open-source software is too often driven from a technology first perspective and not user-first perspective.

Very true … although, tbh, until very recently, I’d say that were the case with all software based ‘products’, both commercial as well as FOSS — and I’m not even sure I wouldn’t argue that the ‘until recently’ doesn’t really hold now either.

There’s a line though.

Take Signal, for instance.

I understand the principle that, if you want uptake on a product that is intended to protect people’s privacy then ‘good enough and appealing’ is what you want — because ‘perfect but unappealing’ won’t achieve your goal of empowering people … because they won’t use, no matter how good it is.

On the other hand, the more features they introduce to persuade users to make use of it in preference to WhatsApp and other (less certain offerings) … gifs, stickers, typing indicators, link previews, a ‘quick emoji reply’ popup, an ‘allow for anyone’ option on sealed messages, … the greater the chance of some bug or other slipping in, leaving it open to exploit.

Edward de Bono coined the acronym EBNE (Excellent But Not Enough) .. a reworking of the concept of ‘necessary but not sufficient.’

However, I feel the corollary of NEBE (Not Ecellent But Enough) also holds — there’s something to be said for Ford’s (alleged) “any color, so long as it’s black” maxim.

So, IMO, Signal is in danger of going too far in the direction of pandering to user desires, rendering the original purpose of the app (a genuinely secure and private messaging solution) diminished.

I think it’s important to appreciate your (and Moxie Marlinspike’s) observation of the necessity to make tech meet the needs of the user rather than vice versa … but I think it needs to be balanced with a judicious amount of “No, the customer is not always right — sometimes they’re just an asshole.”