Scott Knight's Blog

How to Keep Your Mac Awake Using Terminal

Scott Knight — Thu, 01 May 2025 18:54:22 -0500

Recently I was running some tasks on my Macbook Pro I wanted to prevent the machine from sleeping. A quick DuckDuckGo search tipped me off to the caffeinate command that’s built into the Terminal in macOS. If this is a well-known thing in macOS, I’ve never come across it so I thought I’d share here in case anyone else would find it useful.

If you have some reason to prevent your machine running macOS from sleeping, caffeinate gives you a few options. For basic usage, here’s how you use it:

Step 1: Open Terminal

If you’re not sure how to do this, see here (it’s one of the Applications that comes with macOS).

Step 2: Type `caffeinate` and hit Enter

This will prevent your Mac from sleeping as long as you don’t close the lid or exit Terminal. Once you close Terminal, hit Ctrl+C, or close the lid on your machine, the process will exit and normal sleep behavior will resume.

Now, here’s the really handy part—caffeinate takes arguments, meaning that if you add an option after the command, you can change how long it will run or set it to wait for a specific process or command to complete before exiting.

caffeinate -i prevents sleeping indefinitely (prevents idle sleeping)
caffeinate -u -t 3600 prevents sleeping for 1 hour (the -u flag resets the last time the user was active, and -t 3600 defines the time limit in seconds, equivalent to an hour)
caffeinate -s <command> prevents sleeping until a command finishes running. Just replace <command> with your desired command
caffeinate -w <pid> prevents sleeping until a running process completes, provided you define the PID of the process

Resources

For further reading, here’s the original article I found as well as the actual man page for the command.

Homelab Update

Scott Knight — Sat, 27 Apr 2024 09:00:28 -0500

Some time last year I wrote about using a cheap Lenovo Thinkcentre I bought as a lab machine to run my homelab server. Initially I used it to try out a new (to me) Linux distro, which I installed straight onto the machine. I enjoyed test driving it, but it never replaced my primary system.

So, this week I decided to repurpose the ThinkCentre as a general lab server to locally host open-source applications and dust off some of my brain’s cobwebs around server configs and networking.

This YouTube video from TechHut provides a really easy way to get set up if you’re interested in doing something similar.

For now I’m running Ubuntu Server LTS with Docker and Portainer for container management.

On top of this, I installed Cosmos as a dashboard which looks nice and has a really decent OOTB feature set. So far, I dig it.

A tiny view of the Cosmos dashboard. The default color scheme reminds me of Mardi Gras.

Future plans include trying out some new apps (this sure beats more software subscriptions) and adding a NAS to the lab, so if you have any recommendations send me an email reply using the button below.

Keeping Analog Alive

Scott Knight — Wed, 17 Apr 2024 15:28:21 -0500

I’ve written here about my career in the software industry, but I’m also a big fan of analog technology. Digital stuff can be fun, but once you’re ingrained in the tech world for so long and understand how it all (or, at least, most of it) works, it can lose a bit of its magic. For that reason I’ve always been drawn to the machines and devices of yesteryear.

One way I express that interest is through my collection of vintage and antique typewriters. I have a handful of them which I’ve picked up and restored to working condition over the years. I use them for all sorts of things like recording recipes, notes, letters, greeting cards; the uses are endless. The tactile action and sound of the clacking keys as you type makes using them fun, while simultaneously allowing you to slow down, think, and appreciate the process of writing.

A greatly helpful resource I’ve used to clean, restore, and maintain my typewriters is Joe Van Cleave’s YouTube channel. Joe’s been making videos on vintage and analog tech since the 1980’s and has hundreds of videos on just about anything to do with writing, typewriters (their usage, maintenance, collecting), and camera systems. Unfortunately, the number of people knowledgable about these machines is dwindling and it is increasingly hard to find good resources on restoring them. Thankfully, however, there are still folks like Joe who are doing their best to pass their knowledge along.

I admire Joe’s consistency in creating useful content around something he’s passionate about, and his bona fide wealth of knowledge shines throughout each video he posts. If you’re into vintage/antique writing technology or just want to learn something interesting, check out Joe’s videos.

Joe’s Channel: https://www.youtube.com/@Joe_VanCleave

P.S. It’s been a busy couple of months so I haven’t been posting as regularly. I’m hoping to get back to posting more frequently soon. C’est la vie.

Once

Scott Knight — Fri, 01 Dec 2023 15:19:52 -0600

Since I started using Hey for my personal email a while back, I’ve enjoyed following the stuff coming out of the 37signals shop. They have a unique and refreshing take on business, work philosophy, and building tech.

I saw that they’re releasing a new line of products soon over at Once.com. I checked their site and, for now at least, it’s just a placeholder landing page. However, I read through the message and wanted to share an excerpt here because I appreciate the sentiment behind the project:

Something happened to business software.

You used to pay for it once, install it, and run it. Whether on someone’s computer, or a server for >everyone, it felt like you owned it. And you did.

Today, most software is a service. Not owned, but rented. Buying it enters you into a perpetual landlord–tenant agreement. Every month you pay for essentially the same thing you had last month. And if you stop paying, the software stops working. Boom, you’re evicted.

For nearly two decades, the SaaS model benefitted landlords handsomely. With routine prayers — and payers — to the Church of Recurring Revenue, valuations shot to the moon on the backs of businesses subscribed at luxury prices for commodity services they had little control over.

Add up your SaaS subscriptions last year. You should own that shit by now.

SaaS still makes sense for many products, but its grip will slip. Installation and administration used to be hopelessly complicated, but self–hosting tech is simpler now and vastly improved. Plus, IT departments are hungry to run their own IT again, tired of being subservient to Big Tech’s reign clouds.

Once upon a time you owned what you paid for, you controlled what you depended on, and your privacy and security were your own business. We think it’s that time again.

And you know what? I think they have a valid point.

Subscription-based software has really gotten out of hand, and for some essential products I don’t necessarily think users should have to pay for access. Subscriptions have their place, and the benefit really depends on whether the team is regularly and reliably shipping new features to justify an ongoing perpetual cost to the user. Otherwise, to the 37signals team’s point, you’re really just “renting” the software. Sometimes, if a software product is embedded enough into your workflow, the subscription cost almost feels like a ransom to avoid losing your data or work.

Either way, I’m glad to see someone striving to rebalance things in terms of pricing models and a philosophy that’s user-centric. And I look forward to seeing what the Once product suite will have to offer in the near future.

How to Create Custom Terminal Commands in MacOS

Scott Knight — Mon, 06 Nov 2023 16:30:27 -0600

I recently wrote here about some of my favorite personal automations. When updating my a finance automation script recently, I decided I wanted a way to quickly run the automation anywhere. I typically have a terminal open, so I thought a custom terminal command would be the best way to fire off my scripts when I need them.

Unfortunately, the resources are slim out there regarding how to configure custom terminal commands in MacOS. Using an article I dug up (linked below), I was able to figure it out.

Here’s how I was able to get it set it up in a few steps:

1. Create a file to store your custom commands

Open a terminal window and navigate to the Home directory using the command:

cd ~

Now, create the custom commands file using the command:

touch .my_custom_commands.sh

2. Add custom commands

Open the file in Visual Studio Code or your preferred text editor. Note that the dot we added before the filename creates a hidden file, so be sure you’re able to locate and open it properly.

In the new file, add your custom commands using the following structure and save:

#!/bin/bash

function my_function_name() {
    python ~/path/to/my/script.py
}

3. Set up file permissions, if needed

The new file may generate a “permission denied” message. If so, you should be able to fix this by running the following command:

chmod +x .my_custom_commands.sh

4. Make it available in the terminal

If you try to run the command (the name of the function) you just created, you won’t be able to quite yet. We’ll need to let the terminal know to load your commands to each new session.

Open the rc file:

open ~/.zshrc

After the last line, add:

source ~/.my_custom_commands.sh

Save the file, then exit and start a new terminal.

Once you have all this set up, you can add a new custom command any time by simply adding a new function to your commands file. Just note that you’ll need to start in a new terminal session whenever you edit the file so your new command gets loaded up.

Aside from automations, this is also useful for repetitive strings of commands. Here’s one I created to set up virtual environments for new Python projects, so rather than memorizing the extra commands I can just enter create_env and the rest is covered:

function create_env() {
    python -m venv env
    source env/bin/activate
}

As always, feel free to use this as a starting point and make it your own.

(Source: Medium)

How to Automate Anything

Scott Knight — Mon, 30 Oct 2023 17:57:51 -0500

After some interest from my last post where I discussed some of my personal automations, I’d like to expand on the topic of personal automation a bit and discuss how you can automate (mostly) the boring or tedious stuff in your life!

I say “mostly” because while with enough time and resources, you could automate anything with a repeatable process. However, considering our main goal here is to save time, you’ll have to evaluate which things are pragmatically worth expending your time and effort to automate.

Here’s my general process:

1. Choose what you want to automate

An obvious first step, I know. But as I just mentioned, here’s where you sift through your daily life to see where your pain points are. Make a short list of the most tedious things you’d rather not have to do, and then pick out the things that don’t require human skill as part of the process. That is to say, the best candidates for automation are those with a repeatable process.

2. Record the steps

Now that you’ve chosen a task that can be reasonably automated, write out the steps to give yourself an idea of what the automation needs to accomplish. If the task is complex, that’s okay for now. Just write down the general steps as a working outline.

3. Identify the most painful step

From the list you just created, look through the process to find the most pressing bottleneck or pain point. This is likely the best place to start automating. You may have thought we were going to automate the entire process all at once, but in my opinion it’s best to start small and iterate your automation until it captures the overall process over time.

This approach has several advantages:

The effort of automating is minimized, since you are only tackling one step at a time.
You get to test each step to make sure it works fully and accurately before moving onto the next.
The work is spread out over time, so you don’t spend one large chunk of time automating an entire complex process (unless you really, really want to).

4. Choose your platform

Now, for the task(s) you’ve just targeted for automation, you’ll want to consider which platform best suits the triggers/actions involved. For example, my simple “Blog to Social Media” automation is done using IFTTT since there is no complex logic involved. That automation is simply a trigger (a new post is published to my RSS Feed) and subsequent actions (a new post is published to Mastodon and X/Twitter) are executed.

A more complex example might be my financial automation. I’m taking a CSV of my checking account transactions, parsing through that and performing some matching to categorize each transaction, and then saving the result to an Excel file for my records. For this, I opted to write a custom Python script which handles the more sophisticated logic.

The point is, I choose a platform that’s appropriate for the complexity level of the task at hand. It’s not a one-size-fits-all sort of thing.

5. Automate it

You’ve chosen your most painful step to automate, as well as your platform. From here, create the script as required to solve the problem at hand. You may need to consider how often to run the automation or what triggers need to be configured so that the automation is run as needed.

Depending on the scenario, you might only need to run your automation manually (on a case-by-case basis, as my finance automation) or via some other action like my RSS-to-Social script.

As a tip: if you’re opting for a no-code platform like IFTTT or Make to set things up, it still helps to have some surface-level familiarity with APIs and how to use them. This will come in handy for cases in which you need to interact with some service that doesn’t have a built-in set of building blocks.

6. Repeat steps 3-5 until the whole process is automated

Keep working through your Step 3 list until the whole process is done. This will help you automate the most difficult parts of a process, significantly cutting down on your pain points even from the first iteration.

7. Deploy

Set up your automation using the appropriate triggers, as mentioned previously. If you’re using an online service like IFTTT, this is probably already taken care of. If you’ve opted for something more custom like a Python script, you’ll need to consider how to deploy that so it’ll run when you need it.

Good solutions for this could be Netlify, Amazon Web Services, or one of any number of similar cloud hosting services. Find one that fits your expertise and budget.

8. Monitor, iterate, and improve

While this is all called “automation”, you’d be sorely missing a step by just setting everything and forgetting about it. You’ll want to monitor things over time to make sure that, once you’re “done”, everything keeps running as intended.

APIs and services change their available functions, the timing of tasks is prone to change at various points, and many other things may change over time from your initial configuration.

In addition, you may think of ways to improve your automation or use the extra time you’ve saved to add more steps that wouldn’t have been possible when you were handling things manually.

Feel free to experiment, and have fun with it!

Personal Automations

Scott Knight — Wed, 25 Oct 2023 07:36:43 -0500

Since I learned to code, I’ve always been a fan of eliminating boring and/or tedious stuff with automation. This can involve anything from custom scripting (usually with Python, in my case) to using a service like IFTTT, Zapier, or Make (formerly Integromat). With these tools you can connect most common services that expose an API.

While automation use cases are usually born out of some personal pain point or specific challenge, I’m always interested in seeing what other people automate and how they choose do it. I’ve also taken great inspiration from pieces like Al Sweigart’s Automate the Boring Stuff with Python and this classic Github repo.

I recently bought an IFTTT subscription and audited my personal life for a list of things that take up too much time or involve a tedious process.

Here are some of my favorites:

Email/SMS to Brag File (IFTTT): Sending an email or text to IFTTT creates a new row in my career brag file with text I include. (My “brag file” is what I humbly call the personal doc where I track my professional accomplishments so I don’t forget about past wins.)
Blog RSS to Socials (IFTTT): Each time I publish a new post on this site, it gets automatically posted to my Twitter and Mastodon accounts. (Bonus—Buttondown also sends an RSS-driven email to readers subscribed to my posts)
Instapaper to Trello (IFTTT): New articles saved to Instapaper automatically get added to a “future blog posts” list on my personal Trello board.
Trello Board Automations (Trello): Using Trello’s built-in automation capability, I’ve set up rules and buttons that keep my task board organized so I never lose track of any reminders, tasks, or projects. These are driven by due-dates, sources (cards added via email, etc.), and task age (highlighting overdue items and so on).
Smart Home/IoT Automations (IFTTT): Various applets turn on/off lights and appliances using smart plugs driven by my location, times, alarms, etc.
Custom Transaction Categorization (Python): A custom script I wrote using Python which can take a CSV export of my bank transactions and categorize them for my accounting. It can categorize a year’s worth of transactions in less than a second, which is pretty neat. It’s also more secure and private than granting access or uploading my bank info to some third-party company’s database.

Is there anything useful or cool that you’ve automated? Send me an email, I’m interested to learn more about it!

Automate Your Data Governance

Scott Knight — Tue, 10 Oct 2023 00:00:00 +0000

After reading about 23andMe’s recent data breach issue this morning, I’d like to share some thoughts on data governance from my perspective as a Data Automation Engineer.

Data drives just about every aspect of our world, and increasingly so as technology continues to evolve. A growing challenge with this is the art of managing and governing this ever-growing volume of data in a way that ensures the proper maintenance of its quality and security.

In my career as a data professional I have seen how data governance can slip into an afterthought, particularly in small or mid-sized organizations. Luckily, automated data governance can be a powerful tool to ensure data quality, security, and compliance without much effort.

Let’s take a moment to explore the idea and understand how it is poised to shape the future of data management.

What is data governance?

Where data analysis is the practice of collecting, transforming, visualizing, and utilizing data to inform decision making, data governance involves the establishment of policies, procedures, and controls to ensure data quality, integrity, privacy, and security. It’s about defining who has access to what data, how the data should be used, and maintaining compliance with regulatory requirements like GDPR, HIPAA, or other industry-specific standards that apply to your organization.

Traditionally, data governance has been a manual and resource-intensive process which is prone to human error. However, automation can streamline these processes with much more efficiency and effectiveness.

Before we move on, I would like to note that, as with all things involving automation, I do recommend including some element of regular human-led oversight to ensure everything functions as intended. If erroneous automations are left to their own devices and decide to go off the rails, whatever they control will quickly suffer from a lack of regular oversight.

Why should data governance be automated?

Keep pace with scale and complexity

As a business grows, so does their data. While it may seem fine to manage data policy in a smaller organization, this task quickly becomes unfeasible as a company expands both internally and externally; internally, with respect to access control and permissions monitoring, and externally, with respect to data volume and security.

Real-time monitoring

Automation allows you to monitor your data usage in real time, which in turn ensures more immediate responses to violations or discrepancies.

Improved consistency and efficiency

By implementing automation you are simultaneously ensuring the consistent application of your governance policies, reducing the error stemming from human-led governance.

Guaranteed compliance

Automated processes can be configured to comply with the latest data privacy regulations, making sure your compliance never lapses.

How can automation enhance data governance?

Quality monitoring and cleaning

Automations can be configured to monitor data quality continuously, so flagging and remediation of inconsistencies or bad data can be carried out even if you have a constant stream of incoming data.

Access control

Perhaps an obvious place to implement automation is with access rights and permissions. As an organization grows, it should not be granting access rights to individual users as it may have been done with a smaller team. Automations should be in place to manage access with user groups and predefined roles and rules.

Managing metadata

Using automated processes to extract and manage metadata can provide valuable insights into data lineage, usage, and relevance. For example, I implemented automated metadata management on a previous team using SharePoint Document Libraries. This helped us create the versioning, rollback, and artifact creation we needed to pass a CMMI Maturity Level 3 appraisal.

Policy enforcement

Governance policies, such as data retention policies and privacy configurations, can be automated using roles and rules as mentioned before. This allows your policies to be enforced consistently across the organization.

How to get started

1. Assess the playing field.

Conduct an in-depth analysis of your existing data to identify areas that can be automated. Common examples of areas may include those mentioned in the section above.

2. Choose your tools.

Select automation tools that align with your organization’s requirements to integrate seamlessly with your existing systems.

3. Implement and test.

Use the appropriate resources to design and implement automated data governance processes. Be sure to run a sufficient amount of testing to identify bugs and defects before putting your automations into a production environment. This will help you avoid many potential hiccups during your roll-out.

4. Train and integrate.

Train your team and communicate to the organization where appropriate on your new data governance processes. Educating your team is about how it works and who to reach out to for support will help things run smoothly after launch.

Wrapping up

Automating your data governance isn’t just a new trend in tech. It’s a necessity in today’s data-centric world. By learning how to leverage automation, you can ensure your organization maintains a high degree a data quality and compliance, paving the way for improved decision-making and sustainable growth. And in doing so, you’ll watch your data management transform into a seamless, efficient, and secure process.

Leveraging the Power of No-Code

Scott Knight — Thu, 05 Oct 2023 15:45:02 -0500

As a founding fellow of the Engineering fellowship at Coho, I was asked to write a guest post for their inaugural Blueprint newsletter when it launched a few months ago.

I chose to write an informative piece about how low- and no-code tools can be leveraged by non-technical folks to multiply their daily efforts by developing their own software tools.

I still believe this can be a game changer for those not familiar with traditional programming experience. If you’ve ever thought about a software solution for a problem or something you’d want to automate but weren’t sure how to bring it to life, this one’s for you.

Read the full article on Blueprint here

The Right to Repair: Fighting Planned Obsolescence

Scott Knight — Wed, 06 Sep 2023 15:41:08 -0500

“They don’t make stuff like they used to.”

We’ve all heard some form of this claim at some point and, unfortunately, it’s true. What’s concerning is that the decline of product quality and longevity we’ve seen in recent decades is fully intentional.

My family and I like to collect antiques and vintage goods we find at estate sales and shops on the weekends. Recently, this had me thinking: the items we find are noticeably sturdier and of a higher quality than anything you’d find on shelves today. From time to time we’ll even find vintage electronics or machines that still work like they did when they were new (or close to it)!

Many devices today are only good for a few years before they start slowing down dramatically, exhibiting battery life issues, physically breaking due to poor craftsmanship, and so on. To make things worse, the average consumer can’t even get into many of today’s devices to replace a battery or other parts. Myriad examples of these types of situations exist today.

Let’s talk about it.

What is planned obsolescence?

Planned obsolescence is the practice of designing products with a limited lifespan, so that they become obsolete and need to be replaced sooner rather than later. This is done in a variety of ways, such as cheapening materials, making products difficult to repair, intentionally reducing updates that slow down or disable older devices, or removing support for older devices altogether.

What is Right to Repair?

The Right to Repair is simply the principle that consumers should have the right to repair their own devices. This would include the right to repair information, spare parts, and tools necessary to accomplish successful repairs without needing to go back and pay for manufacturer’s service or replacing the product outright.

Obviously, this is not appropriate in all products for practal reasons. The Planned Obsolescence/Right to Repair conversation is primarily centered around the tech industry and its output.

Still, I would argue that we as consumers should have the option to repair most anything we own, if we so wish.

Why does it matter?

This principle is important for several reasons.

Having the Right to Repair helps to extend the lifespan of products, which can save consumers like you and me more money.
Repairing instead of replacing reduces waste and stimulates circular economic activity, which is great for the environment. When products are repaired, they are less likely to end up in landfills (currently a major source of pollution).
Consumers get more control over their devices and are less reliant upon manufacturers.
More jobs would be created in the repair industry.

How can companies transition?

While it will likely take legislative action to cause any significant change to these practices, here is a potential path for companies wanting to create more repairable products:

Use higher-quality materials that are less likely to break.
Make products easier to disassemble.
Provide clear repair instructions, diagrams, and documentation.
Offer spare parts and tools to consumers.
Allow independent repair shops to access diagnostic tools and software.

What can we do about it?

The right to repair is a growing movement. Canada, members of the E.U., and a number of states in the U.S. have passed Right to Repair laws, and other countries are considering similar laws as an instrument of consumer and environmental protection. As the Right to Repair movement continues to gain momentum, the hope is that more tech companies will start designing products that are more repairable.

If you are interested in learning more about the Right to Repair, there are a number of resources available online. You can also contact your local elected officials and ask them to support Right to Repair legislation.

If you’d like to discuss more about this issue, use the button below to email me. I’d love to hear your thoughts.

A Piece of AI History

Scott Knight — Sun, 20 Aug 2023 08:04:39 -0500

Yesterday I went to check out our local library’s annual used book sale, and one of the books I picked up is titled Essentials of Artificial Intelligence by Matt Ginsberg. I thought it was a neat find because the book was written in 1992, and this particular edition was published in 1993 so it provides some interesting insight into what was going on in the early days of AI research.

I’ve written a few posts recently that skew toward some of the more negative aspects of AI, but I don’t want to give the impression that I’m against it; on the contrary, I’m all for using AI and tech in general to better our human condition. It’s why I got into the field myself. I just think we should do it in a measured fashion and stay mindful of the side effects of its implementation, rather than letting those things be an afterthought.

Anyway, I’ll balance the scales here with a piece of AI history by sharing an excerpt from the book’s concluding remarks on the applications of AI. Comparatively, it does renew my appreciation of how far we’ve come technologically over the last thirty years.

AI’s most immediate applications are peaceful: Intelligent interfaces that make computers easier to use for everyone, such as telephones where you can talk in German but listen on the other end in English. Household robots will vacuum our houses some day (not soon, but some day); AI generally holds the promise of freeing us from the drudgery that we can hope to delegate to our machines.

AI is a field with tremendous potential and with tremendous risks—as is any other scientific discipline. The problem—indeed, the very idea—of constructing an intelligent artifact is sure to capture the public inteest. We must be sure that our zeal and enthusiasm is confined to our research, and does not damage our abilities to honor the responsibility that we all have to keep the public well informed of the nature of scientific activities.

— Matt Ginsberg, Essentials of Artificial Intelligence, 1993

The Ethics of Artificial Intelligence

Scott Knight — Wed, 16 Aug 2023 11:32:47 -0500

A couple of weeks ago a family member struck up a conversation that’s become somewhat commonplace in my social dealings. The person (for their privacy I won’t name them here), asked what my thoughts were on “all this AI” that’s being developed lately.

Because I work in the tech industry, AI is something I’m regularly asked about by those close to me as it’s become more of a hot topic in the last couple of years. And my opinion on AI hasn’t changed much since a decade ago when the field’s momentum really began to pick up.

Nonetheless, many of the concerns brought up in these conversations are valid and relate to ethical concerns about our future with this powerful technology. What follows is a selection of the most emergent ethical hurdles I think AI and the tech industry will need to implement solutions for, regulatory or otherwise.

Bias and Discrimination

AI systems are trained on data, and if the data is biased (many studies show that most datasets are), the AI system will be biased as well. This leads to AI systems making discriminatory decisions such as denying loans to members of marginalized communities, reinforcing bias in the judicial process, or denying appropriate healthcare measures to save insurance companies a few dollars at the expense of someone’s life.

To what degree should we be responsible for altering source data to try and swing the pendulum back in balance?

At what point does the act of correction itself constitute or introduce its own brand of bias?

How can we derive benchmarks for something as subjective as “fairness”?

Privacy and Surveillance

AI systems can readily be used to collect and analyze large amounts of data about people at a volume we’ve never seen before. Data can be, and already is, used to track people’s movements, monitor their online activity, and predict their future behavior. We know that much of this is already happening by social media platforms, their advertisers, and many other entities such as governments around the world.

This raises obvious concerns about the future of individual privacy and surveillance, as well as the potential for abuse of this data. For example:

How can we ensure fairness, transparency, and accountability, and that entities using AI-enabled surveillance measures do not violate human rights, such as the right to privacy and the freedom of expression?

What are the implications of using AI-driven surveillance to monitor political protests or other forms of dissent?

Job Displacement

AI systems are becoming increasingly capable of performing tasks that were once done by humans. This could lead to widespread job displacement as machines take over more and more jobs to save employers money.

While this is a natural and valid concern about something that would lead to significant changes in our economy and society, I’m not yet convinced that labor replacement by AI will be as widespread as many folks predict. AI systems are still in their infancy, and while their capabilities are impressive to the average human the technology is still quite prone to error; not only this, but those familiar with how AI works behind the scenes can vouch for the fact that there are still many tricks being employed to make generative output seem more impressive than it is, particularly from that of language models like ChatGPT.

That is to say, AI still has a long way to go to become reliable enough to automate the labor of humans across industries; nonetheless, we should expect this possibility and evaluate our level of risk on an individual level.

Can we use AI to foster the creation of new jobs, rather than just replacing existing ones?

How can we ensure that any transition to AI-powered jobs is fair and equitable?

Loss of Control

AI systems are becoming increasingly autonomous, meaning that they are able to make decisions on their own without human intervention. This raises concerns about the loss of control over AI systems, as well as the potential for these systems to make mistakes or even harm people. While this relates to previous concerns, I believe the ethical implications deserve their own consideration.

In general, perhaps AI developers should implement some form of guard-rail systems into their systems to provide a degree of oversight and course correction when necessary, rather than allowing models to run freely.

Shouldn’t we design AI systems to remain transparent and accountable to humans?

Who should be responsible for managing the oversight of AI systems?

Existential Crisis

As AI systems become more sophisticated, they may begin to question their own existence and purpose. It sounds funny, but as we’ve seen this is a phenomenon that’s already begun to appear on the fringes of AI news. Perhaps it could lead to a crisis of meaning for AI systems, as well as for humans who interact with them.

After all—if AI takes over all of our jobs in the future, what would we do with all of our free time?

A Week with Linux Mint

Scott Knight — Fri, 14 Jul 2023 15:31:48 -0500

Last week I posted here about open-sourcing my decision on which Linux distribution to use on my new lab machine. The overwhelming majority voted for Linux Mint, so that’s what I went with.

Not having used Linux in many years, I couldn’t recall anything particularly good or bad about my experience with the OS. That is to say, I’m going into this with an open mind.

The hardware I’m using isn’t anything special, which personally I prefer in the sense that it will allow me to additionally evaluate my chosen OS on an average system.

If you’re interested, here are the basic specs:

Model: Lenovo ThinkCentre M900 (2017, Refurbished)
Processor: Core i5-6500T
Processor speed: 2.5GHz
Intel HD Graphics 4000
RAM: 16GB
Storage: 256GB
vPro
Support for USB 3.0

The intention here is that it will serve as part of my home lab on which I can run experiments and projects without affecting my primary machines. So, with that in mind, it should be sufficient for its purpose.

Installation

I didn’t have any major issues installing Linux Mint on the machine. LM provides detailed installation docs that were easy to follow, and the process was a standard one: create a bootable USB, boot it up using the BIOS, then install.

One thing I appreciated was that, when booting with the install media before the main installation, you’re actually loading a live boot of Linux Mint, which allows you to preview and check for compatibility issues before you fully commit.

Once I ran the full installation, it probably took less than ten minutes to get fully set up. Another plus in my book.

By the way, I’m running Linux Mint 21.1 “Vera”.

User Experience

The one challenge I had in getting things fully operational was setting up the wi-fi adapter drivers on the system. Unfortunately, the Realtek RTL8188FTV that came with the ThinkCentre (the hardware does not have internal wi-fi support) was not recognized by the OS.

So, after some diligent research I was able to identify a solution that got it working. For those interested, I may write up a separate post on how specifically I accomplished this, but suffice it to say that it took some advanced tinkering to get things running. This is something that could be a major turn-off to

This Github repo was critical in getting the drivers installed and adding the appropriate kernel modules: https://github.com/kelebek333/rtl8188fu

After that minor inconvenience, everything in general has been pretty smooth. The OS itself is snappy and efficient. The themes and color options are a nice touch for those who like customization.

The overall feel is similar to Windows, so it is not alien to someone who might be wanting to migrate to Linux from that domain. Many of the folder actions, start menus, and general functionality are akin to what you’d find in Windows. I’ve heard that Linux Mint is a good starter Linux experience for those coming from a strictly Windows background, and so far I say that notion holds up.

My only small gripe is that some of the software packages seem to be intentionally held back by the distro maintainers from the publisher’s latest releases. For example, I wanted to set up the new Supernova release of Thunderbird, but at this time that does not appear to be possible. However, after some research I realized that this seems to be common practice in the Linux community and should not be unexpected. Some may see this as a negative, but I realize that this may also be a mechanism to protect against zero-day vulnerabilities as well as other issues by waiting a bit before integrating brand new releases. So for now, I’m okay waiting for a bit longer.

All things considered, the experience has been positive after one week. We’ll see how things progress as I continue to build out the machine.

Using Data to Choose a Linux Distro

Scott Knight — Fri, 07 Jul 2023 16:51:29 -0500

The Problem

I recently decided that I wanted to beef up my home lab a bit by adding a small, inexpensive machine that I can use as a sandbox for server projects and testing operating systems. When I realized that StackSocial had a Lenovo ThinkCentre M900 on sale for less than $200, I ordered one.

I also knew I wanted to try something besides Windows and MacOS on this new machine, so I decided to evaluate Linux distros. As it turns out, choosing a Linux distro is a really tall order for someone who’s only superifically familiar with the domain; there are hundreds, if not thousands, of distributions.

So, knowing the membership to be Linux and FOSS fans, I turned to the Mastodon community to help me decide what to try out first.

Data Collection

I posted a poll on Mastodon asking everyone to vote for their favorite Linux Distro.

I didn’t expect to get much engagement as a new member with few followers, but it seems as though this topic is something everyone has an opinion on.

Having set the poll to stay open for a week, I received 58 votes, not accounting for recommendations from replies on the post (I asked everyone with an “Other” vote to specify their preference in a reply).

After a week, I combed through the comments to create the full dataset analyze the results.

Analysis

I used a Jupyter Notebook for the basic analysis. Pandas helped me to import and transform the data, and I used Matplotlib for the visualizations. As there were two types of recommendations, I classified the recommendations as either distros or desktops (what’s the difference?).

From there, I added a few blocks of code to clean things up, sort the data, and create the bar charts.

Results

Now that I had a couple of visuals to evaluate, the results were quite interesting and unexpected. Zorin OS was one of my personal preferences going into the poll, but it hadn’t received a single vote (since the poll closed, I’ve had some comments recommending it).

Linux Mint was by far the winner in the distro category, followed by Fedora and NixOS. So, it looks like I’ll be starting off with Linux Mint.

For desktop environments, KDE Plasma was the most recommended.

I’ve since posted the results on Mastodon, in case anyone who participated was interested in the overall outcome.

I’ve pushed the project files to my Github, but I recommend checking it out here using the Jupyter Notebook Viewer if you just want to see the results.

Why I Wont Be Joining Threads

Scott Knight — Thu, 06 Jul 2023 17:36:44 -0500

Threads, the new app from Instagram (Meta), is designed to be a more “intimate and private” way to connect with your friends. It’s free to use and there are no ads, which seem to be two of the main selling points. So far, though, I’m not convinced that it’s enough to make me want to join.

First of all, let’s talk about the privacy aspect. Instagram has a pretty poor history when it comes to privacy. They’ve been caught allowing their ad partners to collect and share loads of data from millions of user accounts without their consent, and have been criticized for their lax security measures. So I’m not sure why I should trust them with the intimate and private messages between me and my friends.

Secondly, even if Threads were completely private I’m not sure I see the point. Between the “[insert career here] has been killed by AI” and “I discovered X about [insert buzzword]. Here’s what you should know…” posts, I’m already bored with the Twitter echo chamber. It’s become a cesspool of people scrambling over each other to grab whatever clout they can get their grubby little fingers on. If Threads is supposed to be the “new Twitter”, what would I gain by joining?

Finally, I suppose Threads is just another way for Meta to collect data on me and my digital interactions, which I’m not keen to support. As we know, if you’re not paying for a product then you likely are the product (or maybe it’s FOSS).

As the Sharks say: for those reasons, I’m out. I’m not necessarily saying it’s a bad app. I just don’t think it’s for me.

If you’re looking for a more private way to connect with your friends, there are plenty of better options out there. Signal is a great secure messaging option. I’d also recommend Mastodon as a federated alternative to Twitter.

For now, you can find me there.

What do you think? Are you planning on joining Threads?

18 Month reMarkable 2 Review

Scott Knight — Sun, 02 Jul 2023 13:59:37 -0500

The reMarkable 2 found its way to my Instagram feed in early 2022, a time when I was in the trenches of trying to get two companies off the ground while also just starting grad school working toward my MBA.

I’ve always preferred handwritten notes to typed, but a pain point for me has been keeping my notes organized in a way that’s easily accessible later. Having been through several iterations of owning an iPad and Apple Pencil for keeping notes, I didn’t like the amount of distraction iPads can cause with the notifications and other apps.

The reMarkable 2 seemed to answer this struggle, so I ordered one.

Pros

What stood out to me about the reMarkable 2 is that it’s a device designed for focus. There aren’t any additional apps or notifications to distract you from your note taking or writing, you can use it to convert handwritten notes to text, email them around, and sync your notes to your other devices. The user interface is minimal and easy to navigate, and I like the integration with various cloud storage providers.

In addition to these software features, the tactile experience of the reMarkable 2 is hard to beat. Even compared to other devices I’ve used, it’s probably the closest thing you’ll get to pen and paper without actually writing with pen and paper. In my opinion, reMarkable nailed that experience.

Physically, the device is also quite thin and sleek, so it’s easy to tote around to meetings or other places where you might want to bring some documents or take notes.

Cons

In my experience with the reMarkable 2, my only remarkable (pun intended) negatives are primarily the costs. While I think the device itself is reasonable, currently selling for $299 with a $279 refurbished option, the peripherals are going to rack up your order total.

The basic marker, which you need to use the tablet, is $79. However, this doesn’t include an eraser (trust me, you’ll want the eraser), so now you have to get the “Marker Plus” for $129.

I believe that as a critical piece of the system, the basic marker should be baked into the cost of, and included with, the tablet. Then, perhaps an upgraded marker model can be an upcharge if desired.

You’re also going to want to protect your device, and the cover folio is also sold separately starting at another $79.

The good news here is that you can find cheaper stylus and cover folio alternatives on Amazon for much cheaper, which is what I’ve done since the initial purchase.

Overall Opinion

All things considered, 18 months later I still use my reMarkable 2 every week. The stylus nibs last a good while, so after the initial investment there really isn’t much you need to keep the device running other than the optional $2.99 Connect subscription. Connect is reMarkable’s online service that keeps your notes synced to your other devices and backed up in their cloud.

With a battery lasting for weeks on a single charge, I’ve been quite happy with the low-maintenance experience of the paper tablet.

If you’re a heavy note taker and you’re on the fence, I’d recommend giving the reMarkable 2 a shot.

Have you used the reMarkable 2 or a similar device? Drop me an email using the button below, I’d love to hear about your experience with it.

Self Hosted Software List

Scott Knight — Wed, 28 Jun 2023 17:03:29 -0500

I’ve been browsing around for project ideas for an old Dell Inspiron 15 I have laying around. The bones are great, but some of the keys don’t work and the screen has some damage after being dropped on the ground at some point.

I want to keep it in use somehow, and my initial idea was to use it as a homebrew NAS server using TrueNas.

Before committing to that, though, I decided to check out other self-hosted software out there in case there was something else I’d rather add to my home lab.

I came across the list below on Github and wanted to store it here for future reference. There’s a ton of good stuff in there.

Repo Link: awesome-selfhosted

What's Up With Reddit?

Scott Knight — Mon, 26 Jun 2023 15:00:54 -0500

How long can a site continue to alienate its user base and volunteer moderators before its bull-headed CEO corrects course? Reddit may be on track to answer that question for us.

The Situation

In case you haven’t heard, Reddit mods and users have been on a major strike over the last week. Thousands of subreddits (some holding tens of millions of subscribers) have either gone private or otherwise restricted access to protest the company’s recent announcement of its intention to implement some major API pricing changes.

While these changes may have internally seemed like a good idea for stakeholders, this would also force some of the largest third-party apps and services to close up shop.

The Problem

The problem isn’t that Reddit made a business decision. From an executive and stakeholder’s perspective it makes all the sense in the world to steer toward increasing revenue; they realized that many generative AI services like ChatGPT are using their corpus of data and they want to be compensated. Fair enough.

What I see as the crux of the matter is that Reddit has obliterated any trust that its users, mods, and third-party devs had for them as an organization. The new pricing will have a greater negative effect on services who aren’t even using it for generative AI.

Further complicating things, Steve Huffman’s ~~idiotic~~ stubborn response to the Reddit community’s protests has done nothing but dig a deeper hole for the company.

I’m not sure the company’s reputation will fully recover from this fiasco.

What Happens Now?

Ultimately, it’s anyone’s guess.

As of now, Huffman really doesn’t care what the community thinks and shows no intention of backing away from the decision. He thinks that this protest will pass.

And while it may, this whole ordeal is certainly leaving a bad taste in the mouths of many of its users (myself included).

Is There a Downside to AI?

Scott Knight — Tue, 13 Jun 2023 00:00:00 +0000

I had a conversation the other day with my good friend, Kasra Khalili, that stuck with me.

We were discussing the viral rise of AI since the release of ChatGPT and its future potential. AI (Artificial Intelligence, if you’re reading this from the rock you’ve been living under) is the flavor of the month in the tech world, and it is rapidly entangling our everyday lives. From smartphones to self-driving cars, AI is changing the way we live and work.

One thing I’ve noticed lately is that, while many folks (especially Twitter tech bros) are hailing the latest exploits or scrapping to stake their claim as the latest and greatest “ChatGPT Prompt Engineer”, not many are talking about the potentially negative side effects lurking under the surface.

One of my biggest concerns about AI tools like ChatGPT is that it could lead to people becoming overly dependent on it. We’re already seeing this happen with students who use ChatGPT to write their essays in school, and workers who use it to automate the bulk of their daily tasks. This is a dangerous trend, and I worry that it could lead to us losing our critical thinking skills. If we become too reliant on AI to do our thinking for us, we’ll become lazy and our mental faculties will atrophy. AI should be a tool, not a second brain.

Another trend I’m predicting is a sharp increase in what is already a minefield of misinformation out there. Just last year, a group of researchers used ChatGPT to generate a fake news article that claimed the death of a major politician. The article was shared widely on social media, and many believed it to be true, causing a great deal of confusion and panic. This, unfortunately, is not an uncommon story.

Considering this type of content generation in tandem with deepfakes, I wonder how many AI tools will spawn just to combat the rubbish being generated by the other platforms.

Of course, these are just a couple of the potential pitfalls. There are, as we’ve already seen, many benefits to AI-based technology. I’m just saying it’s important to weigh the risks and benefits carefully. I believe we should monitor the negative side effects so that we can take steps to mitigate them, rather than stand by indifferently, waiting for it to take over our role in the world.

Rather than rush the process, it would be wise to take a measured approach to AI development and make sure that it continues to be used for good. Let’s also make sure it doesn’t become so powerful that it poses a threat to humanity (we are sure that’s off the table, right?).

All things considered, I believe the future of AI is bright. If we’re mindful about how we proceed with it, AI can continue to be a force for good in the world.

Scott Knight's Blog

How to Keep Your Mac Awake Using Terminal

Step 1: Open Terminal

Step 2: Type caffeinate and hit Enter

Resources

Homelab Update

Keeping Analog Alive

Once

How to Create Custom Terminal Commands in MacOS

1. Create a file to store your custom commands

2. Add custom commands

3. Set up file permissions, if needed

4. Make it available in the terminal

How to Automate Anything

1. Choose what you want to automate

2. Record the steps

3. Identify the most painful step

4. Choose your platform

5. Automate it

6. Repeat steps 3-5 until the whole process is automated

7. Deploy

8. Monitor, iterate, and improve

Personal Automations

Automate Your Data Governance

What is data governance?

Why should data governance be automated?

Keep pace with scale and complexity

Real-time monitoring

Improved consistency and efficiency

Guaranteed compliance

How can automation enhance data governance?

Quality monitoring and cleaning

Access control

Managing metadata

Policy enforcement

How to get started

1. Assess the playing field.

2. Choose your tools.

3. Implement and test.

4. Train and integrate.

Wrapping up

Leveraging the Power of No-Code

The Right to Repair: Fighting Planned Obsolescence

What is planned obsolescence?

What is Right to Repair?

Why does it matter?

How can companies transition?

What can we do about it?

A Piece of AI History

The Ethics of Artificial Intelligence

Bias and Discrimination

Privacy and Surveillance

Job Displacement

Loss of Control

Existential Crisis

A Week with Linux Mint

Installation

User Experience

Using Data to Choose a Linux Distro

The Problem

Data Collection

Analysis

Results

Why I Wont Be Joining Threads

18 Month reMarkable 2 Review

Pros

Cons

Overall Opinion

Self Hosted Software List

What's Up With Reddit?

The Situation

The Problem

What Happens Now?

Is There a Downside to AI?

Step 2: Type `caffeinate` and hit Enter