculpe pas, mais ça ne prouve rien. Les russes peuvent vouloir qu'on sache que c'est eux.
Replying to @RomainBurrel and @ElieSl
Forte participation à Brooklyn où MLP avait fait 40 voix. Effet Trump?
Quoi, moi, trop hardcore?
youtu.be/dNkYCOW_hLs
Résultats pour New York: Macron à 95%. #NiFachoNiCollabo
♥ 3↺ 1
Replying to @fabricedesre
Ne frime pas trop vite, je n'ai pas encore les chiffres de Brooklyn :-)
97.2% Macron à Brooklyn. #NiFachoNiCollabo cc @fabricedesre ;-)
♥ 5
Brooklyn: participation en hausse de 6 points et seulement 18 votes blancs. Trump, ça rend responsable. #NiFachoNiCollabo
♥ 2
Putain la 3D de France 2, c'est vraiment douloureux.
Sur Twitter, eh bien ça fait des courbes.
♥ 1
Qui est en charge des coiffures sur France 2?
Le renouvellement se fait plus vite en politique qu'à la télévision.
La tronche de Dupont-Aignan! J'adore.
♥ 1
The tide turns today.
♥ 10↺ 1
Replying to @gsnedders
Better: actual results in a broad panel of polling stations. Obviously an estimate, but highly reliable.
♥ 1
MLP a raison sur un point: le combat qui vient est entre mondialistes et enfermés. Le cosmopolitisme est seul avenir qui soit commun.
♥ 4↺ 1
Replying to @rdeltour
Ce soir je fête une victoire cosmopolite. Si demain je dois être dans l'opposition interne à ce bord, j'y serai!
♥ 1
Ce jour j'irai au Louvre avec toi!
♥ 2
Le premier problème du quinquennat: il faut trouver quelqu'un pour écrire les discours d'@EmmanuelMacron.
♥ 5↺ 3
.@EmmanuelMacron Putain, et qui fait ces images floues et sans balance des blancs?
♥ 1
.@EmmanuelMacron Je voudrais dire à @EmmanuelMacron que je suis disponible pour écrire son prochain discours, et tout ceux qui suivent.
♥ 3
Ah, @JLMelenchon s'adresse aux collabos.
La nouvelle dichotomie ouverts/enfermés se superpose pour l'instant à la vieille gauche/droite.
À gauche, nous avons @EmmanuelMacron pour l'ouverture et @JLMelenchon pour l'enfermement.
.@EmmanuelMacron @JLMelenchon À droite, l'enfermement est clairement pris par le FN. Une droite ouverte va-t-elle émerger?
Replying to @edasfr and @EmmanuelMacron, @JLMelenchon
Je pense qu'il nous faudrait plus que des tweets pour régler ce désaccord!
Replying to @edasfr and @EmmanuelMacron, @JLMelenchon
Je pense que c'est une conséquence difficilement évitable de la démocratie. C'est possible d'en sort, mais il faut vraiment le faire exprès.
Replying to @edasfr and @EmmanuelMacron, @JLMelenchon
Justement, dans le mien non plus
Replying to @edasfr and @EmmanuelMacron, @JLMelenchon
Je pense que c'est un attracteur fort de toute dynamique de grand groupe. Non pas qu'il faut s'y résoudre, mais il faut l'éviter exprès.
Replying to @floff01 and @edasfr, @EmmanuelMacron, @JLMelenchon
J'ai de très gros doutes sur les modèles liquides, justement à cause de la représentativité. Pour le coup la démocratie marche mieux AMHA.
Maintenant, on a notre propre Justin.
Quoting a tweet by @JustinTrudeau ↗
Les cop•a•in•e•s du Web: le gouvernement sera annoncé dans une semaine. Si vous aviez le choix de votre ministre de tutelle, qui serait-ce?
Replying to @SamyRabih and @axellelemaire
Je ne sais pas si elle veut enchaîner :-)
Replying to @jwajsberg
Ça je plussoie.
Replying to @OlivierNourry
J'aime bien l'idée danoise d'un ambassadeur (même s'ils se focalisent à tort sur les GAFA), je pense que le ministre peut avoir ce rôle.
Replying to @glazou
C'est peut-être le moment d'être proactif et de le signaler par avance!
Replying to @jwajsberg
On a qu'à dire cinq jours pour décider du programme, puis un pour voir qui colle ;-)
Replying to @null and @borisschapira, @OlivierNourry
Plus recherche, oui. Mais du coup ça fait sens d'avoir un coordonnateur.
Replying to @frivoal and @nitot
J'y ai pensé :-)
Replying to @frivoal and @nitot
Faut peut-être commencer par demander à l’intéressé s’il veut :)
Time To Philo: libéralisme ≠ fascisme
mailchi.mp/timetophilo/li…
Replying to @frivoal and @glazou, @nitot
Une petite campagne de soutien sur les internets et puis voilà.
Replying to @glazou and @frivoal, @nitot
Plus j’y pense et plus je me dis qu’avoir qqun du cru serait bon. Si Tristan ne veut pas (triste mais je respecte) on en choisit un•e autre!
Replying to @obrien and @goldengateblond
Not the whole left, mostly by the illiberal part. It’s a serious shake-up of political orientations; hard to tell where it’ll go.
Replying to @temptoetiam
Généralement la CGT est très heureuse d’aider la police à taper sur la CNT/AIT et aide à séparer le cortège.
Replying to @ElieSl and @GoulardSylvie
J’entends beaucoup son nom! Ça serait un choix très intéressant.
Tous les Américains que je croise m’embrassent, nous félicitent, applaudissent, crient «Vive la France!»
♥ 3
Replying to @null and @nitot
Tu as un bug tracker de prédilection? On pourrait en lister pas mal.
Can one of you kind and properly educated folks tell me what this symbol is?
Replying to @velvetype and @bfluzin, @jb_morizot, @fadebiaye
Merci! (Deux fois, ça a l’air top ShapeCatcher). Ça termine ma collection des symboles privés utilisés dans Word (pour les extraire).
♥ 1
Brilliant tool for when you see a character but don’t know its Unicode name or position: shapecatcher.com. (ht @velvetype & @bfluzin)
♥ 7↺ 3
Replying to @null and @nojhan
Réponse a. J’attends encore que mon clavier de téléphone apprenne à passer en capitale derrière!
Lecture fortement recommandée sur l'axe politique ouvert/enfermés qui vient (logiquement, enfants) à dominer le vieil axe gauche/droite.
Quoting a tweet by @LogicalAnalysis ↗
Validation de la @LaFrenchTech? :-)
Quoting a tweet by @antonhowes ↗
Replying to @mathieugallard and @ElliotLepers
En utilisant une grille à deux dimensions, rajoutant ouvert/fermé, je pense que les 23% très à gauche pour MLP feraient moins tache.
Replying to @edasfr
J'aimerais bien trouver un système pour dépiler efficacement. Pareil pour les tabs.
Replying to @ElieSl
Je parie tout ce que tu veux qu'il ne l'aura pas. Je ne sais même pas comment il peut y croire.
Replying to @slightlylate
If they want to protect them past the end of 2018, they're going to need to level yet up.
I begin to worry that they might.
♥ 2
Replying to @mathieugallard and @ElliotLepers
Plus floue que gauche/droite? Ça fait 6 mois que je n'ai pas vu 2 personnes qui ne s'accusaient pas mutuellement de ne pas être de gauche :)
Moral grandstanding: there’s a lot of it about, all of it bad – aeon.co/ideas/moral-gr…
♥ 1
This is easily one of the best books in web technology. I can't think of a better way to level up as a web developer.
Quoting a tweet by @heydonworks ↗
♥ 1
J'ai failli déposer une candidature @enmarchefr. Quand je vois ce que qui arrive à @manuelvalls je me dis que j'aurais dû :-)
♥ 3
Replying to @heydonworks
I mean it! You're the one to thank. Also, I am now curious about your taste in music.
♥ 1
Replying to @null and @borisschapira
Brutal.
Replying to @null and @dauwhe
It's bad, but the alternatives aren't great (for different reasons).
♥ 1
Replying to @heydonworks
Well reggae, I mean... I'd express an opinion but I don't want to lower the civility of discourse on Twitter.
♥ 1
Replying to @heydonworks and @deanleigh
That's not nice to the fish. I've been told there's great reggae for the past four decades or so. Still waiting...
♥ 1
Is there some kind of consumers' union in the US that you'd recommend to deal with a bank?
Replying to @skdh
Great paper title though :-)
Replying to @auroreberge and @TeamMacronPR, @EnMarche78, @JeunesMacron78, @azizfrancois, @annemarieidrac, @LerougeMarc, @LadroiteavecEM, @chdeveyrac, @RichardFerrand, @delevoye
Bravo, félicitations! J’espère que tu vas gagner :)
Replying to @charlesengelke
I’ll look into it, thanks a lot!
Quelqu’un sait pourquoi il manque encore ~150 #investitures alors qu’il y a eu des milliers de candidatures?
C’est uniquement pour ne pas se mettre en face d’alliés potentiels? Si c’est le cas, c’est décevant.
♥ 1
Replying to @leonor_de_b
Ça atteint un peu le coté «renouvellement»…
Replying to @Manuelas81
Je n’ai pas de problème avec les parachutages tant que ce ne sont pas des accords d’appareils. L’ancrage local des législatives est absurde.
I’m glad I opted against a career in academic philosophy… (via @DailyNousEditor)
Replying to @spignal and @SimonDeDeo
He was quite active in the campaign. I've also seen him around citizen grassroots movements now & then. It's not surprising & a good choice.
♥ 2
Replying to @tabatkins
Dude, seriously, no. No no no no no. You can use 🍋, but that's optional. Aïoli is made with garlic. It's in the damn name!
♥ 2
Replying to @tabatkins
You're hurting gentle kittens in the northern Mediterranean. Mayo is garlic-free aïoli. You could add 🍋 to either.
Replying to @tabatkins
Oh yeah, no one here calling it fancy. In fact, it’s a bloody staple.
But it is ontologically subservient to neither mayo nor lemon.
♥ 2
Replying to @null and @neuroconscience
I have to give it to him, he’s really helping with the whole «don’t let this become normal» thing.
Replying to @jcartillier
Ils pourraient aussi directement écrire les articles, c’est plus simple.
♥ 1
Replying to @OriPekelman and @mounir
Ç'aurait été cool de te présenter toi aussi, ailleurs :-)
Just to when I you think they can't level ☝️ the shit you can't make up.
Quoting a tweet by @WTFJHT ↗
♥ 3↺ 2
Replying to @edasfr
Non, ça a toujours besoin d'être dit.
Replying to @null and @nojhan
Ça fait un bout de temps que je pense à écrire un texte expliquant que pour les changer eux, il faut changer nous. Ils sont ce qu'on en fait
Replying to @null and @nojhan
J'avais plus soft à l'esprit. Si on va jusque là autant rééduquer directement les politiques (et élever la prochaine génération en batterie)
Replying to @csarven
Also, do DOIs really have that much of a point? It certainly wouldn't make my top twenty list of what scientists do wrong when publishing.
♥ 1
Replying to @temptoetiam
🔵
Replying to @Lthere and @velvetype, @bfluzin
Neat!
Replying to @temptoetiam
Ah ouais non ici ça ressemble à deux baguettes! Sinon 🥔 ou 📎?
Replying to @csarven
The problem is that scientists have come to see the DOI as establishing that something is real. It's nonsense.
♥ 1
Replying to @jeroenbosman and @csarven
Sure, that's what HTTP URLs + the Internet Archive are for. There's a platform right there that everyone except science uses. It's absurd!
The Brexit negotiations should include #Eurovision membership. Watch the leavers change their minds in a hurry.
♥ 3↺ 3
À 600.000 voix près, la France gagnait l'#Eurovision!
♥ 1
Replying to @jeroenbosman and @csarven
In orders of magnitude. Compare with Clockss in this case :-)
Replying to @jeroenbosman and @csarven, @CrossrefOrg
It's not for the resolution part but their brutally, unannounced decision to rate-limit their API et doesn't exactly fill me with confidence
«Emergence, free will, and causal responsibility» by @erikphoel. Looks like a great event next week in NYC: meetup.com/yhousenyc/even…
♥ 6↺ 2
This makes it sound like people prefer physical books for fetishist reasons. That wouldn't make my list of problems with ebooks.
Quoting a tweet by @cesifoti ↗
♥ 2
Replying to @ZachWeiner
In comic medium: you, Randall, @Bouletcorp I'd reckon. Then it gets messier (twitter vs blogs vs pop books vs well written papers).
Replying to @leonor_de_b and @AmericanGodsSTZ
Tu as lu le roman avant?
Replying to @CrossrefOrg and @jeroenbosman, @csarven
If the contract is enough to guarantee a stable URL, it begs the question: do we need DOIs? :-)
Replying to @CrossrefOrg and @jeroenbosman, @csarven
I'm not saying it isn't hard, I understand what's involved. Just that I expect infrastructure providers to be able to manage change.
Pulling the plug on the API does not exactly achieve that gracefully is all I'm saying.
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
I understand the principle. But that's no different from requiring a contractually that they provide redirects.
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
Again, I understood the understanding facts, justice not how the contract helps. Or how separating location from archival is sound.
Qu’est-ce que le libéralisme égalitaire? Comprendre la philosophie de Macron theconversation.com/quest-ce-que-l… (via @getdisturb)
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
My point is: if the contract is sufficient to ensure an up-to-date index, then it’s sufficient to ensure permanent URLs.
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
In other words, it would be possible to centralise the contractual obligation without centralising the resolver. The two are distinct.
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
To put it differently: if the contract works, you don’t need the resolver. If it doesn’t, the resolver will have some rot.
♥ 1
Replying to @edasfr and @getdisturb
Euh, non. C’est dire qu’il est moralement nécessaire de lutter contre ce qui empêche les gens de vivre/s’épanouir.
Replying to @jwajsberg and @edasfr, @getdisturb
Ben oui — c’est une évidence non? Tu vois autre chose dans cet article? C’est pas très surprenant en partant de Rawls+Sen non plus.
Replying to @rdeltour and @mediapart
Han, mon abonnement a expiré il y a petit bail déjà :-/
Replying to @jwajsberg and @edasfr, @getdisturb
Ah tu parlais de Macron — en l’occurence sur ce tweet là je m’en fous, c’est la philosophie en question qui est proche de la mienne.
Replying to @jwajsberg and @edasfr, @getdisturb
Après oui, pour l’autre on verra sur les actes — j’attends :)
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
How is that more fragile than centralising the same?
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
They manage things far more intrinsically complex like handling HTTP requests. Why is this special?
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
But you’re drifting towards a discussion of DOI in general. I think the fact that DOI is problematic is basically assumed in Web circles.
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
My question is more specifically about the value that the contract supposedly brings to the setup. I don’t really see it is all I’m saying.
Replying to @trekonomics
I wonder if this isn’t a government picked to win the elections, that will lose the presidency.
♥ 1
Replying to @joewass and @CrossrefOrg, @jeroenbosman, @csarven
Fair enough. Happy to buy beers over this another time!
Replying to @trekonomics
More than he picked the dream PM for much of the center+right, which will win him the election — but arguably the wrong guy for the next 5y.
♥ 1
Replying to @trekonomics
At the same time, you can’t do much if you don’t win the elections to start with — so it might have been the only option.
♥ 1
Replying to @trekonomics
Yes, it will be. I’m not a fan of the PM choice, but the EM candidate in my circo is good so…
♥ 1
Replying to @trekonomics
Yes, weak but not ineffective. My preferred result is a parliament with no majority. But you can’t really vote that :)
♥ 1
Replying to @edasfr
Normalement à ce stade quelqu’un doit te répondre «check-list».
#JeFaisLeTaff
♥ 1
Replying to @null and @nojhan
J'y ai pensé très fort. En plus, vu le bordel de l'échiquier politique il y aura toujours une raison possible.
«Responsive Design for Motion» — excellent and very thorough article from @cookiecrook webkit.org/blog/7551/resp…
↺ 2
I have a bunch of fresh @KeybaseIO invites if anyone’s looking.
Seriously, how haven't we helped Korea away from ActiveX already?
Quoting a tweet by @annabellekorea ↗
Very proud of the work you’ve all been doing on WPT — hard and often thankless. So thank you!
Quoting a tweet by @tobie ↗
♥ 2
Replying to @isabellehanne and @quatremer
Oui, c’est pour ça qu’on «it’s the last straw» aussi.
Replying to @null and @vbsteven, @ndw
So... Basically everything inside seven levels of tables at fixed resolution with huge amounts of font tags and poor accessibility?
Replying to @null and @ndw, @vbsteven
I was doing HTML 20 years ago (sic) and that's how it was more often than not. Just saying, the problem might not be the intent of the tech!
Replying to @gchampeau
Putain c'est l'enfer. Plus les onglets du mobile. Plus l'overflow data Pocket et Evernote. J'aimerais un truc pour trier ce bordel...
Is it just me or has @facebook's automatic timeline curation recently dropped from "noticeably bad" to "spectacularly useless"?
♥ 3
Replying to @roessler and @ndw, @vbsteven, @mavoweb, @cramforce
I'm not sure AMP is what these fine gentlemen are looking for :-)
♥ 1
Replying to @jwajsberg
J'aimerais jamais été frappé par une différence justifiant de se faire chier à changer (et applications devoir changer partout).
Replying to @ElliotLepers
"Ils ont l'argent, on a les gens." Vraiment?
Replying to @quatremer
Ah, de bons souvenirs! Quand je vivais là-bas ils ont fait un trou devant ma porte à l’automne qu’ils ont juste mal rebouché au printemps… 💕
♥ 2
Big turnout for @erikphoel's "Emergence, Free Will, and Causal Responsibility" at the Consciousness Club!
♥ 1
I presume all the "First Rule of Consciousness Club" have been made before, so tonight I'll spare you, dear twitters.
♥ 1
Neat summary of the supervenience argument.
♥ 1
Replying to @rgblong
Though maybe shifted somewhat to the left :-)
Dimensional Reduction pocket.co/sMTfRl
Replying to @trekonomics and @CBSAllAccessTV
Too many subscriptions already. For the Handmaid’s Tale I’d pay a month of Hulu then cancel. Might do the same there, if it’s possible.
♥ 1
«…spawning dozens of subprocesses may be harmful to your shell's equanimity…»
The things you have to worry about these days.
A potential alternative design for JSON Feed github.com/brentsimmons/J… (cc @danbri)
♥ 17↺ 3
Replying to @frivoal and @danbri, @mavoweb
Please consider commenting on the issue :-)
Replying to @danbri
But those were about RDF, no? Things have changed :-)
Replying to @danbri
I think it's worth a shot. Maybe suggest that in the issue? JobPosting is indeed a pretty cool example of what becomes possible.
Replying to @danbri
I hesitate to bring in too much baggage :-)
Replying to @mcnees and @skdh
If you read Amazon reviews for non-Apple music players, a common source of (sincere) five stars is "does not require iTunes, just copy!"
♥ 1
Replying to @hsivonen and @danbri
Ha! LD is RDF-compatible but largely designed by RDF-haters. You can almost completely ignore the RDF bits, many people do.
Replying to @hsivonen and @danbri
But yeah, I certainly did find a lot more use (and interest from people) in the real world than there is in standards.
Replying to @hsivonen and @danbri
I don't think RDFa ignores the RDF bits, it's even in the name :-) Consuming JSON-LD is JSON.parse(src) (which admittedly is disputed).
Replying to @hsivonen and @danbri
Oh, that's crap yeah. But schema.org RDFa is broadly deployed (comparable to OGP in magnitude I reckon) and processes as RDF.
Replying to @stevefaulkner and @hsivonen, @danbri, @w3c
You mean no matter how long and far I go, there'll still be a fork of me alive there?
😹
♥ 3
They email it with their steely knives
But they just can't unsub that list!
Quoting a tweet by @amyw3c ↗
♥ 2
After Brexit, BrexIP?
Can you please stop this maniac, British friends?
independent.co.uk/life-style/gad…
♥ 2↺ 3
Replying to @frivoal and @stevefaulkner, @hsivonen, @danbri, @w3c
I've never been big on royalty, but the idea I'm anyone's consensus is terrifying. I'm also not much into strong and stable URLs.
♥ 2
Actually, if it does turn out that I was designed by consensus, that would explain *a lot*.
Quoting a tweet by @frivoal ↗
♥ 3
Replying to @alanstearns
I think it was discussed after lunch.
J’avoue, j’ai tellement hâte de voir Guaino se ramasser. lemonde.fr/elections-legi…
Replying to @frivoal and @danbri, @mavoweb
Thanks, many good points there.
Replying to @LeaVerou and @danbri
There's nothing wrong with the tech there, that's just Google inciting usage in a particular way.
"Men Without Women", Haruki Murakami.
7 stories. Many small details that stir. "Kino" really touched me. His writing on fire in eponymous.
♥ 2
Noguchi Museum.
Robert Stadler at the Noguchi Museum.
♥ 2
Noguchi
♥ 1
Gide's "La porte étroite" evoked shapes like these when I read it.
Replying to @patrick_h_lauke
Classic
“Mike Freer and Me” by @torgo medium.com/@torgo/mike-fr…
Dear @IpsoNews, do you confirm that, and if so do you condone such behaviour from someone who sits on your advisory panel?
Quoting a tweet by @mickyb273 ↗
Replying to @au_tom_otive and @danbarker, @MalwareTechBlog
It's a bit far for me, but if I were in London I'd definitely go pay that conference a visit!
Replying to @dirkschulze and @AmeliasBrain, @svgeesus
The whole point of that series is for the people to have interesting disagreements over, so I'm not surprised :-)
♥ 1
Replying to @hsivonen
I don't understand your first point. For the second, it seems that you confuse "some" with "all", no?
♥ 1
Replying to @hsivonen and @danbri, @ndw
Ah, so that's how you see it. It's sad. It's so entirely the wrong lens. You're defending against attack from an imaginary "strike force".
♥ 2
Replying to @danbri and @hsivonen, @ndw
Haha, that's a good one!
Still, I'd like to know what to do when someone otherwise smart imagines enemies & ulterior motives everywhere :-(
♥ 1
Replying to @hsivonen and @danbri, @ndw
Sure, but that's not the case here. It's like someone wants to put paragraphs of text on the the web and uses <para>. 1/n
I'm saying "if you just call it <p> instead, then it'll be work with a bunch of existing cool stuff".
To which you seem to respond "oh noes, the HTML brigade wants to force us to use <keygen>, <layer>, and framesets!" I don't get that. 3/3
Replying to @hsivonen and @danbri, @ndw
I did, and I pointed out exactly which negative tricks I meant by that.
Replying to @hsivonen and @danbri, @ndw
Now, if you want to have a serious discussion about this, I'm happy to. It might be simplest if I just explained in code?
Replying to @hsivonen and @danbri, @ndw
You saw that in the context of feeds, I saw that in the context of not having to handle two formats for the same thing, as a developer.
I'd forgotten how adversarial people get in standards, it's not my normal anymore. We may well have different sensitivity now.
♥ 1
Either way, happy to just shake hands and work it out with some code!
Replying to @hsivonen and @danbri, @ndw
I'm sorry if you felt that way but I think you're attaching this to a history I don't in the least see it belonging to.
It's like I said "hey we could maybe use src on the video element" and you retort "you're bringing XHTML 2 back!" Didn’t see that coming...
Replying to @hsivonen and @danbri, @ndw
Except that the «it» you’re describing is not the «it» I’m doing. I’m saying «maybe boots when it rains», you reply «boots, like the Nazis!»
♥ 1
Replying to @hsivonen and @danbri, @ndw
All I’m doing is proposing to align a JSON format with what major consumers and millions of producers already do. You’re bringing RDF in.
Replying to @hsivonen and @danbri, @ndw
Well, that is where our perceptions differ. I know this stack, and I know it doesn’t bring in RDF unless you insist that it does.
Replying to @hsivonen and @danbri, @ndw
I am not analogising RDF and the Nazis, I am analogising your insistence to bring in RDF with some people’s insistence to bring in Nazis.
Replying to @hsivonen and @danbri, @ndw
Sort of parallelising with Godwin, if you will.
Replying to @hsivonen and @danbri, @ndw
In my view of the historical context yes, bringing in RDF is to call for flaming. I don’t understand why anyone would do that.
Replying to @hsivonen and @danbri, @ndw
You keep saying «it had the string RDF»; that’s like telling someone they can’t spell because they say «Referer» in HTTP.
Replying to @hsivonen and @stevefaulkner, @danbri, @ndw
Right. I’ve offered explanations for different PoVs. I’ve apologised if I’ve hurt your feelings. I’ve offered constructive ways forward…
♥ 1
Replying to @hsivonen and @stevefaulkner, @danbri, @ndw
If any of that interests you, the door is always open. If not, well, just stop talking to me.
♥ 1
Replying to @brianleroux and @janl
What if we made a simple, straightforward «Intervention CoC» (maybe with a better name) and rendered it common/normal to just drop the link?
Replying to @brianleroux and @janl
Social norms are there to be created. If I had that intervention.io link (or whatever), I’d definitely drop it in toxic exchanges.
♥ 1
Replying to @janl
FWIW it’s not just JS, it has improved but too much of Web standards is done with that sort of toxicity.
Replying to @janl
One thing that’s difficult is that trying to intervene against aggression is often seen as aggressive. Hence: a dedicated token could help.
♥ 2
Replying to @janl
The one thing that troubles me with the UN idea is that is seems rather work-intensive. I’d contribute some time, but does it scale?
↺ 1
Replying to @janl
You make an excellent point right there.
Replying to @janl
I spent a lot of time being a jerk myself because I thought it was the normal thing. If I can help, please LMK.
♥ 1
Replying to @shepazu
I think a lot of us found ourselves there because of that type of reason. I don’t think it’s healthy for standards, the Web, etc. though.
♥ 1
Replying to @ElieSl
Tu as oublié les guillemets à «philosophes». Ça change vachement le sens, là.
If this is an actual image from one of their symposia, I am definitely becoming a baboonologist.
Quoting a tweet by @gisela_kopp ↗
♥ 2
Replying to @jwajsberg
C’est pas un IBAN mais je crois que la Chase fait ça. En tt cas tu peux déposer un chèque juste en prenant la photo, l’OCR trouve le compte.
At the launch for the book that goes with the beautiful project that is @EverydayAfrica. You should check it out!
Replying to @a11yMel and @melaniersumner, @jensimmons
Is it bad that I immediately started computing the levels of nested tables this would require? 🙄
Replying to @a11yMel and @melaniersumner
I’ll take the curse anyway, that’s where the fun’s at :)
♥ 1
Quand je vois le nombre de bugs dans les formulaires de déclaration d’impôts de @dgfip_officiel, j’avoue que j’ai un peu peur…
♥ 1
Replying to @dgfip_officiel
À l’évidence, mais ces problèmes ne sont pas issus de configurations étranges. Exemple: accepter é mais refuser É sans aucune explication.
Replying to @null and @ndw, @melaniersumner, @jensimmons
You could probably get away with those but those vertical and horizontal lines that don't line up become unwieldy fast.
Replying to @jensimmons and @ndw, @melaniersumner
C’mon Jen, can’t you let some old people have fun coding like it’s 1997? None of ‘em fancy grids on this lawn of mine!
♥ 1
Replying to @jensimmons and @ndw, @melaniersumner
These are awesome — I should try to dig up some of my more preposterous table layouts from back then and do the same :)
♥ 1
Replying to @null and @ndw, @jensimmons, @melaniersumner
Huwah? Your version is at least four months behind! You should upgrade, security and all that :)
Replying to @null and @marcosc
That's a crappy crisis. Go grab B2G and start your own OS or something real, you know.
♥ 4
Replying to @null and @marcosc
Isn't 60 the new middle age anyway? I've been 40 for months and I haven't done anything weird yet.
Replying to @Sniperovitch
Ce que j'aimerais qu'on m'explique, c'est comment on peut revoir Prometheus.
Replying to @G_Devailly and @lshlj
Maybe we should run captioning contests on these until people are laughed into not doing it?
Replying to @null and @doctortovey, @ndw, @asbjornu, @danbri, @hsivonen, @RubenVerborgh, @Twitter
You fix that by showing the edit history :)
Replying to @null and @doctortovey, @ndw, @asbjornu, @danbri, @hsivonen, @RubenVerborgh, @Twitter
I’m not so sure: Facebook, Slack, Skype all allow edits under that sort of assumption and it seems to work!
Replying to @insoumis1986 and @mickael_andrieu
Huh? 1) En quoi le fait d'être une "société privée" a un rapport? 2) As-tu seulement un seul exemple où Google fait mieux?
Replying to @adam42smith
Maybe the reboot?
Replying to @heydonworks
That's compiled code man, not entirely fair
;-)
Replying to @heydonworks
Yeah but clearly there’s been some pre-processing. Mind you, I think React uses prettier, so the original is likely not great either! :)
♥ 1
Replying to @heydonworks
Yeah, that ain’t the source. It looks like that: github.com/facebook/react
Replying to @insoumis1986 and @mickael_andrieu
Tu y perds, pour rien. Par exemple, la branche accessibilité du W3C connait de forts dysfonctionnements endémiques. La plupart de ceux…
Replying to @insoumis1986 and @mickael_andrieu
…qui œuvrent à améliorer la vie des autres dans ce domaine travaillent en dehors de WAI — parce qu’il n’y a pas d’autre choix.
Fox News’s once-unquestioned dominance of cable news looks increasingly tenuous
economist.com/blogs/graphicd…
Replying to @insoumis1986 and @mickael_andrieu, @ElieSl
Évidemment qu’on peut participer au W3C… Ça ne change rien aux problèmes des groupes WAI. Tu n’as pas l’air au courant de leur situation…
Replying to @TomReller and @SciPubLab, @Sci_Hub
The music and film industries solved that by switching to more reasonable pricing practices. Are you considering doing the same?
♥ 2↺ 1
Replying to @TomReller and @SciPubLab, @Sci_Hub
We have different expectations of what an individual might consider «reasonable pricing». For 6 Netflixes you can print 20 pages. 1/2
♥ 3
Replying to @TomReller and @SciPubLab, @Sci_Hub
And get 20% off articles that are typically an order of magnitude too expensive. With respect, I don’t see Deepdyve as a serious attempt.
♥ 1
Replying to @jaffathecake
I always parse that one as «somehing». Then I tilt my head and try to remember wtf «somehing» means.
♥ 1
I've also been wondering how this might apply to self-learning through books versus video MOOCs.
Quoting a tweet by @erikphoel ↗
♥ 1
Replying to @erikphoel
My personal sense is that books work better, but I also don't like courses much (& have no university degree), so I might not be typical :-)
♥ 1
Replying to @namedgraph and @LeaVerou, @danbri
A lot of us tried RDF and gave up, not because it's new but because it has dreadful ergonomics. I mean, something as basic as arrays is pain
Replying to @namedgraph and @LeaVerou, @danbri
JSON-LD makes it usable, but that's mostly because you can ignore RDF is involved. And RDF bugs still spillover in modelling.
Replying to @namedgraph and @LeaVerou, @danbri
Oh, well, if the whole Web stack needs to be built on it then it’s a solved problem: that’ll never happen, so we can just drop it!
Replying to @namedgraph and @LeaVerou, @danbri
Either you can hack it or it won’t happen. Arrogance towards hacking never worked, never will. It’s the only thing that changes the world.
Replying to @namedgraph and @LeaVerou, @danbri
And, by the way, real-world distributed web work is done with what you call «JS hacking», eg. github.com/ipfs/js-ipfs.
Replying to @namedgraph and @LeaVerou, @danbri, @timberners_lee
Both of those are *extremely* hackable. In fact, all we did for the first ten-fifteen years was manage them with hacks.
Replying to @namedgraph and @LeaVerou, @danbri
Nevertheless, the Web persisted.
Replying to @asbjornu and @mazihe, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
It’s easy to validate schema.org. People don’t bother because the point is resilience over correctness — classic Web format.
♥ 1
So May thought Macron would jeopardise his relationship with all the EU for… what exactly? Is she desperate or just plain stupid?
Quoting a tweet by @pierrebri ↗
♥ 2↺ 1
Replying to @namedgraph and @LeaVerou, @danbri
LOL, I can’t believe the degree of arrogance. «They don’t use my pet toy, it must be they’re stupid!!!»
♥ 2
Replying to @namedgraph and @LeaVerou, @danbri
We use LD, a lot. We’d be in that list. Yet we stay as far as possible from treating it as RDF, because only bad things happen there.
Replying to @mazihe and @asbjornu, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
The question in thread was how to ensure an instance matches schema.org. But we can also talk about resilience elsewhere. 1/2
♥ 1
Replying to @mazihe and @asbjornu, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
Schemata have their use, but they introduce coupling and headaches in version management. A well-designed format can do without. 2/n
♥ 1
Replying to @mazihe and @asbjornu, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
See for instance berjon.com/xmlbp-schemata… for part of an old discussion on this topic.
♥ 1
Replying to @temptoetiam
MY EEEEEYES!!!!!
I don’t think that typography was ever cool.
Replying to @mazihe and @asbjornu, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
And no amount of calling stuff «bullshit» can hide a total lack of argument. But hey, you’re entitled to your opinion!
♥ 2
Replying to @namedgraph and @LeaVerou, @danbri
Maybe things wouldn’t escalate if you weren’t so condescending? Including to Scratch, pioneer in making code not just for a detached elite?
♥ 2
Replying to @mazihe and @asbjornu, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
What are those structures and datatypes XML has & JSON doesn’t? This reads like it was written by someone who can’t tell XML Schema from XML
♥ 1
Replying to @mazihe and @asbjornu, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
This document also seems to harbour notions of how interoperability is achieved that are anchored in a previous century, no?
Replying to @mazihe and @asbjornu, @webbr, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
Other than that this all seems solvable; I’ll let you know if I become a consultant in this space again :)
♥ 1
Replying to @webbr and @mazihe, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
It completely depends on the type of contracts. Purely structural? JSON Schema oughta work. Semantic? Use JSON-LD and SemWeb validation.
Replying to @webbr and @mazihe, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
For instance, we completely ignore the RDFness but we have customers who use SHACL on what we output.
↺ 1
Replying to @webbr and @mazihe, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
For our own stuff, we just use defaults everywhere and make processing resilient to errors. That’s a very strong contract, too.
Replying to @webbr and @mazihe, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
Depends of what you mean by «friendly»! If it’s «works with», JSON Schema can do that form of extreme coupling perfectly well. It’ll work.
♥ 1
Replying to @webbr and @mazihe, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
If however you mean «in the spirit of what draws people to JSON» you just can’t. That degree of coupling is painful and costly, always.
♥ 1
Replying to @webbr and @mazihe, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
XML was not designed to be used like that any more than JSON was. It’s not the syntax that’s unfriendly, it’s the overhead from coupling.
♥ 1
Replying to @webbr and @mazihe, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
So if you want to make that world «friendly», you want to revisit your architectural decisions before you revisit the format :)
♥ 1
Replying to @mazihe and @webbr, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
XML doesn’t do any of that either, which is also why people were drawn to it. That sort of stuff was added in later layers, like XML Schema.
Replying to @mazihe and @webbr, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
XML Schema only had limited areas of success because in practice only a tiny fraction of situations call for any manner of tight coupling.
Replying to @mazihe and @webbr, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
For most XML users, XML Schema didn’t make it useful, it made it painful. Especially given its initial interoperability was basically a joke
♥ 1
Replying to @danbri and @mazihe, @asbjornu, @webbr, @ndw, @hsivonen, @RubenVerborgh, @context
I was asking a serious question, Dan :)
Replying to @mazihe and @webbr, @asbjornu, @ndw, @danbri, @hsivonen, @RubenVerborgh, @context
If the architecture works for you & you just want the same for JSON, just use JSON Schema. AFAIK it can validate more powerfully than XSD.
Replying to @mazihe and @webbr, @asbjornu, @ndw, @danbri, @hsivonen, @context
I spent 4 years maintaining an XSD implementation; I took a quick peak at a random zip there, found nothing that looks all that hairy…
Replying to @mazihe and @webbr, @asbjornu, @ndw, @danbri, @hsivonen, @context
XSD is a rather weak validation tool. I saw nothing at a quick glance that couldn’t be automatically converted to JSON Schema.
Replying to @mazihe and @webbr, @asbjornu, @ndw, @danbri, @hsivonen, @context
But then you’d have to pay me a whole lot more than just holding my beer to look at namespaces with embedded version strings :)
Replying to @asbjornu and @mazihe, @webbr, @ndw, @danbri, @hsivonen, @context
It’s an old practice, see berjon.com/xmlbp-hard-to-… or berjon.com/xmlbp-naive-ve… for some bits. Also, @ndw in w3.org/2001/tag/issue… :-D
TFW you have no platform, and one of your rare concrete proposals is a really bad one.
Quoting a tweet by @WTFJHT ↗
Replying to @trekonomics
No idea if it fits, but it made me think of impossible/insufficient time in evolution. Specifically how evolution seems to have explored 1/n
♥ 1
Replying to @trekonomics
more of its state space (or faster, more efficiently) than it theoretically could in the current age of the Universe. Shortcuts or luck. 2/2
♥ 1
Replying to @trekonomics
In any case, I want to read this book.
I would like you to close Twitter and get back to work.
Now is good.
♥ 1
Replying to @trekonomics
Yeah, you picked a damn cool topic with an infinity of damn cool subtopics there.
♥ 1
Replying to @null and @ndw
We’re talking about XML Schema.
Le candidat FN de la première circo de l'étranger a vraiment le nom *parfait* pour son rôle. 👌
♥ 2
Replying to @null and @ndw
Even without trolling, the bit where you said «concise» confused me :) But yeah, it solves some problems, and I’m not debating RNG again.
You complain about CSS. You have *no idea* just how good you have it.
♥ 18↺ 5
Incidentally, they lie in that section: they also measure in 1/8th, 1/20th & 1/100th of an inch. And *all* occurrences are unitless. Enjoy!
♥ 2
Replying to @JibberJim
Now you understand why your Word layout always seems to get somewhat wonky.
♥ 2
Replying to @karlgroves
Truly, we are one glorious nation.
My languages, according to Twitter AI: English (US), Dutch, Tagalog, Spanish, Norwegian, French.
Quoting a tweet by @fantasai ↗
♥ 1
Replying to @null and @ndw
I don't disagree. I've long thought of XSD as the PHP of XML. If you're reasonable and know what you're doing, you can do fine things.
Problems arise when you have to deal with other people's schemata. I don't think those scars are ever going away.
Replying to @null and @ndw
Absolutely. And people use schema design tools that are less than helpful.
Replying to @katecrawford and @TeachGuz
I'm beginning to doubt there is a point at which unravelling turns to unravelled. It's like an abyss made of string.
♥ 1
Replying to @katecrawford and @TeachGuz
Time to weaponise kittens, I guess!
Replying to @seanmcarroll
Replying to @rubin and @valatini
Ceci dit s'ils sont mouillés tous les deux ils peuvent être impeached au même moment (ainsi que d'autres).
Replying to @rubin and @valatini
Je pense que Pence est aussi facho mais peut-être plus compétent. Je ne suis pas sûr du tout de vouloir un impeachment.
Replying to @null and @tomroud, @rubin, @valatini
On a déjà un connard boiteux, ça ne l'empêche pas de grand chose.
Replying to @valatini and @rubin
Aucun président n'a jamais été impeached, donc deux encore moins. Mais bon, on est sur un nouveau territoire.
Replying to @Timmy_dSV and @rubin, @valatini
On ne sait pas. Il a menti sur les liens russes de certains autres pendant la transition.
♥ 1
Replying to @LogicalAnalysis and @aeonmag
Great article, just a nitpick: French does have a spelling bee analogue of sorts (and for good reason too).
♥ 1
Replying to @tobie
How do you mean?
Who you gonna call?
♥ 1↺ 2
Replying to @tobie
You can easily build a queue depiler with async, too. I don't recall which primitives off the top of my head, but it's not hard.
Replying to @tobie
async is ❤️. The only reason we have promises, await, etc. is people don't use it enough 😀
♥ 1
Replying to @null and @amyvdh
I should get a poster of that one to put by my bed.
Replying to @AutomatedTester
Ow
Replying to @C4COMPUTATION and @SimonDeDeo
I think it's operating in a regime entirely its own :-)
♥ 1
Replying to @asbjornu and @LeaVerou, @namedgraph, @danbri, @HydraCG
If you dig through the old work on the Contacts API, you'll find stuff like that.
Replying to @null and @marcosc
It doesn't say what we'd get to do to you. Just sayin', might serve as a motivator.
♥ 1
nextstrain: really cool project to track virus evolution
nextstrain.org
♥ 2↺ 1
Replying to @alexmilowski and @ndw
Lining up versions when dealing with cross dependencies isn't new. The confusion decays, whereas when versions don't match it's permanent.
Replying to @alexmilowski and @ndw
I don't know, they're not giving me koolaid much these days. But if Netscape and React could do it, XProc will live.
Replying to @null and @ndw, @alexmilowski
LOL. I’m not sure it’d do much good to XProc if Netscape were to implement :)
♥ 1
Depression’s Upside pocket.co/xM33M3
@KeybaseIO It’s nice to send emails about more devices, but your app isn’t available for a recent Sony with a decently up-to-date Android?
“Floating labels are problematic” by Adam Silver medium.com/simple-human/f… (via @nhoizey)
♥ 2
Replying to @goetsu and @nhoizey
The contrast issue cannot be fixed with design. If the label has low contrast it’s hard to read; ½
Replying to @goetsu and @nhoizey
if it has high contrast and is used as placeholder, it looks like the field is filled. 2/2
Replying to @goetsu and @nhoizey
Also, most implementations are brittle to variations of zooming, unexpected font change, etc. And when they break your form is screwed.
Replying to @goetsu and @nhoizey
If you increase it always the field looks filled; if you increase on focus then you need to focus to read: pretty bad usability!
Replying to @goetsu and @nhoizey
That’s the problem with floating labels: they try to be both label and placeholder. I’ve never seen an implementation succeed at that.
♥ 2
Replying to @hober and @tabatkins
I only ever solved that by giving up on the cheap part. Same for hats! It takes money to buy boring.
♥ 1
I have become an expert at typing "git push" in Slack right after I've "committed" a message to the channel.
Quoting a tweet by @yoavweiss ↗
♥ 5