I'm Live streaming YouTube again More Freely Than Before
In this blog post, I discuss the rapidly evolving Bard user interface issues and how they compare to Bing's implementations. I am balancing user needs with my own passions, which involve using Jupyter Python running locally or in Linux. To become more advanced SEOs, I am exploring cloud notebook hosts such as Google Colab, Azure Notebooks, and CoCalc. I am also applying principles of Commitment & Consistency and switching from vim to NeoVim to get Github Copilot.
Mastering Cloud Notebook Hosts to Become an Advanced SEO
By Michael Levin
Tuesday, March 28, 2023
I am communicating to our content team the rapidly evolving Bard user interface issues, it trying to catch up with Bing that has set expectations by providing some very good implementations on providing citations and copy/paste links on the responses. Don’t want to derail the meeting, but do want to get in these breaking developments.
I have to balance what the Moz users need versus my passions. I am internally quite convinced that what the Moz users need is cheerleading and a proven formula on becoming slightly more advanced SEOs. And all my work is Jupyter Python running local on your machine, and under Linux in particular. That’s very advanced, no matter how easy Microsoft and myself are making it for people.
None-the-less, I have to dive into this stuff for myself as to not fall behind. Changes in the industry are so fast right now in particular that if I don’t do this recalibrating right now, it will be that much harder to catch up later.
I am basically tweaking my new baseline capabilities. It was expected, but still an awakening it happened so fast, that Python sandboxes were being provided for OpenAI ChatGPT plugins. Get through the first round of this process ASAP! Simplify!
The stuff I’m doing right now is way too advanced for MOZ content. The focus of anything I do has to meet our customers at their comfort level. However, I make the argument that life starts outside your comfort level. I am receiving advice that I use Google Colab as the onramp to Jupyter and I know where that’s coming from, but also I want to lead people to the powerful direction.
Once you’re on Colab or Azure Notebooks or Binder or any of the other of these vast array of hosted Notebook choices, there’s no going to the native install. They already basically spent their willingness to try something new, and they spent it on developing a new vendor/cloud dependency, that of hosted notebook providers. Present the options up front and get to the conclusion that you can do things with the local install that you just can’t do with the hosted versions. Those things include:
- Coming from an IP that’s not the cloud-provider’s
- Being able to change your IP with easy-to-use desktop VPN software
- Being able to do Browser automation with Playwright/Puppeteer
- Processing the data locally so you can double-click results into Excel
My first piece will probably be an example of the types of things you can do with Jupyter. Wet their appetite with some big wins. Dispel their fear and help them to take the first baby steps outside their comfort zones.
My next call is at 1:00 PM, and what I want to do after that call is to live-cast the rest of the day. And so what I should do is roughly script that live-casting session so I can keep jumping back over here to keep myself on-track. Okay, 1, 2, 3… 1?
Practice live-casting again. Do a quickie! Maybe interview Bing and Bard about the many ways to run Jupyter, providing advantages and disadvantages. And let people see me work here in vim live again. Yes! Wrap it up quickly.
BING: list ways to run Jupyter
There are several ways to run Jupyter notebooks. Here are some of them:
- Standard Jupyter servers: You can run one of the standard Jupyter notebooks servers by installing the server in your Python environment, and then running the server and connecting to it via a browser¹.
- IDE integration: You can execute notebooks via your Integrated Development Environment (IDE)¹.
- Hosted services: You can execute notebooks via hosted services¹.
- The command line: You can run Jupyter Notebook within a specific folder by typing
jupyter notebookin the command prompt³.
I hope this helps! Let me know if you have any other questions.
Source: Conversation with Bing, 3/28/2023(1) 4 ways to run Jupyter notebooks - wrighters.io. https://www.wrighters.io/4-ways-to-run-jupyter-notebooks/ Accessed 3/28/2023. (2) How to Launch Jupyter Notebook Quickly | by Benjamin Dornel | Towards …. https://towardsdatascience.com/how-to-launch-jupyter-notebook-quickly-26e500ad4560 Accessed 3/28/2023. (3) How to Use Jupyter Notebook in 2020: A Beginner’s Tutorial. https://www.dataquest.io/blog/jupyter-notebook-tutorial/ Accessed 3/28/2023. (4) Running the Notebook — Jupyter Documentation 4.1.1 … - Project Jupyter. https://docs.jupyter.org/en/latest/running.html Accessed 3/28/2023.
BARD: list ways to run Jupyter
Here are some ways to run Jupyter:
- Use the Jupyter Notebook App.
- Use a command-line interface.
- Use the Notebook Viewer.
- Use the Notebook Compiler.
The Jupyter Notebook App is the most user-friendly way to run Jupyter. It is a web-based application that allows you to create and edit Jupyter notebooks. To use the Jupyter Notebook App, you will need to install Jupyter Notebook and then navigate to the directory where you want to create your notebook. Once you are in the correct directory, you can create a new notebook by clicking on the “New Notebook” button.
The command-line interface is a more technical way to run Jupyter. It allows you to create and edit Jupyter notebooks from a terminal. To use the command-line interface, you will need to install Jupyter Notebook and then navigate to the directory where you want to create your notebook. Once you are in the correct directory, you can create a new notebook by typing the following command:
The Notebook Viewer is a tool that allows you to view Jupyter notebooks without having to edit them. To use the Notebook Viewer, you will need to have the Jupyter Notebook installed. Once you have installed the Jupyter Notebook, you can view a notebook by clicking on the “View” button.
The Notebook Compiler is a tool that allows you to compile Jupyter notebooks into Python scripts. To use the Notebook Compiler, you will need to have the Jupyter Notebook installed. Once you have installed the Jupyter Notebook, you can compile a notebook by clicking on the “Compile” button.
BING: List the cloud notebook host like Google Colab, Asure Notebook and others.
Here are some cloud notebook hosts similar to Google Colab and Azure Notebook:
- Kaggle: Kaggle provides free browser-based Jupyter Notebooks and GPUs⁵.
- Amazon SageMaker: Amazon SageMaker notebook runs on the Jupyter Notebook App¹.
- IBM DataPlatform Notebooks: IBM DataPlatform Notebooks is a cloud-based Jupyter notebook environment¹.
- Jupyter Notebook: You can run Jupyter Notebook in the cloud¹.
I hope this helps! Let me know if you have any other questions.
Source: Conversation with Bing, 3/28/2023(1) Best Google Colab Alternatives in 2021 - Paperspace Blog. https://blog.paperspace.com/best-google-colab-alternatives/ Accessed 3/28/2023. (2) 5 Alternatives To Google Colab For Data Scientists. https://analyticsindiamag.com/5-alternatives-to-google-colab-for-data-scientists/ Accessed 3/28/2023. (3) Azure Notebooks vs. Google CoLab from a Novice’s perspective. https://dev.to/arpitgogia/azure-notebooks-vs-google-colab-from-a-novices-perspective-3ijo Accessed 3/28/2023. (4) Google CoLab Vs Azure Notebooks: Which One Should You Prefer?. https://analyticsindiamag.com/google-colab-vs-azure-notebooks-which-one-should-you-prefer/ Accessed 3/28/2023. (5) Six easy ways to run your Jupyter Notebook in the cloud - Data School. https://www.dataschool.io/cloud-services-for-jupyter-notebook/ Accessed 3/28/2023.
Bard: List the cloud notebook host like Google Colab, Asure Notebook and others.
Here are some cloud notebook hosts:
- Google Colab: A free cloud service that provides access to Jupyter Notebooks and Google’s computing resources.
- Azure Notebooks: A cloud-based service that provides access to Jupyter Notebooks and Microsoft’s Azure computing resources.
- CoCalc: A cloud-based service that provides access to Jupyter Notebooks and a variety of programming languages and software tools.
- Kaggle Kernels: A cloud-based service that provides access to Jupyter Notebooks and Kaggle’s dataset library.
- Binder: A cloud-based service that provides access to Jupyter Notebooks and a variety of programming languages and software tools.
These services allow you to run Jupyter Notebooks in the cloud, which can be helpful if you need to access powerful computing resources or if you need to share your work with others.
They’re great when you want to do exactly what everyone else is doing in exactly the same way as everyone else without the advantages that the power of your desktop brings to the picture. That’s where competitive advantage resides. And it doesn’t really have to be difficult.
You don’t need to use vim like I’m doing here. Let’s separate those issues. I advocate and am passionate about a few different things. They’re not all required out of the starting gate.
There’s some really easy ways to get started.
I fear explaining how to do cloud Jupyter hosts, because you’re leaving all the real power on the table. But why not crawl before you walk? If I lay the issues out very clearly at the outset, I can help folk go from Colab to local Linux-based Jupyter later.
What am I doing?
I am applying the principles of Commitment & Consistency as laid out in the seminal marketing book Influence (a.k.a. The Art of Persuasion) by Robert Cialdini.
I’m likely switching from vim to NeoVim so I can get Github Co-Pilot in THIS environment.
BING: Who is Robert Cialdini?
Robert Cialdini is an American psychologist and academic who is considered to be the seminal expert in the rapidly expanding field of influence and persuasion. He is the Regents’ Professor Emeritus of Psychology and Marketing at Arizona State University and was a visiting professor of marketing, business and psychology at Stanford University². His research is focused on uncovering why people say “yes” to certain requests and how to apply these findings ethically in the business world³.
Source: Conversation with Bing, 3/28/2023
(1) Robert Cialdini - Wikipedia. https://en.wikipedia.org/wiki/Robert_Cialdini Accessed 3/28/2023. (2) Robert Cialdini (Psychologist Biography) | Practical Psychology. https://practicalpie.com/robert-cialdini/ Accessed 3/28/2023. (3) Robert Cialdini, Ph.D.. https://www.robertcialdinibf.com/ Accessed 3/28/2023.
BARD: Who is Robert Cialdini?
Robert Beno Cialdini is an American psychologist and academic. He is the Regents’ Professor Emeritus of Psychology and Marketing at Arizona State University and was a visiting professor of marketing, business and psychology at Stanford University.
He is best known for his 1984 book on persuasion and marketing, Influence: The Psychology of Persuasion. It was based on three “undercover” years applying for and training at used car dealerships, fund-raising organizations, and telemarketing firms to observe real-life situations of persuasion.
Cialdini’s book has been translated into 38 languages and has sold over 7 million copies. He has been elected to membership in the American Academy of Arts & Sciences and the National Academy of Sciences.
Cialdini’s work has been praised by academics and business leaders alike. He has been called “the godfather of influence” and “the most influential social scientist of our time.”
His work has been used by businesses to improve their sales and marketing strategies, and by individuals to improve their relationships and achieve their goals.
What am I doing (again)?
I am live-casting much more freely during the work day than I did before, because in my last capacity at ZD, all my deliverables are closely kept proprietary secrets. In my new capacity at ZD/MOZ, all my deliverables belong to you. And there are some good ones. This is all for my day-job, even though this first step right now is calibrating to the new world order under ChatBots that I must accommodate to, or be left behind. RESEARCH!
I believe in Linux, Python, vim & git as a sort of eternal tech platform. If you know it plus whatever, you’ll never go obsolete. You are future-proofed. I believe that so strongly that having to move off vim just makes my point.
As the world changes, your mastery of vim (muscle memory similar to driving) carries over to the new vi-based platform.