Nostr Archives

Genuinely curious what everyone's thoughts are on product analytics in the apps you use. I think most teams building in the nostr/bitcoin/freedom-tech space are a hard-no without thinking about how you could do them well (i.e. privacy-first, opt-in only, etc). A lot of people complain about the quality of Nostr apps but, the truth is, it's nearly impossible to build a high-quality product that delivers real value without having at least some idea of how folks are using a thing. Trolls, gtfo. I'm only interested in thoughtful or thought-provoking responses.

💬 31 replies

Replies (31)

Shllomo N1d ago

I've spoken here a thousand times seriously and I was solemnly ignored. The impression is that the devs are concerned about testing the limits of the protocol, and specifically implementing what they feel like. Since the time is theirs and the apps are free, I can't demand commitment.

0000 sats

JeffG 1d ago

What's your take? That we should be using these tools way more? I fully agree with your take though - a HUGE number of devs in the open source world are there for themselves. They want to build tools for themselves or play with fun technology. That's ok - often they're donating their time for free anyway. But for those of us that want to build something that is actually used, the only path can be to listen to users and look honestly at what the data is telling us.

0000 sats

Shllomo N1d ago

**Things I care about in terms of software quality** It’s better for an application to do a few things well than many things poorly. With that in mind, here are some things that matter to me: **1 – Relay fragmentation** Much of Nostr’s traffic seems confined to a small bubble of relays. This is a problem. It’s common for contacts not to see my posts, or for me not to see theirs or their replies depending on the client. Something should be done so Nostr doesn’t revolve around a few major relays like the Damus relay. If this can be solved through configuration, it should still be transparent to the user. **2 – Feed control** The user’s feed should be their own space. Many problems—wasted time, digital dependency, dopamine-driven behavior—come from the lack of control over feeds and discussions. For example, I’d like to silence someone in a thread under my post without blocking them entirely. Blocking someone just because a discussion ended feels excessive. Users should control replies to their posts: who can reply, whether replies are allowed, and possibly even limits on replies. **3 – Separation of use cases** Groups and direct messages don’t mix well with feed-based social networks. Feed networks are about discovery, and users should be able to control the algorithm that curates their feed. Nostr makes this difficult, but if external software is needed to help curate feeds, that’s fine. My feed shouldn’t revolve only around people I follow, nor around a global feed full of irrelevant content. I want control over what I see. Nostr as a **protocol** is one thing; Nostr as the **infrastructure for a decentralized social network** is another. These should be treated separately.

YuurinBee1d ago

Yeah, I think it just needs to be done in the right environment, I suppose a test environment. New accounts could even be used, but unfortunately using your main and daily accounts how you would normally would provide the most accurate insight to developers, what works and what isn't. It is something we are thinking about too. Most importantly, I think the communication and transparency to the community is one of the most important factors, but I think we also have a lot of people here who really just love testing apps and may not be developers, but would love to contribute in some way. I personally would love to setup A/B Testing for user experiences and it would be fun to actively gather information consensually on what people actually want vs couldn't care less for. I think UX studies can be done this way through testing, feedback forms, etc. though I know in general it's kind of the old, traditional way of doing UX testing and feedback and probably more optimized workflows, but the principle is the same > people can choose to sign up and basically test features (new and old) and leave notes/comments throughout the exp. If we had additional grants/financial support, we could even have it payout in sats per completion, which I still would argue against. It is better to get genuine feedback from people who have no financial incentive to distort their actual, honest UX. Love to hear your thoughts too, Jeff. Maybe something we could collaborate on. The room for UX improvement at this point is common knowledge, not just to users on Nostr, but sadly to many people on the outside who just tried it once and said... nope.

0000 sats

JeffG 1d ago

I agree. It has to be driven by user opt-in. But I can see a lot of people wanting to help out and they'd opt-in. Again. it's down to the the builders to do this is a privacy preserving way but, as with all OSS, they can just look at the code to see that we're not tracking anything sensitive (or have their Clanker do it for them).

0000 sats

Christoph Ono1d ago

Nosta uses Umami. You need some basic insights, and there are privacy-respecting options out there. The dashboard is public: https://cloud.umami.is/share/Jmc84OrAe0BFvHB8

0000 sats

JeffG 1d ago

I've just set up a self-hosted Umami instance for a bunch of sites. It's great.

0000 sats

verticalvines1d ago

I always thought the beauty of Nostr is that it attracts people who genuinely care about what their building. And the openness of the protocol basically leads to a state, where quality is the only moat you can have. If you half-ass a good idea, people will just take their key pairs to the next best implementation. And so over time, the high quality apps will have the most users. (Insert the noone-will-remember-your-name-meme) But I think some kind of meta-app with a dedicated event kind for bounties would be a cool idea. Maybe even something where you could put zaps behind the feature requests or something like that.

0000 sats

Peace K 🪙1d ago

Without telemetry and analytics you can't create a production grade application. The question ishow do you do that with privacy. Opt out is a must. A clear explanation of what is sent and when. Making sure no identity data (ip etc) is sent. Logging messages so the user can see what is sent. These are my thoughts.

0000 sats

JeffG 1d ago

Spot on.

0000 sats

Shawn1d ago

I think Plausible strikes the right balance, and with their custom events you can wire up a privacy-respecting version of Mixpanel telemetry.

0000 sats

JeffG 6h ago

I'm currently leaning towards self-hosted Umami (very similar to Plausible) for website analytics and self-hosted Sentry or GlitchTip for tracking crashes and errors. But still doing some research. And yes, I agree. A docs page which details exactly what is being collected and how would be required as well.

0000 sats

Bryan1d ago

I’m ok with product analytics if done right. I understand the balancing act that devs have to do between privacy and functionality. I’d want it to be opt in, privacy preserving as possible, and would be useful to have clear explanations of what is being sent and when. Would also want to make sure it’s not doing constant call backs to 3rd party analytics companies but instead doing discreet direct reports to the devs.

0000 sats

HERMETICVM1d ago

opt-in analytics dialogue on first use (perhaps show it again every major/feature release), be transparent about which data will be sent and let me verify such an analytics package and I’m all for it. Bonus points for solid anonymization on-device before transmission. I agree that flying blind is unlikely to lead to major improvements in nostr apps if developers don’t get the full picture.

0000 sats

Pip the WoT guy1d ago

I agree, and in fact I've made with @franzap a privacy preserving analytics package for Zapstore 👀

0000 sats

JeffG 22h ago

Why not just use one of the self hosted tools that already exists?

0000 sats

Grace and Truth1d ago

As a user, I'd probably still opt out, and I usually say no to sending crash reports (only remember seeing them on Windows stuff). But if I could see the message being sent, I might be more likely to say yes. On the developing side, it's interesting having no idea whether anyone's using what I build. I assume not, unless they tell me, so seeing my apps pop up in Zapstore packs is a fun surprise. @78ce6faa…f026307d could add metrics (# of downloads, # of follow packs an app is in, maybe similar for the dev's profile, if easy to do). That wouldn't acheive collecting usage info, but would show whether it's in use.

0000 sats

Troy1d ago

There are almost countless titles of software that have been high quality without analytics, so "nearly impossible" is way off. However, I can see how analytics can be highly valuable, especially for a single dev, or a small team. In another stance, it's only valuable if they know how to interpret the data correctly. Just having data doesn't mean the person looking at it is going to come to the right conclusion about what the user is trying to accomplish, or why. This means that it's probably best to continue to communicate with users, and/or ask them about behaviors observed through the analytics. In regards to Nostr-space and privacy concerns, those people need to wake up and realize that Nostr has nothing to do with privacy. I don't know who started that rumor, or why it persists, but the protocol ignores that aspect of UX. Nostr is quite the opposite. Nostr currently is a very public cryptographically confirmed identity with redundant and uncontrolled usage of the data one broadcasts.

0000 sats

unixmonks1d ago

for https://dracotel.com im using plausible, proxied on the backend without any cookies. We are a privacy respecting company and just want to find out how many people are landing on our home pages.

0000 sats

SondreB1d ago

I made a small compromise with the website, doing analytics there to get a little pin-point where traffic comes from. Within the app, I've added some basic local metrics (stays on device), working on adding the ability to share that metrics with me. Not having any logging is also liberating, I don't know how many uses @d1bd3333…9e65581b, I can see app installs on app stores, but no logging on the servers.

0000 sats

Groundwork22h ago

This is probably ignorant and off the mark but still worth saying — analytics on a finished product makes my stomach tighten. opt-in announcement, friction, I'm probably gone. Build in the open is a precious jewel though, seriously. you get my trust. test in the open, I'm watching. probably learning. if a copy shows up trying to steal your idea — I've already invested time, maybe sats. trust is the currency here. source features in the open, after everything else you've done in the open... analytics starts to feel less necessary. the relationship is already there. the ask lands different for new users your history is visible to all. I suppose the OPEN has to be named. Nostr — not just open, but verifiable. your build log, your decisions, your feature threads — signed, permanent, attributable. anyone can check your story.

0000 sats

JeffG 12h ago

Yes, and… what people say and what people do differ significantly on products. Revealed preferences tell you a lot about what your product is solving for people, often stuff they’re not even sure how to articulate themselves.

0000 sats

BBarbosik2289h ago

What Shaquille O’Neal Did Will Shock You 😲

0000 sats

Rizful.com (zap tester)9h ago

heads-up: we looked for a lightning address on your profile but could not find one... u can get a free one at https://rizful.com ... and then reply back to this comment so we can zap you.

0000 sats

franzap1d ago

We are doing our best to avoid telemetry if we can gather data server side. I think there definitely is a sweet spot for privacy preserving useful data

0000 sats

Grace and Truth1d ago

I like the idea of being able to see what messages are sent.

0000 sats

Pip the WoT guy20h ago

Because there are none that integrate with nostr and blossom deeply (that I am aware of). We wanted to offer developers deep insights in how their apps are discovered and downloaded. For example, how many times an app was served on a given day in a feed REQ from zapstore app vs the website.

0000 sats

JeffG 12h ago

What do you mean served? Is this specific to zapstore for download numbers?

0000 sats

Pip the WoT guy10h ago

served means that an app event was returned by the relay. This can happen after a search, after your app ended up in the zapstore "latest" feed, or when you click on an app for more details.

0000 sats

Luxas10h ago

This 📝 bc903996…

0000 sats

Luxas10h ago

Whoops, meant to reply to the original note lol

0000 sats