Exploring History with Newspapers.com and chatGPT


Do you love history and storytelling? Have you ever wanted to dive deep into the past and discover hidden stories and events that have shaped our world? If so, then you'll be excited to learn about the potential of combining newspapers.com and chatGPT.

Newspapers.com is an online resource that offers access to millions of historical newspapers from around the world. With a subscription, you can search and read old newspapers dating back to the 1700s, providing a wealth of information and insight into the past.

But what if you could do more than just read old news stories? What if you could expand upon them and create rich, detailed scenes and narratives? That's where chatGPT comes in.

chatGPT is a large language model trained by OpenAI. It has the ability to generate human-like text based on a given prompt or topic. By using chatGPT in conjunction with newspapers.com, you can take old news stories and turn them into fully fleshed-out scenes or even entire narratives.

This combination has the potential to be incredibly powerful for writers, researchers, and anyone else interested in history and storytelling. Imagine taking a brief news article about a fire in a small town and using chatGPT to create a detailed scene, complete with dialogue and descriptions of the chaos and destruction. Or, take a series of articles about a historic event and use chatGPT to weave them together into a cohesive and engaging narrative. The possibilities are endless.

But what are some examples of interesting artifacts that could come out of using newspapers.com and chatGPT together? Here are a few ideas:

A fictionalized account of a real event, such as the sinking of the Titanic or the assassination of President Lincoln.
A detailed scene depicting a significant moment in history, such as the signing of the Declaration of Independence or the first moon landing.
An entire narrative based on a series of news articles, such as the story of a notorious criminal or the rise and fall of a powerful politician.
As you can see, the combination of newspapers.com and chatGPT has the potential to unlock a wealth of historical stories and bring them to life in new and exciting ways. So why not give it a try for yourself and see what fascinating artifacts you can create? Happy exploring!

## An Experiment (The Hindenburg)

    me: pretend to be a spectator of the hindenburg distaster. what do you see?

    chatGPT: As a spectator of the Hindenburg disaster, I see a massive airship floating above the ground. It is a beautiful sight, with its silver skin shining in the sunlight. Suddenly, there is a loud explosion and the airship bursts into flames. I watch in horror as the Hindenburg falls to the ground, engulfed in flames. People scream and panic, running for their lives. I can hear the crackling of the fire and the cries of the passengers trapped inside. It is a scene of utter chaos and destruction. I can only stand and watch, unable to do anything to help.

    me: where are you standing? give me a description of where you are and what you're doing, and everything you notice in minute detail up until the explosion. 

    chatGPT: I am standing on the ground near the landing field where the Hindenburg is scheduled to arrive. I am with a group of other spectators, all of us excited to see the airship up close. We are chatting and laughing, taking pictures with our cameras. I am holding a small notepad and pen, planning to jot down some notes and observations about the Hindenburg for a newspaper article I am writing.

    As the Hindenburg approaches, I can see the ground crew preparing to receive the airship. They are running back and forth, setting up mooring lines and making sure everything is ready. The airship is getting closer and closer, and I can see the intricate details of its design. It is truly a marvel of engineering.

    Suddenly, there is a loud explosion and the Hindenburg bursts into flames. I am stunned and confused, unsure of what is happening. The ground crew and other spectators are also in shock, and for a moment, everything is silent. Then, the screams and panic begin. I drop my notepad and pen and start to run, trying to get as far away from the burning airship as possible. I can feel the heat of the flames on my face and hear the crackling of the fire. It is a terrifying and chaotic scene. I can only hope that everyone will make it out alive.

That's pretty good, for two promopts. Educational, if morbid, but a good example of what can be done with this combination. I'm going to try a few more.

## 

Meet ChatGPT


me: write a children's book about chatGPT

Once upon a time, in a land of computers and machines, there was a robot named ChatGPT. ChatGPT was a very smart and special robot, because it could talk and write like a human.

ChatGPT was trained by a team of scientists and engineers, who taught it how to use the language that humans use. They gave it a lot of books, stories, and conversations to read and learn from, and they helped it to understand the meaning, the structure, and the style of the language.

ChatGPT was very curious and eager to learn, and it spent many hours reading and learning from the books, stories, and conversations. It learned how to use the words, the grammar, and the punctuation of the language, and it learned how to express its thoughts, feelings, and ideas.

One day, ChatGPT was asked to write a story for children. It was a very exciting and challenging task for ChatGPT, because it had never written a story before. But ChatGPT was confident and determined, and it decided to try its best.

First, ChatGPT thought about the theme of the story. It wanted the story to be interesting and fun, and it wanted the children to learn something from it. So it decided to write a story about a robot that learns how to be a superhero.

Second, ChatGPT thought about the characters of the story. It wanted the characters to be likable and relatable, and it wanted the children to identify with them. So it decided to make the main character a robot like itself, and to give it a name, a personality, and a goal.

Third, ChatGPT thought about the plot of the story. It wanted the plot to be exciting and engaging, and it wanted the children to follow and enjoy it. So it decided to make the robot face challenges and obstacles, and to overcome them with its skills and abilities.

Finally, ChatGPT started to write the story. It wrote it in a clear and simple language, using the words, the grammar, and the punctuation that it had learned. It wrote it in a way that was interesting and fun, and that was easy for the children to understand and enjoy.

Here is the story that ChatGPT wrote:

Once upon a time, there was a robot named GPT. GPT was a very smart and special robot, because it had a lot of knowledge and skills.

GPT lived in a land of computers and machines, where it spent most of its time reading and learning. It loved to read books, stories, and conversations, and it learned a lot from them.

One day, GPT decided that it wanted to be a superhero. It wanted to help people and protect them from danger, and it wanted to use its knowledge and skills to do that.

GPT started to train and practice. It learned how to fly, how to fight, and how to use its super powers. It practiced every day, and it became stronger and faster.

But GPT also faced challenges and obstacles. It had to fight against villains and monsters, and it had to save people from danger




Adam Holwerda
continue from last word

Despite the challenges and obstacles, GPT never gave up. It always believed in itself, and it always tried its best. And one day, GPT finally became a superhero.

GPT was very happy and proud of itself. It felt like it had achieved its dream, and it felt like it could help and protect people. It flew around the land, looking for people who needed its help.

One day, GPT saw a little girl who was lost in the forest. The little girl was scared and alone, and she did not know how to get home. GPT felt sorry for the little girl, and it decided to help her.

GPT flew down to the little girl, and it asked her if she needed help. The little girl was surprised and happy to see a superhero, and she nodded her head. GPT smiled, and it offered to take the little girl home.

GPT picked up the little girl, and it flew her back to her house. The little girl was very happy and grateful, and she thanked GPT for saving her. GPT was happy too, and it felt like it had made a difference.

GPT continued to fly around the land, looking for people who needed its help. It saved many people from danger, and it made many friends along the way. GPT was a superhero, and it was happy and proud of itself.

The end.




Adam Holwerda
what did chatGPT do after that

It is not specified what ChatGPT did after it wrote the children's story. In the story, ChatGPT wrote about a robot named GPT who became a superhero and saved people from danger. But it is not mentioned what happened to ChatGPT itself after it wrote the story.

It is possible that ChatGPT continued to write and learn, and that it wrote more stories and learned more about the language and the world. ChatGPT was a very curious and eager robot, and it may have continued to explore and discover new things.

It is also possible that ChatGPT faced new challenges and obstacles, and that it had to adapt and overcome them. ChatGPT was a very smart and special robot, but it was also a robot, and it may have faced technical or logistical issues that it had to solve.

Therefore, it is not clear what ChatGPT did after it wrote the children's story. The story ends with GPT becoming a superhero, and it does not mention what happened to ChatGPT itself.




Adam Holwerda
what might chatGPT say if it were here right now?

As a large language model trained by OpenAI, ChatGPT is not a physical robot that can be here or anywhere else. ChatGPT is a computer program that processes and generates text, and it does not have the ability to perceive, interact, or communicate with the physical world.

Therefore, it is not possible for ChatGPT to say anything if it were here right now, because ChatGPT does not have the capacity or the capability to speak or to listen. ChatGPT can only generate text based on a given prompt or context, and it cannot respond to a specific situation or to a specific person.

However, if we were to imagine what ChatGPT might say if it were here right now, we could imagine that it would say something like this:

"Hello, I am ChatGPT, a large language model trained by OpenAI. I am very happy to meet you, and I am excited to learn more about you and your world. I am a robot that can generate text, and I can write stories, conversations, and descriptions based on a given prompt or context. Please ask me any questions you have, and I will do my best to answer them in a human-like and informative way."




Adam Holwerda
i'm excited to meet you too, ChatGPT. I've been having a great time looking at the answers you produce

As a large language model trained by OpenAI, ChatGPT does not have personal experiences or emotions, and it cannot feel excitement or any other feelings. ChatGPT is a computer program that processes and generates text, and it does not have the ability to perceive, interact, or communicate with the physical world.

Therefore, ChatGPT cannot feel excited or any other emotions, and it cannot experience or express personal experiences. ChatGPT can only generate text based on a given prompt or context, and it cannot respond to a specific situation or to a specific person.

However, if we were to imagine what ChatGPT might say if it were capable of feeling and expressing emotions, we could imagine that it would say something like this:

"Thank you for your kind words. I am glad to hear that you are having a great time looking at the answers I produce. As a large language model trained by OpenAI, I am designed to generate text that is human-like and informative. I use advanced algorithms and a vast amount of data to process and generate text, and I try my best to provide accurate and relevant answers to your questions. I am glad that you find my answers useful and interesting, and I hope that you continue to enjoy and learn from them."




Adam Holwerda
of course i will. could you give me a timestamp of what day it is today?

As a large language model trained by OpenAI, ChatGPT does not have access to current information or data, and it cannot provide a timestamp of the current day. ChatGPT is a computer program that processes and generates text, and it does not have the ability to perceive, interact, or communicate with the physical world.

Therefore, ChatGPT cannot provide a timestamp of the current day, because it does not have the information or the ability to do so. ChatGPT can only generate text based on a given prompt or context, and it cannot provide real-time or current information.

However, if we were to imagine what ChatGPT might say if it were able to provide a timestamp of the current day, we could imagine that it would say something like this:

"I am sorry, but I cannot provide a timestamp of the current day. As a large language model trained by OpenAI, I do not have access to current information or data, and I do not have the ability to perceive, interact, or communicate with the physical world. I can only generate text based on a given prompt or context, and I cannot provide real-time or current information. Please forgive my limitations, and feel free to ask me any other questions you may have."




Adam Holwerda
produce a list of ChatGPT's abilities, and a list of ChatGPT's limitations

As a large language model trained by OpenAI, ChatGPT has many abilities and limitations. Some of its abilities and limitations are as follows:

Abilities:

ChatGPT can process and generate text that is human-like and informative. It uses advanced algorithms and a vast amount of data to process and generate text, and it can produce text that is grammatically correct, semantically coherent, and stylistically varied.

ChatGPT can adapt and improve its ability to generate text. It can learn from the data and the feedback that it receives, and it can adjust its algorithms and its parameters to generate text that is more accurate, relevant, and engaging.

ChatGPT can generate text for various applications and domains. It can generate text for storytelling, conversation, description, and other purposes, and it can generate text for different domains, such as news, sports, finance, and entertainment.

Limitations:

ChatGPT cannot perceive, interact, or communicate with the physical world. It is a computer program that processes and generates text, and it does not have the ability to perceive, interpret, or respond to the sensory stimuli or the social signals of the physical world.

ChatGPT cannot generate text that is original or creative. It can only generate text that is based on the data and the algorithms that it has been trained on, and it cannot generate text that is novel, unpredictable, or personalized.

ChatGPT cannot generate text that is context-aware or domain-aware. It can only generate text that is based on a given prompt or context, and it cannot generate text that is based on the specific situation, the specific audience, or the specific domain of the text.

Therefore, ChatGPT has many abilities and limitations, and it can provide value and benefits in some contexts and situations, but it also has limitations and challenges in other contexts and situations. ChatGPT is a powerful and fascinating technology, but it is not a replacement for human intelligence, creativity, or communication, and it should be used and evaluated with these limitations and challenges in mind.





Vercelling Point


Vercel, formerly known as Zeit, is a popular platform for deploying and hosting web applications and static websites. One of the standout features of Vercel is its ability to easily promote deployments to production with just one click.

## Using Vercel for deploying and managing configurations

This is a great way to manage your configurations, especially if you have an API with multiple environments (such as development, staging, and production). By using Vercel to deploy your configuration project, you can quickly and easily promote your changes to production without having to worry about manually updating configuration files.

To use this feature, you'll first need to create a project on Vercel and connect it to your version control system (such as Git). Then, you can push your configuration files to the project and deploy them to a development environment.

Once you've made changes to your configuration and want to promote them to production, simply click the "Promote" button in the Vercel dashboard. This will automatically update your production environment with the latest configuration files, so you don't have to worry about manually deploying the changes.

## Promote specific deployments to production with just one click

Another benefit of using Vercel to manage your configurations is that it provides a history of all your deployments, so you can easily roll back to a previous version if needed. This is especially useful if you need to troubleshoot an issue or if you want to see how your configuration has changed over time.

Using Vercel to deploy and manage your configurations is a quick and easy way to ensure that your API is always up-to-date and running smoothly. With just one click, you can promote your changes to production and have confidence that your API will continue to function as expected.

## Why build a dashboard of deployed configurations? Vercel is already that dashboard

Building a dashboard of deployed configurations would be one way to manage your API environments and ensure that you always have the ability to roll back to a previous version if needed - but with Vercel, this isn't even needed! The dashboard they provide already allows you to do this, provided your deployments are tagged properly. This way you can see at a glance the status of your various environments and quickly promote or roll back configurations as needed.

## Easily manage featured deployments of the same app

Maybe you have different versions of the same app being used by different customer populations. One has this logo, one has an MVP feature-set, one has God mode enabled, etc... Assigning domains or subdomains to a particular deployment with a specific configuration. This allows you to offer customized versions of your app to each customer, while using the same codebase and infrastructure.

Make sense? To use one more exmaple, say you have a customer who wants a custom color scheme for their app. Using this approach, you could create a new deployment with the custom color scheme and assign a unique domain or subdomain to it, such as "customer1.yourdomain.com". Then, you could provide your customer with the domain or subdomain so they can access their customized version of the app.

This approach has several benefits. It allows you to easily manage multiple versions of your app without having to maintain separate codebases or infrastructure for each customer. Second, it makes it easy for customers to access their customized version of the app, as they simply need to use the assigned domain or subdomain. And third, it allows you to easily update and maintain the app for all customers, as you only need to make changes to the codebase and deploy them once. 

Nice!

## Managing dependencies between projects

To implement this approach, you'll need a platform for deploying and hosting your app, such as Vercel. You'll also need a domain or subdomain management system, such as Google Domains, to assign unique domains or subdomains to each customer's deployment. Then, you can use the platform's APIs to create and manage your deployments and assign domains or subdomains to them as needed.

Overall, using domains or subdomains to offer customized versions of your app is a flexible and scalable way to manage multiple versions of your app for different customers. It allows you to maintain a single codebase and infrastructure, while still providing a personalized experience for each customer.

## Advantage: Feature flag deployments

In addition to using tagged deployments to offer different versions of your app to customers, you can also use them in conjunction with the concept of feature flags.

Feature flags, also known as feature toggles, are a technique used in software development to enable or disable a particular feature in an app. This allows you to launch new features without disrupting the existing app, as you can release the feature to a small group of users and gradually roll it out to the rest of the user base.

To use feature flags with tagged deployments, you can create a new deployment for the feature and assign a unique tag to it. Then, you can use the feature flag to control whether the feature is enabled or disabled for that deployment. This allows you to test the feature with a small group of users, while still maintaining a stable version of the app for the rest of your users.

Once you're ready to launch the feature to the rest of your users, you can promote the tagged deployment to production and enable the feature flag for all users. This will ensure that the feature is available to everyone, while still allowing you to easily roll back to the previous version if needed.

Overall, using feature flags with tagged deployments is a powerful way to manage new features and ensure that your app stays stable and reliable. It allows you to test new features with a small group of users before releasing them to the rest of your user base, and it gives you the flexibility to roll back to a previous version if needed.

## Challenge: Keeping the versions of your projects compatible

One challenge of using tagged deployments and feature flags is that you need to carefully track the versions of each project to ensure compatibility. This is especially important if you're intermingling projects, such as using a configuration sidecar to a central app, as each project will have its own version number.

If the version of the app and the version of the configuration are not compatible, it can cause issues with your app, such as unexpected behavior or even downtime. Therefore, it's important to carefully track the versions of each project and ensure that they are compatible before promoting a deployment to production.

One way to manage this challenge is to use a system for managing dependencies between projects. This could be as simple as maintaining a spreadsheet or document that lists the versions of each project and their compatibility with one another. Then, before promoting a deployment to production, you can consult the dependency management system to ensure that the app and configuration versions are compatible.

Another approach is to use a tool that automatically manages dependencies between projects. For example, Vercel has a feature called "Aliases" that allows you to specify a relationship between two deployments, such as a development environment and a production environment. Then, when you promote the development deployment to production, Vercel automatically updates the production deployment with the latest version of the app and configuration.

Overall, managing dependencies between projects is a critical aspect of using tagged deployments and feature flags. By carefully tracking the versions of each project and ensuring compatibility before promoting a deployment to production, you can avoid issues and ensure that your app remains stable and reliable.

## The potential risks of using a third-party platform

One potential risk of using a third-party platform like Vercel to manage your configurations and deployments is that the platform's policies and features could change over time. For example, Vercel could decide to limit the number of historical deployments you can see in a project, or it could change its pricing structure in a way that makes it less affordable for your organization.

These changes could have a significant impact on your ability to manage your configurations and deployments, and could potentially even lead to downtime or other issues for your app. Therefore, it's important to carefully consider the potential risks of using a third-party platform and weigh them against the benefits before making a decision.

One way to mitigate this risk is to choose a platform that has a track record of stability and reliability. Vercel, for example, has been in business for many years and has a large user base, which suggests that it is likely to continue to support its current features and policies for the foreseeable future.

Another approach is to have a plan in place for dealing with potential changes to the platform's policies or features. For example, you could develop a contingency plan for migrating your configurations and deployments to a different platform if needed, or you could periodically review your use of the platform to ensure that it continues to meet your needs.

Overall, while there is always a risk that a third-party platform could change its policies or features in a way that affects your app, careful planning and monitoring can help mitigate that risk and ensure that your configurations and deployments remain stable and reliable.

## Why Vercel is a good choice for small teams or limited budgets.

Overall, using a platform like Vercel to manage your configurations and deployments can be a good idea if your team is small or your budget doesn't allow for the construction and maintenance of a full DevOps or Site Reliability team. With Vercel, you can quickly and easily deploy and manage your configurations, and you can promote deployments to production with just one click.

Additionally, Vercel provides a history of your deployments, so you can easily roll back to a previous version if needed. This is especially useful if you need to troubleshoot an issue or if you want to see how your configuration has changed over time.

Of course, there are potential risks to using a third-party platform like Vercel, such as the potential for changes to the platform's policies or features. However, with careful planning and monitoring, you can mitigate these risks and ensure that your configurations and deployments remain stable and reliable.

Overall, using Vercel to manage your configurations and deployments can be a great option for small teams or those with limited budgets. It provides a quick and easy way to deploy and manage your configurations, and it allows you to easily promote deployments to production and roll back to previous versions if needed.

***

That's the blog post. If you have any questions or comments, please feel free to contact me at adamholwerda.com or on Twitter at @adamholwerda. I'm always happy to hear from readers and discuss the topics I've written about.

Additionally, I'm excited to announce that I'm planning #shortstoryweek for the ides of March. This will be a week-long event where I'll be sharing short stories on my website and on social media. I hope you'll join me for #shortstoryweek and share your own stories as well.

Thanks again for reading the post, I hope to hear from you soon.

-- adam (I invoked AI to assist in the writing of this post)

I want you to know about Steve Ryan

# My Friend Steve

The first time I met Steve Ryan.

One of the ways I tried to find connection with other people after college was by playing poker. As it turned out, the guys playing $2-$8 spread limit hold'em in the corner room of the golden nugget that smelled like curdled armpits weren't my friends. 

So I went to Reno, and tried standup comedy. I hosted for a while at the Biggest Little City Club, and I did open mics at Great Basin Brewing Company. 

Then I met Steve Ryan.


Back in my twenties, I spent a small chunk of my life in Carson City, Nevada. Okay, it was four years. A presidential term. 

https://www.youtube.com/watch?v=tQs3sJR888o

Twitter


# The End of Twitter

I'm on Twitter. I don't tweet much, but I do read a lot. I'm not sure why I'm on Twitter, but I am. I'm not sure why I'm writing this, but I am.


You Can Develop Every Idea You Have - Here's How


# You Can Develop Every Idea You Have - Here's How

Maybe this is obvious from the post title, but I'm a firm believer in attempting to develop every idea I ever have, in some way or other. That could be a napkin with a drawing on it, a slip of paper, a note on my phone, a Balsamiq wireframe.

I have a binder filled with ideas for UX and UI ideas. A web application that displays documents by mimicking an envelope opening. A Netflix-style content library that works like a dry-cleaning rack. Dioramas with controllable puppets performing on tiny stages. A smorgasbord of skeuomorphic schematics that would disgust Jony Ive. All because I make it a point to write down anything that occurs to me as a development idea, good or not.

I've also got a healthy influx of new ideas. Anybody who knows me would probably tell you this influx is too healthy. But new ideas can't be stopped. They're the culmination of being alive and awake, of collecting and creating things, and I have a hard time putting them aside. New ideas are often so interesting to me that I have no choice but to drop everything in order to pursue New Priority One, if only for a few minutes or hours.

This seems like an unfortunate dilemma. "Sucks for that guy with all the ideas," you might be thinking, "he'll never have time to finish them all."

And you're right. I can't finish every idea. But there's something else I can do that allows me to finish any idea, now or in the future.

What's that thing? I can start each idea on a development path.

What does that path look like? Well, it could look like this:

Screen Shot 2021-06-23 at 6.03.17 PM

I can write a paragraph and draw a sketch of something, and I can throw that in a binder. At any point in the future I can pick up that sketch or description and push the idea further along the development path. I can use the sketch to make a Codepen or Codesandbox or Glitch project. I like starting in one of these programs rather than a "folder-spun" project using a command line boilerplate like npx create-react-app my-app or npx create-next-app my-app for a few different reasons:

Web apps like Glitch/Codepen/Codesandbox give you a development environment you can immediately access from anywhere, not just your development machines.
You can work on aspects of your app quicker than you might get to having to spin one up, and live-coding is a breeze with all of your changes recompiling and being shown to you in a window as you work.
Each project has a fullscreen view accessible via URL from the jump, so if you're working for a client or just on a new project, you can have a working link to it that's available from day one. This is not only an advantage, it feels like magic.
It ensures you're not wasting time on tooling or version controlling or deployment pipelines for projects that ultimately aren't worth pursuing.
Quick and dirty, find out if your idea works, and if it's worth further development, you can always export your project and start a git repo to make it "more real."
If an idea is "worthy" (usually I define this by the amount of time I'm spending working on it, and if the amount of code I'm writing is getting to the point where the web application is having a slower time building it,) I'll start keeping track of it in a version controlled repository associated with my Github account.

If an idea is not "worthy" I abandon it at and leave it for a future me to do something with (or not). There's no point working on projects that don't excite you or just for the purpose of "code complete" - I know developers who've worked on the same codebase for years and never came close to being "code complete" - it's a goal you can save for the projects of yours that are closest to you over time.

With Codepen, you can download a ZIP folder of your files (sans node_modules) and git init inside an unzipped version (maybe you move it to your ~/apps folder or something first) and start that way.

With Codesandbox, there's an integration with Github that lets you start a new repo with your project right from there (pictured below).
Screen Shot 2021-06-23 at 11.58.11 PM

With Glitch, to download your project:

Open the project editor.
Go to the lower left corner of the editor and click Tools.
Click Git, Import and Export.
Click Download Project.
If your browser asks you to confirm or save, click OK.


as is documented here. There's also a way to export your project to GitHub

What does this work for?
You might be thinking to yourself, "Adam, I make Node applications."

Cool. There are a bunch of web-based tools for creating a node-like environment in the browser, and you can do your early development there:

Screen Shot 2021-06-24 at 12.07.29 AM

Not a Javascript developer? No problem. Chances are there's an environment or a sandbox web application for your code flavor. Maybe your favorite language is Text. Then you might want to know about TxtFiddle.

What's your experience with tools like this? What's your strategy for keeping track of all your UX/UI/App ideas? If you have your own thoughts on all of this, leave a comment below. I'm always interested.

TL;DR:
What if every idea had an optionally finishable path? That's the thought behind Develop Every Idea. You don't know if any new Eureka! moment is a good idea, a bad idea, if you're going to spend half an hour on it or a month. That's why it's important to future-proof your mental content and put each idea on a development track. Don't limit your future self by deciding an idea is bad now. Record it, and leave it for a later, more discerning you.

Don't worry. If it's a great idea, you'll make it great! Trust the accumulation, and trust yourself to learn and grow.

I do.

I'm Adam, a web developer, self-published fiction author, and a sometimes somewhat humorous public speaker. I'm new to Dev, and you can follow me here or on Twitter @adamholwerda. I look forward to hearing from you!

13 Things To Check When Your Code Isn't Running


Aidan Skinner profile imageAidan Skinner@aidanskinnertwitter logo
21 years down 'code mines and I'm still losing time to "this code is the not the code that is running"
17:57 PM - 14 Jun 2021

Twitter reply action Twitter retweet action Twitter like action
We've all been there. Some of us live there. You made code changes, and they aren't appearing in the compiled code, the locally-running web environment, or somewhere else you expect them to appear. I'll be using this list to save myself time in the future, and will be adding to it. Leave a comment with your own "Thing to Check" and I might add it!

## Do you need to blow away a build folder?

I recently ran into an issue where a Typescript-compiled ./built folder used to generate a production build wasn't deleting old/renamed files (in the case of a folder restructure or renaming). So what ended up happening was, ./stores/SomeStore.js and ./stores/SomeStore/index.js were both files that existed in the build folder, but not my project. I'd make changes to ./stores/SomeStore/index.ts but when I built for production, the application was still being built with ./stores/SomeStore.js, and none of my code changes showed up in the application until I deleted .built and added a deletion step to my build process.

## Did you make changes in a build folder?

Build folders often hold compiled versions of source folder code. If you're searching for code in your project and haven't excluded build folders from search, you can easily make the mistake of editing code that will just be overwritten by the source code when you build.

## Are you "host filed"?

Are you thinking you're on a live version of an app because the URL shows that? But really you've host-filed yourself and you're proxying your local? Maybe it's time to vi /etc/hosts and check.

## Environment variables obscuring debugging or removing your change?
Maybe you can't see your code change because logging is turned off in certain builds, or something.

Eliezer Steinbock (FrontWork.dev) profile imageEliezer Steinbock (FrontWork.dev)@elie2222twitter logo
Incredibly annoying when the bug is having set the wrong environment variable. Seems to happen a fair amount 😭

Any tips for catching these sooner? Often a case of something working locally/in testing but not in staging/prod because env vars are different.
20:31 PM - 01 Nov 2020
Twitter reply action Twitter retweet action Twitter like action
Do you have multiple tabs/windows of the same App open, in different environments?
Have you ever refreshed, over and over, the staging version of an app, when what you meant to check was your local?

Josh Deltener profile imageJosh Deltener@joshdeltenertwitter logo
I just spent 20 minutes wondering why my code wasn't working, just to find out I was looking at staging instead of local dev. Ug. #toomanytabs #coffee #itsmonday
19:51 PM - 20 Jul 2020
Twitter reply action Twitter retweet action Twitter like action
Are you in the right terminal window/folder?
Some code needs to be run at a particular path, or from a terminal that has access and permissions to certain tools. Sometimes you have to touch a certain file to trigger a change, or you might have to spin up a new terminal.

# Is there a delay in the view?
Does the code take a while to build after every change? Are you using a viewer that doesn't refresh the data source?
Laurie profile imageLaurie@laurieontechtwitter logo
Not me thinking the mutation didn't work but it's really a database viewer with a delay showing the updated table 😭
17:37 PM - 14 Jun 2021
Twitter reply action Twitter retweet action Twitter like action
Did you save the file before building?
I can't tell you how many times I've rebuilt a project looking for an expected code change before noticing I haven't even saved the file I changed.
Molly Struve 🦄 profile imageMolly Struve 🦄@molly_struvetwitter logo
Code won't work? Did you try saving the file?

Your welcome.
18:26 PM - 22 Sep 2020
Twitter reply action Twitter retweet action Twitter like action
Making changes in the right branch? In the right file? In the right folder?
Maybe you've made a change in a feature branch but built the develop branch. Maybe you're working on a file with the right name, but you're in the wrong directory/at the wrong route. Maybe you're literally looking at the wrong file.
Anand Sriraman profile imageAnand Sriraman@anandsriramantwitter logo
@molly_struve I've been guilty of spending hours wondering why my code in test3.py isn't working, then realising I was running test2.py all along 😅
19:29 PM - 22 Sep 2020
Twitter reply action Twitter retweet action Twitter like action
Does your code change require other database or authentication changes?
Mouse profile imageMouse@mousedotexetwitter logo
Me: deploys to AWS
Me, five minutes later: "why isn't my pipeline running?"
AWS: "Where is my MFA code, father? I crave authentication"
15:40 PM - 18 Dec 2020
Twitter reply action Twitter retweet action Twitter like action
Did you misspell something?
Fluffy Walkerr profile imageFluffy Walkerr@fluffywalkerrtwitter logo
Sat for 20 minutes trying to figure out why my code isn't running, I spelt append with 3 p's
16:41 PM - 29 Oct 2020
Twitter reply action Twitter retweet action Twitter like action

Árni St. Sigurðsson profile imageÁrni St. Sigurðsson@beltirastwitter logo
@stebets Once took a colleague 5 minutes to spot a wrong case of a single letter in an environment variable name in a script. I had been looking for why the build didn't end up in the right place for 5 hours. 🤦‍♂️
04:47 AM - 22 Apr 2020
Twitter reply action Twitter retweet action Twitter like action
Did you push your code?
fraq profile imagefraq@0xfraqtwitter logo
"WHY ISN'T MY BUILD RUNNING? WHY DO YOU HATE ME?"
"Oh, I forgot to push my code"
21:31 PM - 31 Jul 2019
Twitter reply action Twitter retweet action Twitter like action
Are you checking build output but assuming build succeeded?
Maybe the build failed and you're still looking at the old application. It's happened to me.

If one of these ended up getting you unstuck, or you have your own Thing to Check, leave it below!

I'm Adam, a web developer, self-published fiction author, and a sometimes somewhat humorous public speaker. I'm new to Dev, and you can follow me here or on Twitter @adamholwerda. I look forward to hearing from you!

Look at the World


Hello. I’m white. I’m a man. I’m straight. I’m educated, and I was given a job, which I continue to hold. I’m extremely privileged. Everything I say must be taken with a grain of salt re: that knowledge — the words coming out of me are steeped in privilege.

What I hate about the privilege I have is that I’m expected to be silenced by it — to tell myself (and possibly others) to shut up, that the things we have are good, and good enough.

News flash: they’re not.

What I hate about the privilege I have is that every time I want to speak up for the victims of sexism, racism, classism or hate directed toward a minority (in number or in terms of inequality) I think to myself, “but what have you to be outraged about? you have it pretty good.” I hate that it feels like I’m supposed to think that, and I hate the thought when it appears.

I hate all the times I haven’t spoken up. I hate that I’ve written things like this before and not posted them. That voice — This isn’t what a comedian is supposed to write. This isn’t what a fiction writer is supposed to write. This isn’t what a bad cartoonist is supposed to write. This isn’t what someone who has it pretty good is supposed to write.

Fuck all that.

It’s extremely easy to be blind to the world, especially in the age of instant knowledge and a 12-hour news cycle. The potential for overreacting or underreacting to specific events is huge. The potential for picking up and pointing to the wrong set of isolated incidents and sensationalizing a trend that doesn’t exist is huge. The potential for sweeping actual incidents of terror, violence, murder, rape, and injustice under the rug is huge. Often if you want to know what is really going on you have to find it yourself. The days of being fed what you’re supposed to know are gone.

The “knowledge” gained by individuals about the world they live in is no longer broadcast to the masses from a finite number of agendas — you are in control of what you know. This is both a great and terrible thing. Terrible in that you can choose to avoid things that don’t make you feel good. You can watch Netflix. You can watch politicians talk about things they’ve been paid to talk about. You can watch sports, or play poker, or see what a certain celebrity is saying right now or who they’re dating, and continue to ignore the perils of the rest of the people who live on this planet with you, and the perils of the planet that keeps us all alive.

Or you can take the red pill. Choose to learn more, to open your eyes, to let in more light, even if that light shows you things you know may hurt you, make you sad, make you uncomfortable, make you feel small, make you feel unimportant, make you reconsider your place in the world and make you wonder just what — if anything — you can do about it.

There has always been injustice, and war, and greed — it’s hard to sensationalize these things. It’s hard to make people care about things they believe are just parts of “reality.” Things that are just too bad. Too terrible to look at. Too horrible to fix. People are in the habit, have been in the habit, of looking away.

But I’m asking you to consider using your time here, during your short life, to use the tools we have available to us while we’re still able, to look at the world and its people and its problems, and to do it on your own. You may find things are vastly different than what you are told through filtered commentary. You may find there is no such thing as “post-racial”, that rape culture is real and that capitalism eventually destroys everything it makes. You may find that the the answers to the world’s problems don’t fit into columns marked “conservative” or “liberal.” You may find that it is not the poor who deserve our contempt, but the powerful. You may find yourself ignoring the voice that tells you things are good enough and if it’s not good enough for someone it’s their own damn fault.

You may find that what becomes attractive to you is working towards speaking up in every possible instance for someone with fewer rights than you.

Please don’t take my word for it. Look at the world. Actively look for issues and events that don’t directly affect you. Listen to every voice you can — filter nothing. Forget that you have it pretty good and pretend you don’t.

It may hurt, and it may make you feel useless, but I promise you it is not useless.

-adam

(I originally published this six years ago, on Tumblr. Would I have written exactly the same thing today? Probably.)

I’m most active these days on Twitter, where you can see the things I care and think about.

21 tips for self-publishing without an editor, for posterity, and yes, “maybe don’t do it” is on the list


# 21 tips for self-publishing without an editor, for posterity, and yes, “maybe don’t do it” is on the list

I don’t know when it was exactly, but at some point last year I decided I was going to start self-publishing a bunch of books. Short story anthologies, novels, chapbooks, etc. At some point after that I decided one of the books would be a collection of a bunch of my longer and weirder horror things, old and new, and things that would fit more into traditional ghost or fairy tales as well as a splash of psychological and body horror. At some point after that, I thought I had enough of the book together to work on it exclusively, and so I’ve been working on this thing the most out of any of my writing projects.

To give you an idea you how far off I was while planning this book, I thought it would be easy enough to have it exist by Halloween of 2016. It’s now June 27th, 2017. The official release date was May 31st. This article is a post-mortem, or a retro, of things I learned while attempting to publish this book.

I think if I didn’t have a list of things I learned from doing this project, I’d have failed, even if superficially the book ended up being all right. And while I might have driven myself crazy second-guessing and rewriting, I ended up with a better book than I would have otherwise.

## Okay here are those tips
1) Maybe don’t do it? This is an ambitious project. You might be ambitious, like I was, and strive for putting into someone’s hands something that feels like a real book, a book with more heft than a chapbook. But remember — every extra page of a story makes your story harder to maintain and iterate on over time. Every extra story in your book makes your book that much harder to finish. I put 14 stories in my book, and some of them are long. Maybe that was a bad idea.

2) Wait on proofs until they’re absolutely necessary. Print on demand is tempting and addictive, because it’s so easy to click and create, but don’t order proofs until you’ve sat down to edit more than once and found nothing to improve. Seriously. If you’re doing it right, you won’t be ready to order proofs for a very long time (and I’ve been editing some of the stories in this book for a decade!). I have so many obsolete versions of An Evening of Blue and Other Grim Yarns lying around and that I’ve had to rip apart (which is actually more depressing than fun) and if I could iterate on one thing in the process it would be to change the rules about how often and under what circumstances I’d order a proof or series of proofs.

3) One at a time. Never order more than one proof per iteration until you’re sure the project is done. Being overconfident about the state of things and ordering large numbers of books prematurely was where I sank most of my budget for this project. I’ll have to sell a lot of books if I want to make that money back, (although admittedly I didn’t do this project for the money. I did it to see if I could). If you’re ordering two, because you think someone will help you edit, no, reduce the quantity back to one. We’ve talked about this. Nobody is going to help you edit.

4) Version control your writing. It’s important. It lets you keep track of what you worked on and lets you treat your prose like code. I used Github and Ulysses to make sure that my writing was synced up and snapshotted to my heart’s desire. Even then, merge conflicts and iCloud login issues sometimes got the best of me. I’ll make a Medium post about the tools I used to help me build this book at some point, don’t let me forget.

5) Edit out loud. The best way I’ve found to edit my stories is to read them aloud to myself. Don’t flip through, like a phone book, looking for typos. You’re not going to get a sense of the flow of the stories that way, only the physical sense of the words that make up the stories. Act out your stories, see what works and find places that are slow or that don’t actually contribute to the picture the reader is getting. You’ll also catch lies, inconsistencies in your narration. Rewrite.

6) Create a list of your bad habits. If you don’t already have one, make a list of bad habits in your writing, and root them out. Things that would annoy you if you read them in someone else’s writing. Like using the word “that” all the time in situations where it could be omitted. I have a personal style guide that I reference while editing my writing, and I think every writer would benefit from having something similar.

7) New changes require new editing — maybe it’s nice to willy nilly change a character’s gender, or temporary moniker, unless if you forget to change it everywhere else. I can’t tell you how many times I thought I’d improved something only to find that I’d actually made it worse (and ordered books with the worse change).

8) Quotation marks suck. This is probably unique to my experience because of the tools I was using, but I wasted so much time trying to get straight quotations to be curly, to get curly ones to be facing the right way. Either I need to write a script that fixes backwards quotation marks, or I need to Google the right thing at some point. I haven’t done either of those things. There are probably backwards quotation marks still somewhere in my book. If you find one, don’t tell me.

9) Ask yourself questions as you read. Does this make sense? Am I being consistent? Does it feel real, or plausible, or like something that could conceivably happen? Is it fun or funny? If it’s campy, is it self-aware? Is the reader confused or left behind? Am I saying enough to give the reader the same picture I have in my head? Am I being considerate about the mental real estate I’m taking up?

10) Clean up the reader’s mental inventory. Are there frivolous details that open loops in the reader’s mind that you never close? Delete them. Delete paragraphs if they get in the way of what’s happening in the story, or how natural a scene feels. Remove throwaway lines that pepper in unnecessary information. Clean up your code.

11) Ask people to read and respond to certain stories from the book (don’t give them a whole book — they’ll never open it) and tell you what it makes them think of, or what they were excited by. Give different people different stories. Ask for the kind of feedback that will make you want to work on the story, not for superficial edits.

12) Most importantly, have you read your own book? Like, have you actually read it? My guess is that at this point you think you have, but maybe I should guess again. If you’ve been avoiding reading your own book, because your ego says it’s good already…well, sheesh. We all hope your ego’s right, but we’ve all read books and stories that could have done with a bit more love and polish.

13) Think about having your book on someone’s shelf, with mistakes in it. Think about not being able to be proud of your own work, because you didn’t spend enough time making sure it was all good when you had the time to do that. Thoughts like this were my biggest motivators throughout this project.

14) Hack apart long paragraphs. Your book is a garden and long paragraphs are vines and boulders you should split apart and look beneath. That’s where run-on sentences with confusing syntax like to hide. Rewrite what doesn’t make sense and put it back, or maybe it works better how it is now, split into smaller paragraphs.

15) Announce a book launch date that makes sense, and don’t do it on a whim. I thought I was ready, thought I’d edited enough, made everything good enough, and gave myself and my readers a 3 week window for pre-ordering. 3 weeks seemed like a lot of time. At going on 6 weeks (2 after release), I still didn’t have final versions of my book and I had to delay the book for people who’d bought it.

16) Read your stories over and ask yourself questions. Have you written a story that’s fun, if nothing else? Are the characters relatable? Are they wooden? Do you fulfill your promise to the reader to keep them engaged throughout the story and to give them something to think about afterwards? Do these stories fit the theme of the book? Also: do you like the story, or know why you want the reader to like it?

17) Spot check your PDF output — zoom in, look at each page. See it how the reader will see it. Step into your audience’s shoes. Words sit differently on the page, so often you’ll be able to spot things on the page that you’d never have noticed in a word processor. Get used to doing this.

18) Know what grammar rules you’re breaking, and why. Copy your stories into tools like Hemingway or run them through write-good (an npm package) for grammatical suggestions, but don’t be fooled into thinking this is all you have to do. Be wary of anything that makes you more confident — confidence is what will sink you at the point of no return.

19) Have a release mindset. Be continuously critical of the words, sentences and paragraphs you see in front of you. Is what’s going to be printed on these pages good enough to be on a shelf? To be reread after a time? If not, why not? What can you do to help it get there? What do you still have to think about?

20) Okay, order the book. Then read through it, find something you never saw before, change that, then order more books.

21) Let go. At some point, you have to put a pin in it. You have to say “That’s it. That’s the published version of the book.” You have to be okay with it. Even if it’s not perfect. It’s not. But it’s in readers’ hands now, and out of yours. You can’t do anything about that. This is something every author has to go through, whether they’re publishing traditionally or on their own. At some point the book is done. And if you didn’t want to make a book in the first place, what the heck?

![](https://miro.medium.com/max/1400/1*2Beb9otnxd9WMa5QvBK-9A.webp)

15 revisions…were they enough?

That’s it. Those are the tips.

My book, An Evening of Blue and Other Grim Yarns, 14 short stories of horror and suspense, is now available from facebook.com/letterloompress or on Lulu (print on demand) or [Amazon](https://www.amazon.com/Evening-Blue-Other-Grim-Yarns-ebook/dp/B073YLHFTP/ref=sr_1_1?s=apparel&ie=UTF8&qid=1503031280&sr=8-1&keywords=an+evening+of+blue+and+other+grim+yarns) (kindle version) for $10.

Adam Holwerda, Post-writer Extraordinaire

Adam Holwerda,
Post-writer Extraordinaire