vec

  * Build your software to stream data directly from Client A's API to Client B's API without it ever being at rest on a machine you control.
  * If your users want reports based off their data, then see if you can fetch the relevant data from the client on demand, build the report email in memory, then forget the raw data.
  * Tell your clients you prefer to use their oauth provider of choice to authorize instead of storing usernames and password hashes yourself.
  * Key your access logs off of an (arbitrary and short lived) session id instead of a user id.

I use the term "code smell" in the same sense sense as "`if` statements are a code smell". It doesn't mean never to use them, it just means that solutions that don't use them tend to be consistently better than solutions that do. Solutions that store less (or no) data tend to be consistently more secure than solutions that do store more data. That doesn't mean those solutions always exist, just that one should prefer them when they do.

I am convinced, however, that it is a very useful heuristic. The platonic ideal of a perfectly secure database is a completely empty one. Every step away from that requires individual justification. That doesn't mean your database will be small, but it should mean the database is as small as possible.

I treat permanently storing user data (any data, for any reason) as my solution of last resort. That doesn't mean I don't use it. It just means I have to convince myself I don't have any better ideas before I do.

Having a database is a code smell.

If you absolutely have to have a database, then having a `users` table (or equivalent) is a code smell.

If you absolutely have to have a `users` table, having any columns in it other than `id`, `username`, and `password_hash` is a code smell.

...and so on.

Admittedly this isn't necessarily easy in a company setting. It's in direct conflict with what the sales & marketing people want, and unfortunately that means your job is sometimes to fight with the sales & marketing team. But it's not complicated.

Yeah, probably. If you're operating a dangerous piece of equipment the onus is on you to do it safely. Bikes are more dangerous than shoes, cars are more dangerous than bikes, and commercial vehicles are more dangerous than cars.

It's not actually that difficult to drive safely around cyclists. Go the speed limit, pay attention, and don't pass unless you can do so safely. Yes, this means sometimes you will go 15 in a 45mph zone for a couple of blocks. If someone finds that burdensome enough to risk killing someone else then I have no problem calling that criminal neglect.

Taxi/Uber/Lyft is, what, $5-15 per one-way trip? That's between $10 and $50 per day to commute. And that's before errands, weekend activities, and unexpected emergencies.

My spouse and I have a six-figure household income and I'd have a hard time finding an extra ~$10k/yr for transportation costs. It would be a complete nonstarter on minimum wage. Snarky or no, "be rich enough you don't have to worry about it" is a really shitty piece of "advice".

But predicting the past is much easier. We have a privileged position to see that, as implemented, Bitcoin is a pretty shitty currency (and a decent high risk investment vehicle). It's way too volatile, transactions are very high friction, and both of those problems look like they'll continue to get worse as Bitcoin continues to mature. More over, it's looking like none of this will actually avoid state capture now that a state actor (China) has decided they care.

No piece of software in human history has been perfect on day one. The real world always has a few surprises. But Bitcoin doesn't have a viable mechanism for change. No central authority and no privileged user means all but the most glaring bugs are unfixable.

There is, however, a degree of respect for that hazard, demonstrated in concrete safety practices, that can override that presumption.

I don't think Equifax has demonstrated that respect. To a layman's view, not many companies do. It's possible that the amount of respect necessary for a hazard as large as the one Equifax created is too onerous for a for-profit entity to realistically implement, but I don't know that there's a fundamental reason that sufficiently paranoid engineering practices couldn't make this moral.

I couldn't agree more with the rest of your comment, but I draw an almost opposite conclusion.

We should discourage people, good and otherwise, from running credit bureaus. Collecting and holding sensitive personal data on millions of people is inherently a mistake, and after the last few years of data breaches it's getting hard to argue that it's an innocent one.

Companies have proved time and again that they can't be trusted with huge datasets on the general public. They don't have much incentive to apply the level of paranoia necessary to actually protect data that valuable, and even when they do they rarely have the level of InfoSec skills and, more importantly, culture to actually pull it off. In practice we see 2 or 3 nines of reliability when we actually need about 5 nines.

We should stop waiting until leaks happen. Holding sensitive personal data on hundreds of millions of people ought to be congressional hearing level scandalous whether that data has leaked yet or not.

Automated defenses will attack North Korean troops. They could also attack lost North Korean children, South Korean military convoys moving to the new front 50 miles north a week after the bombing of Seoul in 2034, and hikers living in Unified Korea in 2067.

"Indiscriminate" doesn't necessarily mean "unjustified" or "illegal", but it's still enough of a marked category difference to be worth some separate ethical consideration.

[1] https://www.youtube.com/watch?v=Erp8IAUouus

What it would prevent is the illusion that providing my SSN on demand is in any way an indication that I am in fact the human being attached to that number and not some third party.

I don't see a problem with the IRS continuing to use it as-is. I would have no more incentive to lie about my SSN than I currently do about my legal name or my mailing address, for example. In fact, I might have a significant disincentive if I knew that my boss, my HR manager, the auditor at the IRS, and anyone else who cared could spot check whether it matched what I told them all my name was. Besides, the IRS has been dealing with endless attempts at tax fraud for a century and a half now. One more awkward but plausible avenue for cheating is likely to be background noise against their existing caseload.

I do agree that would probably be a de facto ban on the industry, or at least a radical restructuring thereof. My suspicion is that the industry is actually not a profitable enterprise once all the externalities are properly accounted for. But I'd be happy to be proved wrong, and I'm more than willing to give the industry a chance to figure out a model that works.

Edit: Completely off topic, but this is the nicest internet argument I've had in a long time. Thanks, scrumper, for being an excellent debating partner!

Assume a credit rating agency can lower a lender's default rate from X% to X'% but introduces an additional Y% of fraudulent loans. If X - X' > Y, that's a good deal for lenders no matter what the absolute size of Y is.

With air travel the people bearing the risk are the direct customers. If I'm uncomfortable with the safety record of airlines I can simply not fly. How do I opt out of the risks imposed by a credit rating agency?

The closer analogy, I think, would be if cargo planes were routinely crashing in populated areas. The air freight company and its clients may well be completely comfortable with their loss rate, but the unaffiliated third parties living underneath would have a pretty justifiable reason to complain.

> Truth is, nothing can be done safely. Not travel, not business. There is always a tradeoff between risk and benefit.

This is absolutely true, but as currently constructed credit agencies get to reap the benefits while pushing the risk off onto the general public. If we regulate to internalize that risk and the agencies still think it's a good risk/reward tradeoff then that's fine. If they no longer think it's a good tradeoff that's fine too.

> So you don't like the bureaux; well, nobody does really. But what is an alternative which allows lenders to compare applicants on a like-for-like basis with some degree of confidence?

There's several suggestions elsewhere in the thread, but I'm going to go out on a limb and suggest: nothing.

On a macroeconomic level, easy availability of credit increases average growth but also increases total risk, and therefore volatility. It's not apparent to me that that's a good trade.

On a microeconomic level, the benefits I receive from being able to get credit easily have to be weighed against both the suboptimal personal accounting I have to engage in to keep my credit score up and against the long tail risk that a third party can open a line of credit in my name. It's not apparent to me that that's a good trade either.

Maybe lenders should be much more conservative with who they loan to. Maybe they should price in a much higher default rate. Maybe individuals should seek a single line of credit from their bank of choice instead of expecting expenses to be financed individually. I don't know, I'm just some guy with 2 semesters of college-level economics. What I do know is that the status quo isn't sacred and the arguments explicitly in support of it seem to boil down to "growth is good and change is scary", which seems like really weak rationale for giving someone most of us don't even have a business relationship with essentially unilateral control of millions of people's financial fates.

I know this will never actually happen, but I sincerely wish the Social Security Administration would publish a complete official database of real name to SSN mappings. It wouldn't impede their use as Social Security Numbers, but it would make it extremely obvious how inappropriate they are as a proof of identity. Maybe then we'd finally be forced to come up with something a bit less insane.

It may be the case that once all the privacy related externalities are pushed back onto credit reporting agencies then their business model is no longer viable. I, for one, am fine with that outcome. If that's the case then credit reporting agencies were never a net good for society in the first place, and all the regulations are doing is exposing that truth.

This has always struck me as an odd anti-regulatory argument, though. "Governments shouldn't use their power to distort the market. Therefore regulations are inherently bad. Instead we should distort the market using a different branch of the government, but in a much more capricious and unpredictable manner."

Adding accountability to the general public via regulation may make them more bureaucratic. It may make them less. That's up to the invisible hand to decide, once their incentive structure matches their real world impact.

When a private company has a massive failure that impacts someone other than its customers, then the impacted parties have no recourse whatsoever.

You're not an Equifax customer. Banks and big businesses are, and they don't really have a strong incentive to punish this sort of behavior.

Weakening the formal power of government officials doesn't make that power dissolve, it just displaces it to somewhere even less visible and answerable to the public.

A system where only a handful of people have to wade through a bureaucratic mess with alarming regularity can get very dystopian very fast for those few.

PHP templates advertise, and do in fact provides, most of the same benefits as JSX. There are almost no new concepts to learn, the syntax is the same as the language proper, and developers have the full expressive power of that language at their disposal wherever they want it.

In practice, however, this seems to fail more often than not in a couple of ways. First, inexperienced developers (i.e. all developers at some point in their career) have a hard time resisting the urge to inline their business logic into their presentation logic. And second, the resulting templates are as often as not illegible, often requiring a reader to trace substantial chunks of tangentially related procedural logic in order to understand what will actually be rendered.

It's important to remember that PHP templates seemed like a good idea at the time. They did, and do, have real short-term benefits, and talented developers can leverage those benefits to great effect. But it turns out those benefits come with real long-term penalties, and besides, many developers are too green or too pressed for time or just don't care enough about the task at hand to wield that power consistently well enough to reliably reap the benefits in the first place. The PHP community has spent the last decade painfully crawling out of that tar pit; I'm not thrilled to see the JS community happily diving into it.

On the other hand, JSX is extremely difficult to read. I can glance at a relatively complex handlebars template I wrote two years ago and instantly understand what it renders. To do the same with a similar JSX component requires spending ten minutes to trace the control flow, understand all the helpers, and mentally re-inline all the loops and conditionals. It's conceptually simple, which is attractive, but in practice that seems to lead to noisy, nonlinear render functions that have little to no structural similarity to their final rendered DOM.

It's also worth noting that JSX is easy to pick up for javascript developers. It's next to impossible for the kind of design-focused frontend people who know HTML and CSS inside and out but only ever needed enough JS to write the occasional jQuery selector. If you're certain your project will never have any styling input from people who aren't first and foremost programmers then JSX might be a great fit, but the rest of us actually have to work with our teams.

* Yes, I know, JSX is technically not a template.

Well, but when's the last time you installed Windows or OSX on clean hardware? Especially one where Windows needed a third-party driver to read the disk? Neither are difficult, mind you, but it's not a task that's suitable for "grandma".

As an apples-to-apples counterpoint, I've been a full-time Linux user for a decade now, but I recently had occasion built a Windows gaming rig for the family. Getting from zero to first boot wasn't difficult, but I was shocked and dismayed at how nothing Just Worked. Instead, I had to go manually download drivers for half a dozen common pieces of hardware.

I'm not an expert, but I'm definitely not a novice, and while there might have been a short window in which Windows had hit peak compatibility, it seems that the numerous hoops I had to jump through to accommodate a mainstream video card and a simple software RAID may well indicate that the moment has passed.

We've been a multiethnic society since the first European boats landed and have redefined who counts as which race several dozen times since then. Even if there were some not insane standard for a "purebred" white person, it would be incredibly unlikely for any arbitrary bigot to happen to meet that standard.

On the one hand, I too would like our leaders to be as informed as possible, to have access to the most expert advisers available, and to have any tools that will aid them in understanding the choices they will have to make. Even the leaders I disagree with. Competence is a nonpartisan value.

On the other hand, advisers are only valuable if their advice is listened to. Past a certain point, one has to cope with the possibility that they aren't actually being called on to advise and inform, but merely to provide a sheen of respectability to whatever the leaders wanted to do in the first place. This is a gradient, not a bright line, but it's still a threshold that can be crossed.

I'd love to be corrected, but I've been paying fairly close attention and I can't think of a single issue, big or small, where this advisory council convinced the president to change his stance. If an advisory body can't do that, then what purpose does it serve?

I can kind of understand why Apple would want the entire world to only work with their proprietary connector, but for the life of me I still can't figure out why anyone else, especially auto manufacturers, went along with it. It's a horribly short-sighted decision for products like cars which regularly have a useful lifetime measured in decades.