Apparently my complaints lead this announcement:
https://discuss.logseq.com/t/why-logseq-ai-and-how-to-preserve-privacy
They are developing local "AI" features and they said they are focusing on that, but they just published the code of OpenAI integration on feat/ai-lab branch atm. #logseq
It's Western media propaganda, South Africa will welcome Putin just like many other African countries a few weeks ago:
Apparently my complaints lead this announcement:
https://discuss.logseq.com/t/why-logseq-ai-and-how-to-preserve-privacy
They are developing local "AI" features and they said they are focusing on that, but they just published the code of OpenAI integration on feat/ai-lab branch atm. #logseq
Ireland’s plan to criminalize “hate speech” takes a “guilty until proven innocent” approach
https://reclaimthenet.org/irelands-plan-to-criminalize-hate-speech?utm_source=fediverse
I'm not sure if you are replying to me but just in case when I said "they ignore feedback" I was of course not referring to posts here but to the Feature Requests category that they setup with votes.
Some ideas from their protocols are interesting, namely servers being relays only (Nostr) and DNS domains as handles (BlueSky).
I don't think the problem is billionaires but the fact that economic power easily become political power and grows in a positive feedback loop. Billionaires are a symptom but the problem is a pathological social system.
And a big part of this is people seeking for wealth, social status, visibility etc even at the lowest possible level of the pyramid.
I think a better advice is giving less importance to the narrative that everyone creates about themselves and give it to the really important things in life, including being morally upright and principled, even if no one will ever know, even if no one appreciates you, even if that means sacrificing wealth.
Telegram access is restored in Brazil after judge strikes down censorship order
https://reclaimthenet.org/telegram-access-is-restored-in-brazil?utm_source=fediverse
@wikimediaitalia come mai #Wikipedia ha cancellato senza fornire alcuna spiegazione la pagina relativa a #Byoblu , il primo e unico canale televisivo al mondo finanziato con donazioni dei cittadini?
Something that KDE software could provide independently from the models are some confirmation dialogs that summarize a task.
In the example above it would be a dialog that says "Move N PDFs into <folder>?" and it lists files with a conventient UI.
These dialogs could be used by any program, not specifically by a ML model, but they makes sense with ML to be sure that the model is interpreting correctly the prompt.
Something that KDE software could provide independently from the models are some confirmation dialogs that summarize a task.
In the example above it would be a dialog that says "Move N PDFs into <folder>?" and it lists files with a conventient UI.
These dialogs could be used by any program, not specifically by a ML model, but they makes sense with ML to be sure that the model is interpreting correctly the prompt.
It is clearly impossible when it comes to ChatGPT but there are many simple models that can run locally with a small impact on resources like this one, that could be useful for semantic search:
https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2
Other ideas are tasks like "find all the PDFs about Foo and copy them in a folder on the desktop". Things like this should be possible without big models like GPT4.
Yes but they are totally ignoring the feedback, here there are the details:
Criticizing China’s lockdowns are a “microaggression” and should be reported through online portal, UK university says
Matrix.org powers a federated annotation platform for the Web:
First of all, thank you for the summary.
About this part:
> I think this needs a bit of communication from the Team to confirm their intentions and stem any potential misinformation spreading.
I thought the team had a problem with communication in general so I didn't care much in the past but my complain about OpenAI integration has another point that is not really about communication:
They are not respecting the Feature Requests section on the forum **at all**. The request for OpenAI integration has been made recently and it is mentioned in this summary.
There are tons of requests that better fit Logseq core features. Some are hard to implement, some less, but they are more than one year old and with many votes.
For me this is not about communication, this is just taking the money and implement what they want.
?
The Fediverse is... federated i.e. decentralized but not distributed and it is based on Web domains (DNS) anyway, so it is only natural to use the original website as the source of information instead of mirror that same information using multiple bots just because RSS/Atom and AP uses different formats.
It's a minor release for a "pre-1.0" phase in which the product is still not considered complete.
Generally a major release can break compatibility. With 0.x.0 releases it is similar as in "assume the interfaces subject to change before the 1.0 release".
For example this is the WordPress site from the AcrivityPub plugin author: @pfefferle
No other server involved, but it needs to "speak" AP.
I am suggesting the same experience for the end user but behind the scene the Fediverse instances are reading the RSS feed of website and then presenting them as AP actors that only publish. If this will be made in a standard way, we would see each other's interactions on feeds.
#Mathematics #Macroeconomics #Engineering #ComputerScience #Programming
#Ecology #Environment #Democracy #Freedom #Equity #Liberal #Socialism #PostKeynesian
#Inkscape · #Gimp · #Blender · #VLC
#WordPress · #Mastodon · #PeerTube:
searchable