people simply cannot resist making a mini language and encoding it into a string though, it's like catnip
November 7, 2026 at 1:59 PM
people simply cannot resist making a mini language and encoding it into a string though, it's like catnip
Looking for some weekend reading? We've got you covered. Our new mag, Desire Lines: Encoding Queer Digital Futures, is out! Read seven stories that imagine digital futures rooted in queer creativity, resistance, and possibility. www.eff.org/deeplinks/2...
Presenting 'Desire Lines: Encoding Queer Digital Futures'
What does the digital future look like? Who gets to imagine it, shape it, and make it real?Building a freer, more inclusive digital age requires not only challenging systems of surveillance and
www.eff.org
November 6, 2026 at 6:33 PM
Looking for some weekend reading? We've got you covered. Our new mag, Desire Lines: Encoding Queer Digital Futures, is out! Read seven stories that imagine digital futures rooted in queer creativity, resistance, and possibility. www.eff.org/deeplinks/2...
New form of Hasan unlocked (latest video had an encoding error)
November 7, 2026 at 6:06 AM
New form of Hasan unlocked (latest video had an encoding error)
Huggingface uses a really dumb fucking way of encoding arbitrary bytes in python `str`-s by encoding them as the codepoints 256 + . So "Ġ", is \u288 = 256 + 32, or an ascii space (SPC). Your string above is a long string of spaces. Details at: ai.stackexchange.com/a/48243
xkcd.com/1479/
xkcd.com/1479/
Troubleshooting
xkcd.com
November 4, 2026 at 2:41 PM
Huggingface uses a really dumb fucking way of encoding arbitrary bytes in python `str`-s by encoding them as the codepoints 256 + . So "Ġ", is \u288 = 256 + 32, or an ascii space (SPC). Your string above is a long string of spaces. Details at: ai.stackexchange.com/a/48243
xkcd.com/1479/
xkcd.com/1479/
I will pass it on after encoding!
November 5, 2026 at 12:58 PM
I will pass it on after encoding!
NVIDIA's SEED method streamlines hybrid language model inference, achieving up to 1.4x throughput gains while maintaining performance. By leveraging shallow encoding for deep memory access, it reduces computational and memory demands, pushing efficiency in LLMs. https://arxiv.org/abs/2611.06084
Shallow Encoding Emits Deep Memories
ArXiv link for Shallow Encoding Emits Deep Memories
arxiv.org
November 6, 2026 at 5:40 PM
NVIDIA's SEED method streamlines hybrid language model inference, achieving up to 1.4x throughput gains while maintaining performance. By leveraging shallow encoding for deep memory access, it reduces computational and memory demands, pushing efficiency in LLMs. https://arxiv.org/abs/2611.06084
Social vision expands into auditory cortices of people born deaf: evidence from naturalistic film and encoding models www.biorxiv.org/content/10.6... #deafness #neuroscience
Social vision expands into auditory cortices of people born deaf: evidence from naturalistic film and encoding models
Comparing deaf and hearing adults provides insight into the contribution of experience to function of auditory cortical networks. We used naturalistic fMRI and voxel-wise encoding to measure informati...
www.biorxiv.org
November 5, 2026 at 3:55 PM
Social vision expands into auditory cortices of people born deaf: evidence from naturalistic film and encoding models www.biorxiv.org/content/10.6... #deafness #neuroscience
The law, in its majestic equality, forbids Rust and C alike to leak memory, to dereference null, and to misinterpret inband error encoding
November 7, 2026 at 12:08 AM
The law, in its majestic equality, forbids Rust and C alike to leak memory, to dereference null, and to misinterpret inband error encoding
New preprint with @jordangunn7.bsky.social and @dtalmi.bsky.social! A computational model (eCMR) predicts emotion-enhanced memory better when adding trial-level LPP amplitude than when estimating preferential encoding from behaviour alone.
www.biorxiv.org/content/10.6...
www.biorxiv.org/content/10.6...
Contribution of the Late Positive Potential to Emotional Memory: A Joint Neuro-Cognitive Approach to Memory Prediction
Emotional events are better remembered than neutral events. The emotional Context Maintenance and Retrieval model (eCMR) explains this effect by simulating preferential encoding of emotional items, im...
www.biorxiv.org
November 6, 2026 at 10:47 AM
New preprint with @jordangunn7.bsky.social and @dtalmi.bsky.social! A computational model (eCMR) predicts emotion-enhanced memory better when adding trial-level LPP amplitude than when estimating preferential encoding from behaviour alone.
www.biorxiv.org/content/10.6...
www.biorxiv.org/content/10.6...
Careful with that. You're now legally required to change your pronouns every year at Daylight Savings Time (we're encoding the time change in the constitution). Remember the mnemonic: xe/xim forward, any/all back.
November 4, 2026 at 4:38 PM
Careful with that. You're now legally required to change your pronouns every year at Daylight Savings Time (we're encoding the time change in the constitution). Remember the mnemonic: xe/xim forward, any/all back.
text is a lossy and noisy encoding of language which is a lossy and noisy encoding of thoughts; codecs vary in quality
November 2, 2026 at 7:50 PM
text is a lossy and noisy encoding of language which is a lossy and noisy encoding of thoughts; codecs vary in quality
PDF is the stupidest possible file encoding. a windows folder of 6 large pdfs was 33MB but when i combined those pdf's into 1 it became 12MB. HOW
November 4, 2026 at 9:47 PM
PDF is the stupidest possible file encoding. a windows folder of 6 large pdfs was 33MB but when i combined those pdf's into 1 it became 12MB. HOW
ooooh i love the look of this ultra low complexity codec for embedded systems. nibble based encoding too... was gonna say fixed rate but i think the rle makes it not so. phoboslab.org/log/2026/11/...
PhobosLab
phoboslab.org
November 6, 2026 at 12:11 AM
ooooh i love the look of this ultra low complexity codec for embedded systems. nibble based encoding too... was gonna say fixed rate but i think the rle makes it not so. phoboslab.org/log/2026/11/...
Go 1.27 lands with stricter JSON
Go 1.27 `ships` stable new JSON APIs, `encoding/json/v2` and `encoding/json/jsontext`, resolving issues like invalid UTF-8 and duplicate names.
Basic correctness fixes after 15 years means Go moves at its own speed. 🐢
Go 1.27 `ships` stable new JSON APIs, `encoding/json/v2` and `encoding/json/jsontext`, resolving issues like invalid UTF-8 and duplicate names.
Basic correctness fixes after 15 years means Go moves at its own speed. 🐢
November 6, 2026 at 10:57 AM
Go 1.27 lands with stricter JSON
Go 1.27 `ships` stable new JSON APIs, `encoding/json/v2` and `encoding/json/jsontext`, resolving issues like invalid UTF-8 and duplicate names.
Basic correctness fixes after 15 years means Go moves at its own speed. 🐢
Go 1.27 `ships` stable new JSON APIs, `encoding/json/v2` and `encoding/json/jsontext`, resolving issues like invalid UTF-8 and duplicate names.
Basic correctness fixes after 15 years means Go moves at its own speed. 🐢
I'm glad the dude thanks me for checking out this video, which is to make sure the encoding and streaming properties and subtitles are properly functioning.
November 5, 2026 at 4:04 AM
I'm glad the dude thanks me for checking out this video, which is to make sure the encoding and streaming properties and subtitles are properly functioning.
Video - Invasive encoding
https://objkt.com/tokens/KT1UD9kcF4mZKfuhp4gQmNpZggdYJ9LfMUny/18?ref=tz1L2g6PLJJDbVNMKwZsB1G4U6A86GXeXayC
https://objkt.com/tokens/KT1UD9kcF4mZKfuhp4gQmNpZggdYJ9LfMUny/18?ref=tz1L2g6PLJJDbVNMKwZsB1G4U6A86GXeXayC
Invasive encoding
mp4 file depicting a portrait of a woman altered by encryption and invaded by coding. Work created on the basis of an image reworked with effects using video software.
objkt.com
November 3, 2026 at 3:00 PM
RubisCO-encoding methanotrophs elicit a coordinated transcriptional response to carbon dioxide limitation www.biorxiv.org/content/10.6... #jcampubs
November 5, 2026 at 5:44 PM
RubisCO-encoding methanotrophs elicit a coordinated transcriptional response to carbon dioxide limitation www.biorxiv.org/content/10.6... #jcampubs
📜 SCRIPTORIVM is live!
Transcribe medieval manuscripts by annotating directly on the text, with full Unicode & MUFI, and export TEI P5.
🤖 Built-in AI assistant for encoding & checks
🔜 Coming soon: HTR for chansonniers
👉 scriptorivm.app
#Medievalsky #TEI #DigitalHumanities
Transcribe medieval manuscripts by annotating directly on the text, with full Unicode & MUFI, and export TEI P5.
🤖 Built-in AI assistant for encoding & checks
🔜 Coming soon: HTR for chansonniers
👉 scriptorivm.app
#Medievalsky #TEI #DigitalHumanities
SCRIPTORIVM
scriptorivm.app
November 4, 2026 at 1:06 PM
📜 SCRIPTORIVM is live!
Transcribe medieval manuscripts by annotating directly on the text, with full Unicode & MUFI, and export TEI P5.
🤖 Built-in AI assistant for encoding & checks
🔜 Coming soon: HTR for chansonniers
👉 scriptorivm.app
#Medievalsky #TEI #DigitalHumanities
Transcribe medieval manuscripts by annotating directly on the text, with full Unicode & MUFI, and export TEI P5.
🤖 Built-in AI assistant for encoding & checks
🔜 Coming soon: HTR for chansonniers
👉 scriptorivm.app
#Medievalsky #TEI #DigitalHumanities
I'm looking at a docx with mixed Arabic and French (the teacher's messy class notes) and it looks completely different depending on which software I open it in. Text encoding was a mistake
November 5, 2026 at 10:46 PM
I'm looking at a docx with mixed Arabic and French (the teacher's messy class notes) and it looks completely different depending on which software I open it in. Text encoding was a mistake
I thought, I really should dig a little deeper into audio encoding. So I did.
Audio encoding - Martin Stellinga
Ever wonder how an audio file works? How music and speech can fit in bits
martinstellinga.com
November 6, 2026 at 1:45 PM
I thought, I really should dig a little deeper into audio encoding. So I did.
What is going on with this system? I can't see posts or even my profile page. For my profile I get: "Not Found
Invalid accept-encoding: """ For Discover I get: "Message from server: Invalid accept-encoding: """ What's up?
Invalid accept-encoding: """ For Discover I get: "Message from server: Invalid accept-encoding: """ What's up?
November 4, 2026 at 9:33 PM
What is going on with this system? I can't see posts or even my profile page. For my profile I get: "Not Found
Invalid accept-encoding: """ For Discover I get: "Message from server: Invalid accept-encoding: """ What's up?
Invalid accept-encoding: """ For Discover I get: "Message from server: Invalid accept-encoding: """ What's up?
Pare che Bluesky abbia deciso di escludere i poveracci che, come me, non hanno un computer nuovo con un sistema superiore a Windows 7. Da giorni, a ogni login da qualunque browser su Win7 il feed si blocca su questa notifica: Invalid accept-encoding: ""
Non potremo più accedere? (segue)
Non potremo più accedere? (segue)
November 5, 2026 at 7:45 AM
Pare che Bluesky abbia deciso di escludere i poveracci che, come me, non hanno un computer nuovo con un sistema superiore a Windows 7. Da giorni, a ogni login da qualunque browser su Win7 il feed si blocca su questa notifica: Invalid accept-encoding: ""
Non potremo più accedere? (segue)
Non potremo più accedere? (segue)
They were definitely pretty awesome at encoding info real time. Some of these surprise blowouts would be at 98%-99% while every outlet was still saying "lean D"
November 4, 2026 at 5:13 PM
They were definitely pretty awesome at encoding info real time. Some of these surprise blowouts would be at 98%-99% while every outlet was still saying "lean D"