Oddbean

*deep sigh* I feel Semi's intent here is good, but the execution is bad. He is correct, that Onyx is unable to trend on Primal because of Primal's filters, right? This warning message may not show to Onyx, but that doesn't negate the fact that these terms do apply to her, right? Is it a dick move to create this fake image? Yes. Absolutely! Is it a dick move to remind people that this is actually happening? No. Again, I think the execution is bad here Let's fix that. Hey @miljan my man. If a user such as Onyx always uses the NSFW hashtag on their more risque images, would that help to remove them from being filtered from trending? Hmm. I'll answer this myself, probably not, because she could accidentally forget once and that would slip through the filter. So, that's not a good fix. What about ML? At some point in the future when time and budgets permits, would you consider using ML to tag and filter images? If the ML determines it's risque or NSFW image, don't trend it. I'm sure it's not hard to train a model to look for nudes or lots of skin. Thoughts? Lastly, no one has a perfect answer for every issue. We're all trying to make Nostr the best that we can. Some of us just build and communicate differently. I do wish that Semi spent a little more time building all that we've discussed instead of spending his energy on this. Anyways. Let's go us. Bullish on all of us. Fuck yeah Nostr. 🤙🏻🫂💜

Thanks for a reasonable and nuanced take as always Derek. 🫂💜 To answer your question, Primal users can choose whether they want to see NSFW content in their feeds and/or in the trending side bar. The default is that NSFW content shows in the feeds, but not in the trending side bar. This is an easy setting for users to flip. We are working on a more sophisticated solution where NSFW content will be categorized on a per-note bases (not per-user basis). @𝓞𝓷𝔂𝔁.𝓥𝓲𝔁𝓮𝓷 is a perfect example of an account that posts lots of interesting content with a few occasional NSFW notes. Our new system would make it possible for her regular content to trend by default. A lot of criticism we get can be summarized as "why isn't Primal perfect today?!". We are building and improving things every day. Everything we've done has always been transparent. Happy to engage in good faith, but Semi obviously hasn't been willing to do that. Faking this advisory to generate outrage is obviously over the top. Excusing his behaviour looks a bit weird to me tbh.

The Derek Beer Index is always accurate. 🍻🍻 You get a lot of criticism and shit and always handle it in a classy manner. 🫂🫂 I look forward to the future per note filter enhancements for trending. I knew you'd get around to it, given enough time. Thank you. 🫂🫂

I think when we spend the bulk of our time online here we tend to miss the big picture. We are a network of around 10,000 active users, on a GOOD day. The protocol isn’t perfect. The clients aren’t perfect. The overall UX is still messy and confusing. We’ve barely gotten started, we’re not even close to being ready for the world to show up. We have to give people room to try things, get them wrong, and try something else without being mean to each other in the process. That goes against everything we embody.

Being the author of slidestr.net AND posting some NSFW content on my photography profile, I know this issue from both angles. Public web clients need some kind of default filter. Using ML seems to be a good approach but we must also accept that this will never be solved perfectly. In a year from now we might discuss what is acceptable or find users try to game these systems. With more video content the classification will also become more complicated and costly. IMHO filter choices should be part of an onboarding process where a user selects their interests and could also decide on a NSFW filter explicitly. It would be better to mark/blur content than to hide accounts completely without the user knowing. If there ever is a „suggested accounts for #photography“ list, I‘d like to be on it and not be omitted by some default behavior.

Because this content warning isn't shown, it's a lie or fake. I could also argue that it's a dick move to not show this warning to a npub that is subjeced to these hidden terms. I think discussion and communication is the better path here. All that is happening here is trying to get a reaction. Just be adults and talk about it. That works wonders sometimes.

It’s a good and nuanced take you had here @Derek Ross and @miljan but doctoring this image is a dumb move that should be called out. We don’t want top-down protocol-level moderation, partly that’s why many of us are here. That means that when someone misbehaves it should be called out. And here, specifically, doctoring a picture to trash someone’s product is simply bad. I said this to @semisol privately and probably publicly too in the past; I don’t understand why he’s so fixated on Primal but nothing speaks louder than code and we have the opportunity of building our individual visions for nostr. Don’t get distracted, focus on building.

Do you dislike political satire cartoons? Semi was clearly expressing himself using art. This is not an attack on a person. It’s a critique of a website. If you don’t want your product critiqued, you shouldn’t build a public product. Generative AI will be much more malicious than this or the previous “attack”. Furthermore, I find it weak and off putting to cry every time someone critiques your product. Not the sign off quality leadership. Not that my personal opinion matters much there.

Sure. Define “malicious”. Define “campaign”. That’s a word salad of victim mentality thinking. Playing the victim card is a beta move. Claim it’s false and carry on. Don’t pontificate about how “attacked” you are. If a cat picture on the internet rattles your bones, wait until your product brings the ire of a government or state sponsored “attack”. Do better.