Connection Problem S03E15: The internet is regional stacks
Thanks for the overwhelmingly positive feedback to receive this newsletter at the beginning of the weekend rather than the end. So it shall be, schedule permitting.
×
×
Hindsight is 20/20
I never had the pleasure to meet John Perry Barlow in person, but I consider him one of the great internet thinkers, if not a flawless one. I had never considered this before reading this piece on Slate, but when he wrote “A Declaration of the Independence of Cyberspace” in 1996, he may have shaped the development of the web tremendously but with some omissions that we pay the price for today: The Incomplete Vision of John Perry Barlow.
×
New ThingsCon
Tiny bit of self-promotion: We relaunched the website for ThingsCon. That is all.
×
Tone deaf, with blind spots
Meanwhile in cryptoland, a crypto conf in Hamburg announced Alice Weidel, who is part of AfD's leadership, as a speaker. (AfD, for those not following German politics, is a recent right wing populist movement and party in the style of the US Tea Party and Trump's MAGA movement.) Weidel was labeled as an "economist and blockchain entrepreneur", her political label was conspicuously absent.
(Someone on Twitter mansplained to me that actually this is just lack of nuance in finding this despicable, because really the community just has to work out its Overton Window. Which tellingly is a theory about what political candidates can get away with saying publicly, not what's okay to say. Who knows, there's a chance I might have misunderstood his Twitter replies because of a language barrier.)
It appears they removed Weidel from their website, and possible from their speaker list. However, as someone who runs and curates conferences from time to time I promise you don't put someone up on the speaker list without at least a cursory check of their bios.
This should go without saying, but of course it doesn't:
The apolitical is highly political. If any person or organization tells you they are apolitical, make a run for it. What they're really saying is that they don't care, and don't take responsibility for the things they're putting out into the world.
×
The internet is regional stacks
I like how increasingly the big tech companies aren't just referenced as GAFA (Google, Apple, Facebook, Amazon), with the occasional M for Microsoft appended pitifully at the end. It appears instead that the focus in the Western discussion is opening up to a more global perspective, which in times of a balkanizing internet is direly needed. I see more and more variants of GAFA/BAT popping up on my radar: GAFA(M) plus the Chinese tech giants Baidu, Alibaba, Tencent. We'll see how the exact spelling will work itself out over time. GAFABAT? BATGAFA? FATAGAB?
(By the way, where did GAFA start? I first encountered in in French politics, but surely they weren't the first?)
×
OMGDPR
The brilliant and kind Chris Adams is instigating a meetup to figure out what the GDPR means for all of us who run stuff on or with the web. It's called OMGDPR. I'm so jealous of the punny name, I can't even describe it.Well played, sir!
×
Not so smart, city!
(1)
There Is No Such Thing as a Smart City. Bruce Sterling writes in The Atlantic and, as so often, puts his finger right there on the pulse:
"If you look at where the money goes (always a good idea), it’s not clear that the “smart city” is really about digitizing cities. Smart cities are a generational civil war within an urban world that’s already digitized. It’s the process of the new big-money, post-internet crowd, Google, Apple, Facebook, Amazon, Microsoft et al., disrupting your uncle’s industrial computer companies, the old-school machinery guys who ran the city infrastructures, Honeywell, IBM, General Electric. It’s a land grab for the command and control systems that were mostly already there."
(2)
Speaking of smart stack cities, the German city of Duisburg signs smart city partnership with Huawei to work together on smart city development. What could possibly go wrong? The memorandum of understanding "provides a framework for further discussion in which the two partners will commit to the city’s priorities of expanding the Wi-Fi network, enhancing traffic management, and improving e-government cloud solutions for the city’s services to the general public." It's unfathomable to me how cheaply city administrations are handing over the keys to their infrastructure.
(3)
The Atlantic covers the plans of Google subsidiary Sidewalk Labs (essentially their smart city division) to build a good chunk of Toronto "from the internet up". And there's a lot of interesting stuff and a lot of very questionable stuff in there, especially as far as governance is concerned.
When reading these kind of reports, the very first thing I wonder: What will happen to this infrastructure and the people who've come to rely on it ten years after the initial funding has run out? Because it will run out and new funding won't just magically manifest itself. And then what. Insert your own cyberpunk dystopian image here.
×
Decentralized, compartmentalized
Chris Dixon writes about the importance of decentralization to move to the next stage of the internet (in his words stage 3, as opposed to the current, highly centralized stage 2). I think I agree with this sentiment—the current state of affairs certainly seems a little sub-prime.
That said, Dixon says the way there leads via blockchains; which may or may not be true.
Which brings me to a third aspect—and one with which I violently disagree. Going from his rock solid arguments about the inherent risks and weaknesses of high levels of centralization, and his call for greater decentralization, Dixon comes to the conclusion that the financial incentives of cryptocurrencies are the carrot to get developers et al on board. This transactional model, tied directly to financial payoff, seems risky. What's more, it seems banal and tired and outdated.
The incentive should be, I don't know, bigger. Think along the lines of "here's a network that you can build the future and your lives on if you just help maintaining it a little." Is that not incentive enough? Is it too vague? Offering people pocket change for cleaning up seems to sad for anything we can refer to as "stage 3".
×
Facing the computer age
The archives of Design Quarterly are available online as free PDF downloads. "Very much like the children who are caught between the old math and the new, today’s designers have to face the computer age, turn away from security of the familiar and learn to adapt the new methods." Peter Seitz, 1966.
What a gem!
×
Face value
I just learned that Veronica Belmont produces a podcast with (for? by?) Mozilla called IRL, and this episode on facial recognition and biometric data is ace.
×
Malicious AI
This report on The Malicious Use of Artificial Intelligence, building on "a 2 day workshop held in Oxford, UK" (not many details available, but the participant list seems legit at first glance), explores the potential for malicious uses of AI, and how to tackle/prevent them: "Less attention has historically been paid to the ways in which artificial intelligence can be used maliciously. This report surveys the landscape of potential security threats from malicious uses of artificial intelligence technologies, and proposes ways to better forecast, prevent, and mitigate these threats."
Something beautiful I found in this report:
"Increasingly realistic synthetic faces generated by variations on Generative Adversarial Networks (GANs). In order, the images are from papers by Goodfellow et al. (2014), Radford et al. (2015), Liu and Tuzel (2016), and Karras et al. (2017)."
×
Edgy AI
In her newsletter, Stacey Higginbotham discusses AI at the edge (which is industry speak for on the device as opposed to in the cloud), where AI is currently predominantly happening. For example, Amazon & Google have traditionally been very cloud-focused: That's to be expected in internet companies. (It's even reflected in the thinking of Google's smart city unit, Sidewalk Labs, which designs cities to be built "from the internet up".) But there are strong arguments to be made to put AI closer to the edge:
Ai at the edge could have "tremendous impact on privacy and the speed of improvement in connected devices. If a machine can learn without sending data to the cloud, then that data could stay resident on the device itself, under user control."
Because what doesn't have to go to the cloud doesn't have to be under anyone else's control. And for many contexts that's a very good thing indeed. (I know some smart people at Google and presumably the other big tech companies are working on that, too; different schools of thought exist within each of these organizations.) I, for one, look forward to seeing more happen at the edge.
×
Mostly white male AI
One of the topics at the forefront of socially conscious folks in the current AI and machine learning debate is that AI can reinforce bias through hard-to-spot issues like biased training data. So it's all the more disappointing to see Udacity's Artificial Intelligence and Data Industry Advisory Board to be an all male, mostly white male line-up. It would appear hard to find a group less diverse than this.
As part of due diligence, I'd like to add that I found this image on Twitter and when I checked out the official announcement the image wasn't there; the list of names appears to check out though. It seems like the image was removed and replaced by a placeholder photo of a server rack (not joking).
×
Creepy AI
Boston Dynamics specifically designs their robots to look disturbing, don't they?
×
Not AI, still creepy
A map of the world after four degrees of warming. So friggin' depressing.
×
Testing Open Leadership Map Use Cases
Mozilla has been working on an analytical framework called the Open Leadership Map. It's a set of self-assessment tools and guidelines to help organizations work in the open. As someone who's been working very much in the open, this comes pretty natural to me, but there were a ton of pointers I found very useful when I ran through some of these simple tests. For example, when looking at ThingsCon, we should get better at building stronger systems for accountability and for empowering local stakeholders. Two thumbs up.
A note by the way, even though I've been working more closely with Mozilla and this coincides with me mentioning more Mozilla projects: I'm not on a mission to promote Mozilla (even though I do like and promote some of their initiatives), and I'm not paid to do marketing on their behalf (but have been paid in the past to do research, and may continue this research) - it's just that as I learn more about their work, projects like this are surfaced to me in ways they weren't before, so I'm getting a better overview of the many things going on there.
(Mozilla, I love you, but your online presences make it really hard to get that kind of overview.)
×
RIP, Lucia R.
×
Something nice to end on
Image: David Nadlinger - University of Oxford. (Source, via Kottke)
Look at this: "a photo of a single strontium atom suspended in an electric field taken by David Nadlinger". Because f*** yeah, science.
×
As always, a shout out to tinyletter.com/pbihr or a forward is always appreciated!
×
Have a wonderful weekend.
Yours truly,
Peter