the fact that these tools enable a developer to do all these things are an attribute of the tool and not the user imo. so i guess just look out for people who and learn quickly and ramp fast, nice to work with etc etc.
Absolutely not a crypto person but I've been following this team called noice.so, so far I think they've been doing the right thing in the space, pretty new and think they are operating at the right level of abstraction.
This is actually a good example of what I'm railing against.
Giving your users airdrops is almost always a perverse incentive, and trying to figure out how to onboard to Farcaster as "Step 1" in your product journey is extremely limiting.
So I was wondering of things like the Doris blogpost, this paper and sqlcoder are still relevant/what extra does this approach offer vs trying to build a over mcp?
You summarize the past convestion in this thread.
- Start with a overall summary in a single paragraph
- Then show a bullet pointed list of the most interesting illustrative quotes from the piece
- Then a bullet point list of the most unusual ideas
- provide a longer summary that covers points not included already
- Finally, Step by step/phase by phase understanding of the ideas discussed above
algolia has done the same for their discourse forum, moved everything to discord and removed all old posts.
Upon asking this was a response from the team:
"The Discourse content is no longer available. Much of it was 5+ years old and no longer reflected current SDKs and APIs. We're glad to help you here."
> I'm shocked so many companies use Discord for official purposes.
Going back in time you can replace it with: Discourse, forums, website, IRC.
New generations of devs / manager decide to use "the current tool" to connect with their users. Too bad they also think it a good idea to nuke the older channels.
This time, the "current tool" doesn't allow searching via the web. Most discord forums I'm on are basically black holes in which questions keep getting repeated.
Exactly, it doesn't even help the owners themselves, because people will keep asking the same things over and over again. It's not like a forum where you can easily search by topic.
It’s about the same as the desktop app. It doesn’t feel like a web app. The desktop app is better because it doesn’t feel out of place.
It’s not a bad UI/UX IMO, but it can take some getting used to. For the notifications I have to check several options such as “silence @everyone and @here”. Sometimes I find the updates annoying. But it’s among the best chat UIs I’ve tried.
It's definitely good UX for a chat app. It was good UX from the beginning, when they had a dark mode before a light mode and made voice chat not require installing anything (which was many peoples' first real-world exposure to WebRTC technology).
It's poor UX for a support forum where you want communication to be one-to-many as much as possible (to spread info to the largest number of people with the smallest amount of info producer effort as possible).
Those actually are bad UX, imo. In the servers I've seen implement them, they devolve into basically chat anyway. People still re-ask questions because the search functionality is the same as searching chat, which people weren't doing before. The layout still favors short messages over in-depth posts.
(I do find Discourse to be among the worst web forum software, and I can see similarities between Discourse and the Discord forum-style channels, so some of this may come down to personal preference. Discourse always felt too recency-biased and ephemeral to me compared to e.g. MyBB or phpBB, for example.)
Keep in mind also that it's still behind the walled garden of Discord's authentication and hosting. Imagine if you had to install the Tapatalk app to actually access any InvisionFree/ZetaBoards/Tapatalk/etc forums back in the day. It was annoying enough that mobile users were nagged to, but it would've been unthinkable to require desktop users to. (And being able to launch the app in your browser, while removing friction to "installation," isn't the same thing as actually being on the web-- search engines can't index Discord channels, Discord messages don't have human-readable URLs to share elsewhere, etc.)
Edit: Also, very relevantly to this thread, you can't just grab a webpage snapshot to archive a Discord channel like you could with forum threads. You've got to either take a screenshot (while dealing with scrolling) or scrape the data via the API, being careful not to trip bot protections or violate ToS, and then figure out how to present it separately.
I definitely understand it. I have some really old blog posts with tutorials for a deprecated version of my software and I get a lot of people complaining about things not working based on those. I've had to add disclaimers to the most popular ones.
Hey! It’s more like comparing apples to apple pie.
BGE-M3 is a fine-tuned embedding models. This means that they’ve taken a base language model, which was trained for just language modeling, then applied further fine-tuning to make it useful for a given application, in this case, retrieval.
ModernBERT is one step back earlier in the pipeline: it’s the language model that application-specific models such as M3 build on.
i don't know but i think the answer should some combination of google sheets/excel. any guides around that would be helpful. I think i would just want to use gsheet at this stage.
