Episode Two Hundred and Four: “Everybody”; The Webb Scale; Algorithms 

by danhon

0.0 Sitrep

8:36pm on Tuesday, March 17, 2015 “St. Patty’s Day” in America, which is the day all branded social media accounts turn green (and orange, I suppose) and make attempts to be culturally relevant.

1.0 “Everybody”

Via a non-work-related Slack team (how many of you are doing that? Have a shoot-the-shit, non-work-related Slack team? I’m interested, so do drop me a note), news of Leap (“Your daily commute. Redesigned”)[1].

Leap – from looking at their website – is the latest attempt of what I’ll loosely call “the tech industry” or “startups” to “disrupt” the transportation industry. The job to be done that Leap is trying to disrupt, for lack of a better term, is “[taking] the hassle out of getting to work.”

Their offering is quite simple: one “express” route (which appears to be the same as a Muni route), a simple timetable (7-10am and 5-8pm on weekdays) and “$6 for a single ride ($5 if you buy in bulk)”. Or, says their website, you can “use pre-tax commuter benefits and *ride for less than $4*” (their emphasis).

The bit that feels symptomatic of a general trend in how what’s broadly seen as “the tech industry” – and more specifically, the bad, non-inclusive bits of “the tech industry” is specifically where Leap says:


Use your iPhone or Android to board. Or if you don’t have a smartphone, you can print a paper pass.

I just want to stop here and say a) technology, in general, can be and has been shown to be a democratising and empowering force. Yay for technology! But technology gets a bad reputation when it says “everybody” and doesn’t actually mean everybody, *especially* in a place like the Bay Area, where there are already inflamed tensions about integrating into an existing community.

Because here’s the thing: Leap isn’t for everybody. It’s for commuters. It’s for “getting to work”. It’s not for going to the doctor or dropping your kids off at daycare or trying to get to the vet or picking up groceries or getting downtown to sign up for benefits or meeting a court date.

(Also: take a look at the imagery used on the Leap site. Look at the kind of people shown. Think about the kid of jobs that those people have. Leap isn’t just for *all* commuters, it’s implicitly for a specific *set* of commuters: the kind of people who aren’t, say, service workers. You know: people like you and me who read this newsletter and work in Business Town[2].)

I had this same issue back in episode 119[3], where Bridj (whose website has since been updated), another startup in the “making [public?] transportation better” and if you don’t want to read the actual long screed, you can just tl;dr it as “when you say everyone, you better fucking well mean *everyone*”.

Otherwise – a phrase that I’m learning from the new job – the optics just don’t look good.

The bigger issue, of course, and the one germane to San Francisco is that this is a bunch of capital going in to fix a specific problem for a small class of people, and that public transit in San Francisco is just generally not in a good state and that maybe the entire system needs to be invested in so that it works for *everyone* and so that not insignificant amounts of capital are wasted on short-term solutions for small groups of people.

In other words: Hey, if anyone from Leap is reading? I’m calling you out. If You Don’t Mean Everyone, Don’t Say Everyone.

[1] Leap – Your daily commute. Redesigned.
[2] BusinessTown
[3] Episode One Hundred and Nineteen: Bridj Laziness; Deciding In Public; Terrible Tools : Things That Have Caught My Attention

2.0 The Webb Scale

Smart fellow Matt Webb wrote a few days back about a richter scale for system outages[1] – I had originally wrote “network outages” here, but systems is much better because “systems” is bigger and more descriptive than “networks”, and systems comprise networks, but anyway.

Webb’s thoughts are around quantifying the effect of various interconnected (and frequently unanticipated connections) systems falling over in various ways. He suggests, for example:

Less than 2.0. Not distinguishable from normal network noise, like a call dropping, webpage not loading, or a computer crash.

and at the other end,

8.0. Major network freeze, can be recovered with time or reboot; major human impact. Examples include the 2008 credit crunch where bank lending gridlock precipitated the global financial crisis; power network outage major enough to require black start; the Icelandic volcano that grounded European flights for 6.5 days.

which, mmm, is such a crunchy set of ideas it just makes you want to eat his brains. But anyway, this stuck in my head because a few days later (ie: today) I tweeted that Google Apps Is Down is the new Dog Ate My Homework.

And, related: this *is* happening more and more, and there will be more black swans. I feel a bit weird and doom-saying honestly, but over there in that not-work-Slack-team, we’re busy talking about Yet Another Vulnerability in OpenSSL, another critical but unloved part of internet infrastructure, and on the drive home from the pub, the EyeSight system in our car (that is the stereoscopic camera system that does lane detection and automatic, rubber-band cruise-control and emergency stopping) turned itself off *for no reason*, and then turned itself back on *for no reason*, but at the same time as it turned back on, the Automatic[2] dongle did its happy beep which means that the connection between the car’s OBD-II port and one of our iPhones over BTLE was active, and that just screams *there is a bit of software in this car that has a bug in it and something just restarted and fuck knows what it is and I bet it can’t be fixed by taking the car into service* and god I don’t know what to do anymore.

At least my toaster doesn’t do this. Yet.

[1] A Richter scale for outages (12 Mar., 2015, at Interconnected)
[2] Automatic: An Auto Accessory to Make You a Smarter Driver

3.0 Algorithms

A photo taken by Paul Mison of a Charles Schwab bus shelter ad that read:

Say hello to your personal investing algorithm.
Introducing Schwab Intelligent Portfolios.
intelligent.schwab.com [1]

This means: algorithms are something that computers do, not a series of instructions that can be carried out by any information processing platform, including humans. Algorithms are now something that people want, because hey, those high frequency trading guys make money, and they use algorithms, so I want some of that. It is not clear, from cursory reading of the intelligent.schwab.com service, what such Intelligent Portfolios are doing (but it is a nifty responsively designed site with those little circle button things that fill in whilst you scroll down the page, so, yay?)

[1] Paul Mison on Instagram: “algorithm”

9:10pm. And tomorrow, I think I’ll write about shoes that talk too much.