Jay_millerjay's Categories

Jay_millerjay's Authors

Latest Saves

Today's Twitter threads (a Twitter thread).

Inside: ADT insider threat; Billionaires think VR stops guillotines; Privacy Without Monopoly; and more!

Archived at: https://t.co/nu1HbReiEX

#Pluralistic

1/


This Wednesday, I'm giving a talk called "Technology, Self-Determination, and the Future of the Future" for the Purdue University CERIAS Program:

https://t.co/po5IivZyr4

2/


ADT insider threat: If you build it they will spy.

https://t.co/kJrmtu8L3S

3/


Billionaires think VR stops guillotines: TARP with tasps.

https://t.co/MIKwvsICkr

4/


Privacy Without Monopoly: Podcasting a reading of the latest EFF whitepaper.

https://t.co/R2sl75y4rb

5/
The #Missouri Department of Natural Resources has begun the rule-making process and are attempting to change the definition of groundwater so CAFOs can build more easily in Missouri. 

The public comment period for this rule change runs Feb 16th through March 25th.🧵

Please email the #Missouri Department of Natural Resources stating why you oppose removing "perched water tables" from the definition of "groundwater table".

If you submitted a comment in November please re-submit it now as previous comments will not be included for this round.

Email
[email protected]
Reference that you are writing in response to a proposed rule change to 10 CSR 20-8.300 Design of Concentrated Animal Feeding Operations.
Talking points follow ⬇️

The rule change is being done to provide a direct benefit to a private applicant (United Hog Systems) that has a permit application pending before the Department.
There is no lawful reason to justify the distinction for the Department to exclude perched water...

...from its definition of groundwater table in regards to CAFOs, while also including perched water in the definition at all other wastewater treatment facilities.
All water bodies in Missouri need to be protected by the Missouri DNR.
Day 14 #31DaysofML

🤔 How to pick the right #GoogleCloud #MachineLearning tool for your application?

Answer these questions
❓ What's your teams ML expertise?
❓ How much control/abstraction do you need?
❓ Would you like to handle the infrastructure components?

🧵 👇

created this pyramid to explain the idea.
As you move up the pyramid, less ML expertise is required, and you also don’t need to worry as much about the infrastructure behind your model.

To lear more watch this video 👉
https://t.co/EqJNDmTfhV

#31DaysofML 2/10


If you’re using Open source ML frameworks (#TensorFlow) to build the models, you get the flexibility of moving your workloads across different development & deployment environments. But, you need to manage all the infrastructure yourself for training & serving

#31DaysofML 3/10


Deep Learning VMs provide managed, click-to-deploy VMs for processing data & training the model
🔹 Popular ML frameworks pre-installed
🔹 Reduces the overhead of managing & allocating compute & storage required
🔹 But you figure out how you’ll serve those models

#31DaysofML 4/10


Kubeflow - OS project for deploying ML workloads on #Kubernetes
🔹 Helps configure a multi-step ML pipeline including pre-processing data, training & serving the model
🔹 Run it on-premise or on any cloud
🔹 You’ll still need to configure where it’s managed

#31DaysofML 5/10
the whole point of Dunks was you could go cop them at VIM whenever you wanted for $65. this shit is like having to enter a raffle to buy milk.


like seriously why not make a ton more of them if they're gonna be so sought-after? they land at outlets? so? nike still makes money off that.

the only reason to keep making them so limited is that they KNOW all that matters is the profit on the flip and if they were readily available FEWER people would want them, not more

the whole system is super broken, but it's just gonna go the way it goes, because at this point it all caters to the secondary market. the only reason Nike can sell Jordan 1s for $200 is because the people buying them can flip them for $500

adjusted for inflation, a $65 AJ1 in 1985 is like $160—and modern-day AJ1s are made from cheaper materials in factories staffed by cheaper workers. they don't HAVE to be $200 retail. but the secondary market nuked the whole concept of what sneakers are "worth"