I actually disagree with you. I think security and ease of use are mutually exclusive.
I can't think of a single counterexample where security did not have an effect on usability.
Now to be clear, I absolutely believe we should have a lot more security than we do, and we should be giving up usability to do it. But I think that it is necessary to give up usability.
1. HTTPS everywhere has zero impact on the usability of a website, but makes visiting those sites much more secure.
2. Apple's App Store model for installing applications improves security and is also easier to use for the average user.
3. 1Password with Touch/Face ID can actually be easier these days than entering emails and passwords manually.
4. And more…
Security doesn't always have an immediate effect on usability, and even when it does it isn't always negative.
You have to install a plugin on every browser and device. That's not user friendly. Also some sites break and you have to understand why.
> 2. Apple's App Store model for installing applications improves security and is also easier to use for the average user.
You have to have an account on the app store. You can't log into the same account on more than five machines. You have to put your password in every time you install an app.
That's all a lot harder than clicking on a link in an email.
> 3. 1Password with Touch/Face ID can actually be easier these days than entering emails and passwords manually.
You have to install 1password on all your devices. You have to mange your passwords with 1password. I'm an expert and even I have a hard time getting it to catch every update and every password field. 1password is a usability nightmare. Having the same password everywhere is a lot easier (and obviously less secure).
You're looking at one aspect after the initial setup and ignoring maintenance.
"Usability can be described as the capacity of a system to provide a condition for its users to perform the tasks safely, effectively, and efficiently while enjoying the experience." (https://en.wikipedia.org/wiki/Usability)
Notice the presence of "safely" there. You should be penalizing the alternative in each of these examples for the non-zero risk of data breaches and getting pwned. Then we can have a more productive debate about which cost is greater.
Again I’m not saying it’s bad, quite the opposite. I’m just saying we have to acknowledge it’s always a trade off.
First, does that always hold true?
Second—if we pretend it’s an axiom—careful and competent approach may allow to shift such degradation to aspects orthogonal to usability (instead of being perceivable by end user, failure modes could cause overhead, say, to maintenance or engineering teams).
> I can't think of a single counterexample where security did not have an effect on usability.
Sticking with the "mutually exclusive" criterion, wouldn't ssh vs. rsh serve as a counterexample? How about https vs. http?
Both replacements brought us greatly improved security, with no effect on usability in the vast majority of time I've spent with them.
Https is worse from a usability standpoint because of the need for trust stores and keeping them up to date. While it's mostly taken care of by the browser, it can still cause problems when you don't have anything in your truststore that can verify the other certificate.
Ssh vs rsh suffers a similar problem. If the certificate on the other side changes, you have to update your local config to allow the new certificate and will either be blocked or have warnings until you do so.
Pretty much with anything that requires certificates, usability is worse because you have to maintain the trust store in some way.
In both cases you mentioned the effects are very negligible and certainly worth the tradeoff, but strictly speaking, the usability is worse.
(Edit: remove many words)
That phenomenon gets stronger the less-known the truth is, because people will more readily look for reasons to reject a contrarian/minority view than a familiar one. It's important not to supply spurious such reasons, and that means articulating the truth (or what you believe is true) in a way that avoids gratuitous admixtures.
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
I'm going to take a break for 81 days now, unless I need to respond to something specifically about something I've written. We'll see what this place looks like then. I'll probably follow it with a longer break.
You've been at this longer than I have, so I don't feel like I have much to offer by way of insight, but let me try one thing. Any large, open internet forum has a serious upper bound on how good it can ever get. Misinformation and ignorance float in like open sewage. There's nothing we can do about that. It's frustrating. Actually it's crazy-making, because the more we all do to improve the community and make it better, the more it invites an influx of sewage, because sewage-spewers always want to spew in the highest-quality environment they can find, and there are zero barriers to entry here. (Lest I hurt anyone else's feelings, I should add that I'm not calling the other commenter a sewage-spewer, even if he may be wrong about the tradeoff between security and usability. Rather, this phenomenon is best described in the general case.)
Given that there's zero we can do to prevent this, the only agency we have is around how to react to it. There, I think we've learned something. Misinformation is not best combated by attacking the misinformed. One of my teachers once told me: "I did a Ph.D. in psychology, and here is what I learned from my Ph.D.: that punishment is not good for learning." Rather, misinformation is best combated by relating with the misinformed person while correcting the misinformation. How do you relate with a misinformed person? That's up for grabs. I don't think we've learned any formulas for that; the only thing I feel pretty sure of is that it can't be faked. Clearly, though, "you're full of shit" is not the way to get there.
Attacking the misinformed person puts their nervous system into flight-or-fight mode, which on the internet reduces to fight-mode. Worse, it fires a polarizing laser into the community, because while some readers will agree, others will be on the opposite side of the question, and still others will be neutral on the question itself but will freak out in response to the attack. Attack invites counterattack, and what gets lost in this process is the open state in which people are available for new information. Fight mode is like a muscle spasm, while receptivity requires relaxation.
The model underlying how we moderate HN doesn't have to do with politeness. I agree that enforcing politeness does not improve epistemic functioning; it just creates a fear of breaking the rules, and fear is a shut-down state. Meanwhile in other people it provokes a 'wtf? fuck politeness' reaction—and quite rightly.
The policy here isn't about politeness—nor 'civility', which is a word we stopped using a year or two ago for similar reasons. If I had to pick a word as of right now, I might pick 'openness'. We're interested in what practices of behavior are the ones that can encourage each other to stay in an open state—that is, a state in which we're able to hear each other and exchange interesting information.
The bottom line is that any large internet forum is going to be an 'echo chamber of vulgar misconceptions'—any attempt to prevent that is doomed—but we have the opportunity to be a somewhat interesting echo chamber of vulgar misconceptions, that is, a community which is able to respond interestingly, and the way to have that is to practice openness and the things which encourage openness in others.
Security and ease of use are not by default mutually exclusive. Depending on circumstances satisfying both may be a challenge, but it’s not a blanket impossibility and we could probably hold certain companies up to a standard that corresponds to their multi-billion valuations.
A virtual “slider” on which you can only dial up security by hurting usability is only a useful map, not the territory.