= Floats in AT Protocol =
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
Floats in ATProto - HackMD
Note: I was part of the IPLD team that created the specs/information that is linked from this FAQ.
hackmd.io
March 31, 2026 at 8:00 PM
= Floats in AT Protocol =
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
I am documenting datamodels in a "customProps" object because FU TypeScript.
March 28, 2025 at 4:58 AM
I am documenting datamodels in a "customProps" object because FU TypeScript.
perhaps...
the datamodels are pretty different so I don't think the protocols would ever merge
of course, a client may implement both & developers may build bridges between the two networks
the datamodels are pretty different so I don't think the protocols would ever merge
of course, a client may implement both & developers may build bridges between the two networks
May 5, 2023 at 7:58 PM
perhaps...
the datamodels are pretty different so I don't think the protocols would ever merge
of course, a client may implement both & developers may build bridges between the two networks
the datamodels are pretty different so I don't think the protocols would ever merge
of course, a client may implement both & developers may build bridges between the two networks
When we talked to Anthony Eden for Data in the Wild, he gave us a basic but essential tip on working with #datamodels
April 24, 2025 at 11:00 AM
When we talked to Anthony Eden for Data in the Wild, he gave us a basic but essential tip on working with #datamodels
#AI #Tech #WomenInSTEM #Humor #DouglasAdams #Bots #Robots #AIAssistants #DataModels #Bar #Coffee #Beer #BehaviouralPsychology #Data
February 14, 2025 at 11:00 PM
Our new #webinar on Thu, May 8th is waiting for your registration! Find out more about what #Propylon experts will be exploring during the webinar and register here:
event.webinarjam.com/register/58/...
#standards #machines #AI #XBRL #standardization #datamodels #SDOs #standardsbodies #thefuture
event.webinarjam.com/register/58/...
#standards #machines #AI #XBRL #standardization #datamodels #SDOs #standardsbodies #thefuture
April 17, 2025 at 11:57 AM
Our new #webinar on Thu, May 8th is waiting for your registration! Find out more about what #Propylon experts will be exploring during the webinar and register here:
event.webinarjam.com/register/58/...
#standards #machines #AI #XBRL #standardization #datamodels #SDOs #standardsbodies #thefuture
event.webinarjam.com/register/58/...
#standards #machines #AI #XBRL #standardization #datamodels #SDOs #standardsbodies #thefuture
This is my recommendation: use API to manage test dataset.
The 3 main reasons are:
- test teams don't have to know implementation detail of datamodels
- some events may be emitted with API instead of manual SQL sync
- dog fooding
The 3 main reasons are:
- test teams don't have to know implementation detail of datamodels
- some events may be emitted with API instead of manual SQL sync
- dog fooding
Orchestrating API calls is IMO the easiest solution especially on distributed architectures (where usually creating entities generates events that are required by other services etc).
Smthing like:
- a format for the input data (yaml or similar) in git
- a program that read the files and calls APIs
Smthing like:
- a format for the input data (yaml or similar) in git
- a program that read the files and calls APIs
May 31, 2025 at 11:44 PM
This is my recommendation: use API to manage test dataset.
The 3 main reasons are:
- test teams don't have to know implementation detail of datamodels
- some events may be emitted with API instead of manual SQL sync
- dog fooding
The 3 main reasons are:
- test teams don't have to know implementation detail of datamodels
- some events may be emitted with API instead of manual SQL sync
- dog fooding
This #MITResearch paper explores how bias in training datasets, a known issue, becomes even riskier with emerging #AI models. Their method improves outcomes for minority groups without compromising performance. A step forward for fairness in AI, but much work remains! #AIEthics
Data Debiasing with Datamodels (D3M): Improving Subgroup Robustness via Data Selection
Machine learning models can fail on subgroups that are underrepresented during training. While techniques such as dataset balancing can improve performance on underperforming groups, they require acce...
arxiv.org
December 15, 2024 at 7:35 PM
This #MITResearch paper explores how bias in training datasets, a known issue, becomes even riskier with emerging #AI models. Their method improves outcomes for minority groups without compromising performance. A step forward for fairness in AI, but much work remains! #AIEthics
Thanks to @vahiddousti.bsky.social for the opportunity to talk about large #DataModels in #PowerBI #SemanticModels 👍🏼
Check out the full video of our meeting yesterday with @biinsight.com about "Dealing with Large Data Models in Power BI".
Link: youtube.com/live/mqwutn2...
Link: youtube.com/live/mqwutn2...
Dealing with Large Data Models in Power BI
YouTube video by Data BI
youtube.com
January 29, 2025 at 5:36 AM
Thanks to @vahiddousti.bsky.social for the opportunity to talk about large #DataModels in #PowerBI #SemanticModels 👍🏼
Hey bluesky does anyone know how to do physical datamodels to help me out with my coursework?
March 19, 2024 at 8:51 PM
Hey bluesky does anyone know how to do physical datamodels to help me out with my coursework?
Shivin Dass, Alaa Khaddaj, Logan Engstrom, Aleksander Madry, Andrew Ilyas, Roberto Mart\'in-Mart\'in: DataMIL: Selecting Data for Robot Imitation Learning with Datamodels https://arxiv.org/abs/2505.09603 https://arxiv.org/pdf/2505.09603 https://arxiv.org/html/2505.09603
May 15, 2025 at 6:05 AM
Shivin Dass, Alaa Khaddaj, Logan Engstrom, Aleksander Madry, Andrew Ilyas, Roberto Mart\'in-Mart\'in: DataMIL: Selecting Data for Robot Imitation Learning with Datamodels https://arxiv.org/abs/2505.09603 https://arxiv.org/pdf/2505.09603 https://arxiv.org/html/2505.09603
ChatGPT , Claude , Copilot and Gemini walk into a bar ... open.substack.com/pub/madhatte...
#AI #Tech #WomenInSTEM #Humor #DouglasAdams #Bots #Robots #AIAssistants #DataModels #Bar #Coffee #Beer #BehaviouralPsychology #Data
#AI #Tech #WomenInSTEM #Humor #DouglasAdams #Bots #Robots #AIAssistants #DataModels #Bar #Coffee #Beer #BehaviouralPsychology #Data
ChatGPT , Claude , Copilot and Gemini walk into a bar ...
A dimly lit bar. A neon sign outside. No, it’s not Hotel California. It says Come as you are. The bartender walks over and greets them.”Hi! Welcome! Correct me if I am wrong , So you guys are like rob...
open.substack.com
January 15, 2025 at 5:50 PM
ChatGPT , Claude , Copilot and Gemini walk into a bar ... open.substack.com/pub/madhatte...
#AI #Tech #WomenInSTEM #Humor #DouglasAdams #Bots #Robots #AIAssistants #DataModels #Bar #Coffee #Beer #BehaviouralPsychology #Data
#AI #Tech #WomenInSTEM #Humor #DouglasAdams #Bots #Robots #AIAssistants #DataModels #Bar #Coffee #Beer #BehaviouralPsychology #Data
specialized policies, but selecting data naively may actually harm downstream performance. To address this, we introduce DataMIL, a policy-driven data selection framework built on the datamodels paradigm that reasons about data selection in an [3/8 of https://arxiv.org/abs/2505.09603v1]
May 15, 2025 at 6:05 AM
specialized policies, but selecting data naively may actually harm downstream performance. To address this, we introduce DataMIL, a policy-driven data selection framework built on the datamodels paradigm that reasons about data selection in an [3/8 of https://arxiv.org/abs/2505.09603v1]
Build #LowCode #EnterpriseApps using #NaturalLanguage! Create #APEX Apps with #GenDev: apexadb.oracle.com/ords/r/tech/...
Learn about APEX #AI Assistant to create custom SQL #DataModels + sample data.
Join @jaysonhanes.bsky.social next Tuesday!
#oracle #appdev #autonomousdatabase #adb #adw #atp
Learn about APEX #AI Assistant to create custom SQL #DataModels + sample data.
Join @jaysonhanes.bsky.social next Tuesday!
#oracle #appdev #autonomousdatabase #adb #adw #atp
May 30, 2025 at 6:23 PM
Build #LowCode #EnterpriseApps using #NaturalLanguage! Create #APEX Apps with #GenDev: apexadb.oracle.com/ords/r/tech/...
Learn about APEX #AI Assistant to create custom SQL #DataModels + sample data.
Join @jaysonhanes.bsky.social next Tuesday!
#oracle #appdev #autonomousdatabase #adb #adw #atp
Learn about APEX #AI Assistant to create custom SQL #DataModels + sample data.
Join @jaysonhanes.bsky.social next Tuesday!
#oracle #appdev #autonomousdatabase #adb #adw #atp
While it's always been possible to create a table from #PowerPivot-backed #DataModels, it has required some black magic and #DAX knowledge to make it happen. The DAX Table Monkey in our Monkey Tools #Excel add-in builds it for you with just a few clicks. 🐵 Learn more about it at mkyt.ca/daxtable.
May 13, 2026 at 4:36 PM
While it's always been possible to create a table from #PowerPivot-backed #DataModels, it has required some black magic and #DAX knowledge to make it happen. The DAX Table Monkey in our Monkey Tools #Excel add-in builds it for you with just a few clicks. 🐵 Learn more about it at mkyt.ca/daxtable.
Just spent the evening chatting with the digital devs behind some massive German VTT systems. The pain of migrating legacy PDFs into strict Foundry v13 DataModels is completely universal right now.
August 15, 2026 at 10:21 PM
Just spent the evening chatting with the digital devs behind some massive German VTT systems. The pain of migrating legacy PDFs into strict Foundry v13 DataModels is completely universal right now.
Python Dicts unter Kontrolle: Modelle statt Rohdaten in 2026
Ein praktischer 2026 Guide, um Python Dictionaries unter Kontrolle zu halten: rohe JSON Payloads in dataclasses, Pydantic Modelle, TypedDict Contracts und klare Mapping-Typen umwandeln.
#python #pydantic #datamodels
Ein praktischer 2026 Guide, um Python Dictionaries unter Kontrolle zu halten: rohe JSON Payloads in dataclasses, Pydantic Modelle, TypedDict Contracts und klare Mapping-Typen umwandeln.
#python #pydantic #datamodels
Python Dicts unter Kontrolle: Modelle statt Rohdaten in 2026
Ein praktischer 2026 Guide, um Python Dictionaries unter Kontrolle zu halten: rohe JSON Payloads in dataclasses, Pydantic Modelle, TypedDict Contracts und klare Mapping-Typen umwandeln.
www.alekseialeinikov.com
July 8, 2026 at 7:21 PM
Python Dicts unter Kontrolle: Modelle statt Rohdaten in 2026
Ein praktischer 2026 Guide, um Python Dictionaries unter Kontrolle zu halten: rohe JSON Payloads in dataclasses, Pydantic Modelle, TypedDict Contracts und klare Mapping-Typen umwandeln.
#python #pydantic #datamodels
Ein praktischer 2026 Guide, um Python Dictionaries unter Kontrolle zu halten: rohe JSON Payloads in dataclasses, Pydantic Modelle, TypedDict Contracts und klare Mapping-Typen umwandeln.
#python #pydantic #datamodels
Stop Passing Raw Dicts Around in Python: Safer Models for 2026
A practical 2026 guide to keeping Python dictionaries under control by converting raw JSON into dataclasses, Pydantic models, TypedDict contracts, and clear mapping types.
#python #pydantic #datamodels
A practical 2026 guide to keeping Python dictionaries under control by converting raw JSON into dataclasses, Pydantic models, TypedDict contracts, and clear mapping types.
#python #pydantic #datamodels
Stop Passing Raw Dicts Around in Python: Safer Models for 2026
A practical 2026 guide to keeping Python dictionaries under control by converting raw JSON into dataclasses, Pydantic models, TypedDict contracts, and clear mapping types.
www.alekseialeinikov.com
July 6, 2026 at 4:40 PM
Stop Passing Raw Dicts Around in Python: Safer Models for 2026
A practical 2026 guide to keeping Python dictionaries under control by converting raw JSON into dataclasses, Pydantic models, TypedDict contracts, and clear mapping types.
#python #pydantic #datamodels
A practical 2026 guide to keeping Python dictionaries under control by converting raw JSON into dataclasses, Pydantic models, TypedDict contracts, and clear mapping types.
#python #pydantic #datamodels
Shivin Dass, Alaa Khaddaj, Logan Engstrom, Aleksander Madry, Andrew Ilyas, Roberto Mart\'in-Mart\'in
DataMIL: Selecting Data for Robot Imitation Learning with Datamodels
https://arxiv.org/abs/2505.09603
DataMIL: Selecting Data for Robot Imitation Learning with Datamodels
https://arxiv.org/abs/2505.09603
May 15, 2025 at 5:07 AM
Shivin Dass, Alaa Khaddaj, Logan Engstrom, Aleksander Madry, Andrew Ilyas, Roberto Mart\'in-Mart\'in
DataMIL: Selecting Data for Robot Imitation Learning with Datamodels
https://arxiv.org/abs/2505.09603
DataMIL: Selecting Data for Robot Imitation Learning with Datamodels
https://arxiv.org/abs/2505.09603
If you build #DataModels and want to sharpen your DAX skills, this is where those "how do I actually make this work?" moments get answered. Access comes with my #SelfServiceBI (skw-t.com/ssbi) or Excel Fundamentals (skw-t.com/excelfun) Academies, or join the AKAs on their own at skw-t.com/aka.
Self Service BI Academy
Prepare to change your company's reporting game forever! In the Self Service BI Academy, Ken Puls will teach you how to use Excel, Power Query, and Power BI together to build robust Self Service BI…
skw-t.com
April 7, 2026 at 7:17 PM
If you build #DataModels and want to sharpen your DAX skills, this is where those "how do I actually make this work?" moments get answered. Access comes with my #SelfServiceBI (skw-t.com/ssbi) or Excel Fundamentals (skw-t.com/excelfun) Academies, or join the AKAs on their own at skw-t.com/aka.
I've gotten so used to fast big datamodels that I am quite amazed at how quick Prometheus runs into problems when adding a few labels to a metric in a relatively low traffic application. I wonder if that would improve with different storage underneath.
May 13, 2025 at 12:41 PM
I've gotten so used to fast big datamodels that I am quite amazed at how quick Prometheus runs into problems when adding a few labels to a metric in a relatively low traffic application. I wonder if that would improve with different storage underneath.
SQL and relational datamodels 🫶🏽
February 24, 2025 at 9:37 PM
SQL and relational datamodels 🫶🏽
Feed: "PyPI recent updates"
By: hello@kumo.ai on Friday, May 8, 2026
By: hello@kumo.ai on Friday, May 8, 2026
kumo-api 0.88.0
RESTful datamodels for Kumo AI
pypi.org
May 8, 2026 at 6:37 AM
Feed: "PyPI recent updates"
By: hello@kumo.ai on Friday, May 8, 2026
By: hello@kumo.ai on Friday, May 8, 2026
Thx for that.
There is more XML out there than you think, notably in "back-ends".
Legacy from the 00s.
See it as a Tribute to the Longevity of some systems, datamodels, designs.
Data Lives Longer....
There is more XML out there than you think, notably in "back-ends".
Legacy from the 00s.
See it as a Tribute to the Longevity of some systems, datamodels, designs.
Data Lives Longer....
October 31, 2025 at 9:05 AM
Thx for that.
There is more XML out there than you think, notably in "back-ends".
Legacy from the 00s.
See it as a Tribute to the Longevity of some systems, datamodels, designs.
Data Lives Longer....
There is more XML out there than you think, notably in "back-ends".
Legacy from the 00s.
See it as a Tribute to the Longevity of some systems, datamodels, designs.
Data Lives Longer....