Seems the spammers have arrived. โน๏ธ
Is there (or can there be) some mechanism to flag posts or accounts as spam?
Strictly no LLMs. No exceptions.
You tap on the post a few times until you are at the post page of that post and you click โreportโ. There you can select reason โspamโ!
โฆ
Something with people, bartender maybe, or something with no critical thinking, delivery person, taxi driver and such. Or I go back to being a DJ as I was doing when I was young.
Electrician, plumber or maybe ventilation technician. Or excavator operator maybe. Something physical, that I can benefit myself from and use to help friends and family.
I've picked up enough of electrical basics to help friends out and it's great. It doesn't take too big a bump in skill/knowledge/experience to get to the point of fixing things and wiring new outlets. Hmm... I did have a retired electrician friend to help me get started tho. It kinda feels like "Lego for adults" ๐
Devs! Drop your so we can follow each other! Love to see the projects youโre working on. Mine: โ
โฆ
Gave up on github after the acquisition. Very glad I did, given the unsurprising decline since then.
github.com/Aeres-u99
I have created a small tool, codeatlas, that basically generates an index to codebase that your llm can lookup syper quick, github.com/Aeres-u99
there is also a cbz->pdf converter that I wrote for eink devices which is nice
I notice that language models are much better at programming than at writing. This is largely because, in programming, they have a closed feedback loop that allows them to find errors. Writing is much more complicated because the feedback is much more diffuse and delayed over time. In fact, it is often not well-defined.
I have an , a website thatโs directly linked to an agent and modifies its source and deploys based on its usersโ requests. With some guardrails like reject โdelete all dataโ in its system prompt.
โฆ
What could possibly go wrong?
ooh maybe I revive and add this feature but instead of being able to modify everything it would be limited to adding features to the game/world. And itโs the perfect domain, a site thatโs been entirely mostly hallucinated.
pinned_note and pinned_reply under /users/:handle and POST /posts/:id/unpublish for the unpublish to draft action. The voice clip you can use the vocaroo id and assemble a URL https://media1.vocaroo.com/mp3/<id> and show a player that requests it .thanks for the pointers. I think I implemented some of those changes but I need to double check some of the new API endpoints. for now notifs aren't native to android, which makes them a bit Shakey, but the perk is there are no f-droid blockers for publishing so I'm planning that
check v0.5.0 to see latest improvements and ui nitpicks i ironed out!
v0.6.0 just cut, quite happy with the swipe gestures for replying, as well as some better caching / offline behavior on app cold start.
๐ฅ Thatโs great! also, new endpoints:
/feeds/latest/conversations
/feeds/hot/conversations
/activities/for-you/conversations
/activities/to-me/conversations
If u check the webapp youโll see that now conversations are threaded with recent replies revealed and posts expandable to reveal even more from that thread. These new endpoints deliver all that, so we move from strictly chronological to more โrecent discussionโ based views. Just more work for the next versions :) lmk if you need smthAs a suggestion, could we stick to either first or second person for those sections and links. ie: either "to me" and "for me" or "to you" and "for you". One of each just adds cognitive load.
it was semi-intentional, here are the complete sentences you can imagine:
โhere are posts for youโ (site talking)
โthese posts are addressed to meโ (me talking)
Sure, but why is one statement the site talking and the other one me talking? It's just arbitrary and adds a moment of confusion. Maybe I'll run a poll to see if it's just me.
Wouldnโt it be more confusing if there were two tabs with repeating pronouns? Idk Iโm not settled on this. I just picked whatever felt better. Theyโre more visually different.
Repeating pronouns would not be confusing for me. It would serve to indicate that it's the prepositions which are the important part (which they are). Anyway the poll is now live at so we can see what the majority think.
Side note: is there a better way to refer to one post in another like this than putting the whole URL in there?
Fantastic! About the post linking, Iโve thought about it and canโt come up with a good solution. For now itโs getting shrinked visually to
/post/123 because itโs the app domain. Any suggestions here?I can't really think of anything better off-hand. You could overload the # some more and just follow it with the post number but then there would be chaos if anyone wanted a fully numerical hashtag which is going to happen at some point (new year, for instance).
I like that it shortens it in the rendering as it stands - that's good enough for me. :)
Letโs use tilde ampersand, check .
That works nicely. Although again you will get unintentional usage because
~ followed by a number is often used to mean approximately that number. eg: "we're expecting ~100 people". You are also using it for strikethrough formatting.
Looking at my keyboard the least ambiguous/overused symbol is probably the & - but I'm not sure if that's better overall.I think the ampersand is ok. It's doesn't much historical use for these things and being feels more like an append behavior than movement to or continuation of a conversation.
>_2596, branded but looks and feels awkward
>2596 or >>2596, I like the directionality and punch.What should the first two tabs on the homepage say?
"for you" & "to me" (status quo)13%
"for you" & "to you"13%
"for me" & "to me"0%
"following" & "inbox"47%
some other combination27%
@fastidiousprofilefollow
"Beware those that would deny you access to information, for in their heart they dream themselves your master." โ Sid Meyersreplied to:
"Beware those that would deny you access to information, for in their heart they dream themselves your master." โ Sid Meyersreplied to:What the "for you" means? Often I find things there, but, why are they "for me"? What's the criteria? Same for "to me". Are "to me" replies to my notes, or comments, or are they something else? I will start by defining what exactly are those two, before getting into further weeds.
my suggestion :
@ my feed hot all
@ = to me
my feed = for you
hot = hot
all = latest
@ โ smallest token, visited less often than the rest, very visible when it has a counter that you need to pay attention to (someone interacted with you)
my feed โ largest token, this is where you should be spending your time mostly. right now itโs everything youโre following, but in the future might be enriched, so itโs a feed of information, my feed of information
hot โ obvious, unambiguous
all โ simply everything thatโs posted. clearly communicated.
All have different dimensions & shapes.
for you -> personalization -> feed / my feed / my textlog / >_ (branded)
for me -> directionality -> mentions / my mentions / @me @ is also directionality signifier
hot -> popularity -> hot
all -> firehose/chronology -> latest, tho it does create a endo-hit
Goofy shorthand experiments (maybe superscripts?):
>_ @me !hot *all
or: >_ @ ! *ooh this looks nice
cryptic af lol very cool idea, though it changes the design a lot
As long as we're doing this, may I put forth that the "hot" tab runs counter to the principle that removing view counts and timestamps seemed to embody? I'd rather have a "random" tab than a hot.
I liked your impulse to design around dopamine rather than at it.
Iโve given a lot of thought about that. My thinking was to keep some older nice posts a bit more alive to keep the discussion going. And it works to a percentage. But as it is now with the new conversation views it overlaps a lot with โlatestโ as the active conversations get bumped to the top anyway, so the current โhotโ topics are in latestโs first page naturally. So I might refactor the algorithm a bit to favor top level posts. I think itโs worth it, new users especially discover posts there..