Hacker News Reader

Comment by andybak | Hacker News Reader

Not enough information here to know if they made the right call.

I can imagine situations where this is to the user's benefit and the company could handle the data ethically.

Example: Situations where the user would be likely to have very unreliable connection and be on the page with the email field for some time with other data entered that they wouldn't want to lose.

It might seem a bit contrived but I've had similar scenarios in the past.

EDIT - as others mentioned localStorage is a better solution for this problem.

View on HN

I think there's a very reasonable general expectation on the web that your data isn't used until you've clicked submit/save/etc. I don't think that this practice could be claimed to have "informed consent", even if the privacy policy mentions it explicitly.

You're right there are places where this sort of thing can be useful, but they're rare, and could even be handled client-side in many cases.

    I think there's a very reasonable general expectation on the web that your data isn't used until you've clicked submit/save/etc

I definitely have this expectation and would be angry if it was violated.

I wonder: do non-engineers have this expectation? In 2021?

I've been involved with web development since the more or less earliest days. Back when websites were mostly static pages, and eventually spouted some forms here and there.

But, would a non-greybeard engineer have this expectation today now that UX is generally so different and "app-like" and anything you see in your web browser can basically be assumed to be in constant communication with the server(s)?

Please note that I don't think the current state of affairs is really an improvement. In fact, I more or less hate the "modern" web. I'm just curious about how user expectations have changed.

> Example: Situations where the user would be likely to have very unreliable connection and be on the page with the email field for some time with other data entered that they wouldn't want to lose.

Data entered in a form don't require an active connection. If submitting the form fails you can resubmit it.

Usually such issues start to appear when companies decide to implement their own custom UI from the ground up in a way that the result can only be described as undefined behaviour and things like input fields reset their state when you look at them the wrong way.

> Data entered in a form don't require an active connection.

Such an important property of the web. Server sends a description of what it wants, client renders and handles interactions. Things like server side rendering are anti-web.

Server-side rendering, streaming video, and such aren’t incompatible with internet philosophies in the slightest, just not aligned with the intent of the www. Apps, even the ones that aren’t merely web views, are more www-like in this regard.

That's why LocalStorage exists.

Possibly beneficial and useful features can still be dark patterns. Whether or not there is any potential user-value is irrelevant to the ethics of design decisions like these.

If you're dealing with user data you should never compromise a basic respect for that user's intent. Designing for implied consent where consent should be explicit (e.g. sharing personal data) is unambiguously a dark pattern.

I know of a situation where someone refused to do something maybe 1/4th as shady as this. It sounded semi-reasonable to me.

In this case, the boss simply assigned the task to someone else. Employee stayed employed.

Then save it to local storage

Then you can't send them reminder emails, can you? No, this was pretty sleazy, and quite probably in violation of the GDPR.

Reply on news.ycombinator.com