Two-factor authentication on Wikipedia for admins and up.

Jimmy Wales’ Wikipedia account got hacked the other day, and it turns out a pile of others did too. So two-factor authentication is being made available for everyone with powers from administrator up on any Wikimedia wiki. Go to Special:Preferences and set it up.

(If your account got hacked and has been locked, go to Steward requests. There’s a bit of a queue, please be patient … else it’s time to fire up the powerless sock account.)

It’s still a bit fiddly, so is being rolled out slowly. (The aim is to have it available to all users in due course.) Authentication methods include mobile phone, Google Authenticator and emergency backup numbers you can print out and keep on hand (“scratch codes”). BWolff (WMF) notes:

If you lose your scratch codes and your 2fa device, and you can prove who you are beyond doubt (what “beyond doubt” means I’m not sure, but I guess committed identity is a good choice), then a developer will remove the 2fa from your account. However, please don’t lose your scratch codes.

I use two-factor at work (GMail, Github, AWS) and it’s just fine. This is basically a really good idea.

Note that AutoWikiBrowser will be a bit fiddly, you will need to set up a BotPassword. (AWB plans to support OAuth soonish.)

At least avoiding another Tubgirl is Love incident won’t require distributing RSA keyfobs to the user base. (Though WMF wants to support fobs too.)

Update: Tim Starling on what actually happened. tl;dr change your password and SWITCH ON 2FA, IT’S IMPORTANT.

Attack of the 50 Foot Blockchain.

I’m writing a short book on Bitcoin, blockchains, smart contracts and why all this garbage is garbage. I hoped to have it out by now, but it turns out writing is work! My target is 500 usable words a day. Currently at 16,000 words of draft, I expect this to hit 20,000 (almost certainly not more than 25,000) and then I’ll cull it to size.

I’m occasionally ranting about it on my Tumblr. You can read the tag in reverse order or chronological order.

(and no, I probably can’t actually call it Attack of the 50 Foot Blockchain. Suggestions welcomed.)

Yes, my edits on cryptocurrency-related articles have helped a great deal in the research …

FAQ answer: Sadly, Amazon Kindle only accepts filthy fiat.

Podcast with me on “Neoreaction A Basilisk.”

Eruditorum Presscast: David Gerard (Neoreaction a Basilisk 1). In which Phil Sandifer and I talk for seventy-five minutes on the wonders of Eliezer Yudkowsky, Mencius Moldbug and Nick Land, the end of the world and the monster lurking in the labyrinth of every philosophy. I dispute that I was “reading neoreactionaries before it was cool” — life is too short to read Moldbug. Phil read him so you don’t have to. Go order a book, it’s really good.

“Neoreaction a Basilisk” by Phil Sandifer is kickstarting.

I’ve spent the last six months editing a book. Phil Sandifer found himself writing about “A genre dominated by, in effect, an AI crank, an extremist technolibertarian, and whatever the fuck Nick Land is” and I begged to preview it. I ended up researching, editing, copyediting and helping with the publicity. It has been six months of solid and hearty yuks and lulz and a sheer delight.

The kickstarter is up now (announcement). So far it’s landed about $1500 in twelve hours; people seem quite keen to get this book. And let me assure you that the stretch goal essays are also things the world needs.

There are also excerpts ([0] [1] [2]) and images of what the conspiracy zine and full colour editions will look like. (If I had $70 of actual money spare I’d be sending it in to get the conspiracy zine and colour editions, which look to be gorgeous productions.)

“Or, to put it another way, this is a book that uses Eliezer Yudkowsky, Mencius Moldbug, and Nick Land as a loosely stitched together foundation on which to build an oddball philosophical structure made of bits of Hannibal, China Mieville, Alan Turing, Thomas Ligotti, John Milton, and a futuristic AI that will torture you for all eternity if you buy a mosquito net.”

edit: and at $3000 in the first 18 hours, Phil decided he’d better preview the $4000 essay, “The Blind All-Seeing Eye of Gamergate.”

Internet fundamentalism doesn’t actually work.

There’s an antipattern that Internet social sites tend to: people get nostalgic for the old times, and think that if they can just get rid of these annoying newcomers they’ll get the old site back again just like they remember.

This never works.

The primary fallacy is that the reason it was interesting back then was that nobody knew what they were doing and what would come next; attempting to encase that in carbonite is unlikely to achieve the desired effect. The secondary fallacy is that they themselves are different people now.

(I’m thinking of two sites I frequent — RationalWiki and LessWrong — going through precisely this angst right now. And of another, where a Usenet newsgroup from 15 years ago just revived itself on Facebook, and it appears to be working precisely because we’re different people now so it’s interesting again and we don’t know what happens next.)

“RETURN TO THE MISSION WILL CURE ALL” just won’t work out like you think it will. All communities have a life cycle and will die; recovering yours will be at least as uncertain as striking out with it originally was.

Wikipedia: We’ve won. No tail-lights. Now what?

Wikipedia has won. Wikipedia is the encyclopedia anyone actually consults, ever. In fact, it’s the first in history that everyone actually reads, rather than just having fond high school memories of. Wikipedia now defines what an “encyclopedia” is in popular conception. Wikipedia conventions now shape the English language.

So we don’t have tail-lights to chase any more. What sets our direction? Do we just drift? What does “encyclopedic” actually mean when we can’t just point at Britannica and assume people will understand? “Do more of whatever it is we’re already doing” is the default for new recruits, but with no direction we risk going over a cliff.

It’s 2014. What is an “encyclopedia”?

For Wikimedia in general, we have a good vision statement with clear implications, and a mission statement which is only slightly adverb-hobbled. Neither has the necessary level of detail to accurately explain Wikipedia to the world as well as to ourselves.

Alec Conroy noted a few years ago, in a fantastic foundation-l post on the terrible content filtering idea:

The further we can get away from the model of elementary schools and towards the model of the global universities, the better.

This gives a conceptual model to work to: Wikimedia as the sum of all university libraries. Obviously, this is far from the complete answer — we already do both more and less than that — but it’s the level of vision we need to chase to automatically know what to do next.

What other examples do we have of conceptual goals on that level?

Edit: Hacker News comments.

What I did on my not-a-holiday to SF.

My passport got back in time and off I went last Wednesday afternoon, to help Wikimedia hire their next Chief Communications Officer, in my capacity as press volunteer of nine years. It’s no longer just “press officer”, but a C-level job, so much more powerful and able to wreak more damage. So, days of interviews, we run them through the combine harvester both ways, they get grilled and tested and we demand they produce shrubberies. The reason to bring an experienced volunteer in is because it is literally impossible to do the job without leveraging volunteer efforts — you simply can’t scale. Apparently I asked highly suitable questions.

Let me reassure my American readers that United’s customer service is every bit as legendary in the UK as in the US. They were actually completely nice, but the trololol comes in (a) making “economy” into fourth class rather than third class (you want “economy plus”, for another $117, if you’re over five feet tall) and (b) having a total of two toilets for the two sections. The pilot’s announcements were mostly doggerel poems about flying.

Arrived at SFO and through border control in about two seconds — I knew wearing the Wikipedia shirt was a good idea. (Seriously, get a shirt — it opens doors. The world loves us!) At 5:40pm (or 1:40am the next day by my body clock), SFO is all but deserted. Called WMF, taxi to meet Sue and James at the office. James offered me one of the several hundred varieties of coffee on hand, I mimed cramming a coffee pod directly into my arm. I got the cup of caffeinated mud I desired. (Jet lag is a major theme of the next week and what I laughingly thought of as a circadian rhythm is still scrambled.) Off to hotel, checked in, discovered the room wifi sucked. #firstworldproblems

Every tech loft conversion in the world is identical. I realise they were modelled on the SF ideal, but I could not tell from the place if I was in Market Street or Old Street. Buy a tube of instant extruded tech loft product, squirt it over the walls and the big silver air conditioning pipes just slide into place! Sprinkle with MacBooks to taste. My visit was marked by a disconcerting lack of culture shock.

To Wikimedia Thursday morning! … for a meeting that was cancelled. So I spent the day hanging out on Facebook. (And discovered Deepak Chopra’s been busted sockpuppeting Wikipedia for the past five years.) Apparently no-one wanted to interrupt me in case I was working on something important, bah. Off with Danese to some coffee shop I forget the name of for an exquisitely nice iced coffee. Wander back (via another couple of tech lofts identical to every other tech loft … really) for the Wikimedia Christmas party, where I got to catch up with all manner of people I knew only online. There exist photos.

Courtesy jet lag, I could get away with caning it once (and delighted in doing jumping jacks at a severely hungover Oliver, who is half my age so should have done rather better), so off to Wikimedia Friday morning for some serious candidate grilling. We set up a particularly evil potential PR disaster (which I won’t detail until all this is over) to see how they handled it. I told them how usually I just post about six paragraphs of “this is the situation, this is what we’ve done before, this is how to get this one to die” to the mailing list, but this time they had to work out the answers. A test of their PR instincts and whether they could ask the right questions. Then a two-hour meeting to go over how they all did.

Then out to the bars with Jed to do what any two techies in a bar do, i.e. complain about work. He took me to Zeitgeist, a rock’n’roll bar filled with pretty young hipsters with script tattoos on their feet. Tell you what, if you’re in SF you can’t complain about American beer. Oliver notes that script tattoos are the direct 2010s equivalent of the 1990s tribal tattoo; so for the 2030s we’ll need tribal script tattoos, in fluorescent and glitter. (I’ve said it, now it’ll happen.)

Back in on Saturday morning for more candidate interviews. Then more dissection of the day’s many interviews … then down the upscale bar with Geoff and Jove for all manner of confidential on-topic (I assure you) discussions.

Sunday morning, I woke up and realised my plane was going at 7:30pm, not 2:30pm. I would have spent the day at SFMOMA, but it’s shut for renovations! Bored, I eventually just went to the airport to leech their wifi. SFMOMA has a lovely shop at SFO, including their catalog for a mere $20. YOU MUST GET THIS.

Got back Monday afternoon London time and went to work Tuesday. Still falling asleep and waking up at random times. Oh, there’s still credit on my BART card.

I am slightly amazed how brilliant just about everyone is at Wikimedia. They put out job specs that say “magical flying unicorn pony that farts rainbows, must have rocketed from Krypton as an infant” … and then they get them. I was amazed how good the CVs were. But the same goes for the people working there. They work like they’re at a hot startup, not a dot-org. Because they’re changing the world.

Go and test the Visual Editor. It’s meaningfully testable.

The Visual Editor is live and in testing on English Wikipedia.

People are past the stage of shouting IT’S A DISASTER AND THE WHOLE PROJECT MUST BE SHUT DOWN (that was last week); it’s reached that sweet spot on the upslope of the logistic curve of progress where the basic functionality is in place, it’s usable enough to beta properly, and all the little finicky bugs are coming out. (“Yes, it edits now, but I WANT A PONY. A nice one, mind you.” “IF YOU CAN MAKE A PONY, A UNICORN IS TRIVIAL.” “YOU CALL THAT A UNICORN? YOU’VE GLUED A CORNETTO TO THE PONY.”) And the devs are actually keeping up with them. Holy crap, they’re successfully flying this beast TO THE MOON.

So, experienced editors: please go test the visual editor on your gnarliest articles. Report any breakage, big or small. Save a bad diff, self-revert then report the breakage. Give Oliver something to do with his copious reserves{{cn}} of free time. (In between coming up with “cornetto unicorn.”) Update: and please check and report problems with newbies using the visual editor. (HT Nemo.)

(The visual editor is somewhat annoying, but it works well enough that the bug reports are useful. I find it easier, if slower, for correcting a typo; harder for adding a sentence with a reference. But the main reason experienced users should be testing it is that it’s important — and experienced users will give it the serious kicking it needs to get the bugs out.)

ACHIEVEMENT UNLOCKED.

U.S. Attorney Carmen Ortiz is fighting to hold on to her job, and to avoid professional disciplinary proceedings. FUCKING LOL.

(Here are the Carmen Ortiz and Stephen Heymann petitions, if you’re a US citizen and haven’t signed them already.)

Congress spent the last year wondering every day “is this the next SOPA?” Now they’ve found out what is, and it’s the dude who started that one too. No wonder they’re panicking.

(When Aaron Swartz met the US senator. This is why they killed him. You don’t understand just how much they loathe and despise the Internet.)

It sucks so much Aaron isn’t here to see all this, and laugh and laugh. Guess we’ll all have to make up the difference. It’s incumbent upon every one of us to FUCK SHIT THE FUCK UP. In an orderly, creative and productive manner. So, what’s a good project? I’ve spent the last week depressed and pissed off; it’s time to get moving.

PR people in the EU: obey the bright-line rule, or suffer legal consequences.

I’d heard of the German ruling, but hadn’t realised it meant that editing with a commercial COI, violating the bright-line rule, could be a legal hazard in the EU, violating the Unfair Commercial Practices Directive. In Germany your competitors may be able to sue you for it, and in Britain you could be prosecuted. That’s as well as the media hanging you and your client out to dry.

(The Signpost article claims with no justification that the ruling “complicates” the issue of COI editing, rather than making it much simpler, i.e. damn well listen to us and don’t do it. Some PR people may not like it, but that’s rather a different thing.)

PR industry: “Our bad actions are Wikipedia’s fault.”

Fig. 1: Your client’s reputation when you get busted.

Yet another PR agency is blatantly busted doing the thing we patiently warn them against over and over, with the consequences we warn them of over and over.

The apparently-unanimous industry response, per PR Week: “It’s all Wikipedia’s fault, they should make it easier for us to spin.”

PRCA in particular appear to have turned their opinion 180° since June, when they heartily endorsed the CIPR/WMUK guidelines.

Guys, this really doesn’t help your case.

(CIPR have strongly dissented. PR Week didn’t get around to asking them.)

“it does get to the heart of the accuracy and lack of control of social media” – yeah, it’s accurate and you don’t control it. — Denny de la Haye

Update: Followup from PR Week. They’re also doing a print piece.

Please do not email me an infographic to run in this blog.

Have you ever wondered how some infographics seem to show up across the blogosphere all at once? It’s because the PR people who make these things actively promote them to bloggers.

I fully concur with Tom Morris’s excellent post on the matter, Infographics are porn without the happy ending. I’ll just steal the good bit, explaining why these things are actually seriously problematic, and why you’d have to be a bit of a cock to send one to anyone to do with an organisation like Wikimedia:

If you make an infographic, you are basically saying fuck you to blind people, fuck you to the Googlebot and often fuck you to people with colour-blindness. And you are definitely saying fuck you to people on slow connections. If you are paying £4 a megabyte to get data in Paris (yeah, I hate you too, Orange), putting an infographic where text could do the job isn’t just a giant fuck you but a waste of actual money. And by the time you notice, you can’t complain. If you are out in India and your only connection to the WWW is a phone we Westerners called shitty and threw away about three years ago, the infographic is completely inaccessible to you.

If I run a graphic here, it’ll be either because it’s pure decoration, because I think it’s important to getting the point across or because the image is the point. It will not be because five lines of text is best expressed in 200 obfuscated kilobytes.