20230321
This is the first episode of $cashed_out. I don't really know what I'm doing here, so bear with me.
Feedback on today’s piece.
Here's my final article. What do you think? Tell me your analysis of what I could have prompted you with to get your responses close to this final version.
🤖Your final article effectively discusses the challenges of solving the IT silo problem, the limitations of siloed organizations, and various approaches to tackle the issue. It also highlights the vendor perspective and the Silo Unifier's Paradox.
To achieve a more accurate response closer to this final version, you could have provided more detailed and specific prompts related to each section of the article, such as:
Explain the challenges of IT silo problem and why there's no single person responsible.
Discuss the drawbacks of siloed organizations and how it affects the overall system.
Describe various approaches to fix the silo problem and their limitations.
Share the perspective of vendors in solving silo issues and how they navigate through the challenges.
Explain the Silo Unifier's Paradox and its implications.
By breaking down your requests into specific sections, you can guide the AI to provide responses more closely aligned with your desired output.
Asides
ChatGPT - summarize meetings in real-time. I write notes and it just logs them. I can add things to come back to, etc. at anytime I can ask it summarize the context so far and suggest questions and next steps to ask. Looking at the realtime summarization and also asking it questions I could ask to clarify things would be interesting. You’d have to tell it the full setup at the start so that all you have to do is type into. “Act a meeting notes taker, and also a consultant. I will type notes from a meeting I’m in and I want you to summarize the notes and output two questions a consultant might ask about the meeting so far.” And then at the end, ask it to pull together and summarize it all.
Do platform engineers install, update, monitor, and manage kubernetes? Or does some other role do that? Does it differ between on-premises, unmanaged kubernetes in public cloud, and managed kubernetes in public cloud?
From the State of Spring 2022 Survey
Use of Kubernetes in Spring environments continues to rise, jumping another 8 points in 2022 to 62%—after a 13% increase from 2020 to 2021. Almost half (45%) report running production Spring workloads on Kubernetes in the public cloud, while 34% run production on Kubernetes on-prem, making it the dominant platform overall.
The larger the company, the more important reducing financial costs of hosting becomes. Almost half (47%) of companies
with more than 10,000 employees see this as a benefit of native compilation, and it is recognized as a benefit across all industries—with the exception of Government (9%) (possibly because hosting is less important in government IT).
#books
Verso, O’Reilly Cloud Native Transformation booklet, Tours of the Black Clock, The Psychology of Money: Timeless lessons on wealth, greed, and happiness, Housel, Morgan
Questions for interview of authors of DevSecOps in Practice With VMware Tanzu
Coté will give his idea of what Tanzu is.
As you wrote this book, who did you imagine the users of TAP were? Developers, operations people…others?
How do the various tools make how those people work together better?
You mention each advance exposes a new set of problems. Cloud → Containers → Kubernetes. What problem is Kubernetes exposing?
Is a “secure supply chain” more than a build pipeline that ensures you use trusted third party libraries? Does it include runtime security as determined by security and secrets management? At what point does the idea go to far and end up being everything?
From the book: “Now, let’s shift our perspective to that of an enterprise architect. Let’s call her Alana. She’s tasked with optimizing the time to value for developers, setting the technical direction, and creating and enforcing a set of standards that balances consistency and security with developer productivity and satisfaction.” - when it comes to this type of EA governance, is all that’s been missing this whole time a way for the EAs to actually write code to enforce/provide governance? That is, the only tool EAs had was mandates and internal wiki pages, not actual templates/accelerators to start developers on the right path (on the “left” side of the pipeline), but then enforcement on the “right” side of the pipeline.
“When the developers do not have the required awareness or there are no enterprise-level guidelines on how to build application container images, every team might end up with their own standards and practices. With such a lack of knowledge and controls in place, large organizations may end up with several suboptimal application container images, which may lead to a waste of network bandwidth and storage for every container image push and pull operation.” It seems to me that if you can control how an app is put into containers…shouldn’t we stop developers from building their own containers for production deploys? What am I getting wrong there?
I know that all of the API and gateway packages are different, at different layers of the stack, and used by different people. But, at some point, don’t we need to stop layering API gateway packages on-top of things? I look at the API and routing schemes out there and they all basically operate like the Internet does. Instead of creating virtual networking, a service mesh, kubernetes networking, and then layering on Spring Cloud Gateway…shouldn’t we just aspire to fix the Internet…or at least some of those layers? This seems like a failing of kubernetes: sure you could wrap whatever language you wanted around the kubernetes “service mesh,” but it should be a direct pass through semantically and name-by-name…right?
From the start I really liked the API portal. It seems to be solving a problem every development team has, even if there’s just one team - especially if there’s thousands of teams. Public APIs (open source and otherwise) and such are easier to find and understand. But, when it comes to inside a company, it’s hard to find APIs inside your organization, who owns that API, find documentation, and get started using them. As I recall, Spotify came up with Backstage because of the problem the API portal is solving. I’m a little worried though that if developers haven’t put in the time to fix this problem for, like, 40 years, they’re not going to start now even when there’s good tools. I hope I’m wrong! (Am I?)
I now this is totally naive, but I’ve never really talked through naive to informed. It seems like air gaping your infrastructure/platform/dev stack software should be a simple problem to solve: write your software so that it doesn’t make calls to the public internet. In those places that it does, change the address to one on your local network and put something like Harbor there and you’re done. And yet, making something air gapped is a REALLY BIG DEAL. What’s up with it being so tough? Is it technology, complexity…support is too hard…what?
What is a kubernetes “distro”? I used to think of it like a Linux distro: you could point to a disk image and say “there it is!” Or, when you unbox and open up a new Mac, it has a distinct OS distro on it, MacOS. But I get the feeling that a kubernetes distro is more like am open ended recipe in a cookbook. You can specify a bunch of ingredients, but what matters is the quality of those ingredients, the skill of the cook, and the way the food is consumed. It seems like what matters for a kubernetes “distro” is the installer, the updater, and the selection of “ingredients” the user/owner makes. Is that a good way of thinking about it? Or is it exactly like Canonical, RedHat, and Suse Linux where there definitely is a distro? If it is a good way, are we happy with that?
I’ve tried to understand Cartographer several times - and Cora made me feel like I mostly understood. But it’s still a weird notion to me. It doesn’t seem to actually do anything except sequence your actual build and hold onto a “bag” of data along the way. What’s your explanation of what it is and why it’s helpful? What are other approaches to it?
In addition to the obvious, things I’d like to see in a book about platform engineering
Have only a little bit of definitional discussion. Like, keep the “macro economic headwinds mean this topic is really important for companies to stay alive” stuff to a minimum, and just make a definition and move on without too much story-telling.
It looks like there’s plenty (in a good way) of talk of patterns, actual things to do and ways to think, and, what I’d call “tactics.” It looks actionable. My mindset is always “after this talk, reading this book or article, when the audience goes back to work on Monday, what can they start doing new/different?”
A lot of examples and anecdotes from NOT tech companies, please minimize Spotify-talk – we get it! What’s more valuable is how these practices can be applied at F500/G2000 companies and governments, esp. how to get there from traditional inf./ops. This is where case studies and stories are valuable.
While you don’t want to rely on a single technology, I’d want to use existing tech/products/projects as references. Like, if you don’t mention Kubernetes and Backstage a few times, it just seems totally weird. Sure, they may come and go, but they’re good analogy/examples/prototypes for people to learn from.
I’d spend at least a page or two discussing burnout for platform engineers. While they’re busy reducing developer cognitive load, if they’re expected to manage the stack of software from IDPs/Backstage, to middleware, to CaaS/Kubernetes, to coordinating with security and governance, also build tools and “secure software supply chains,” and then meet with (if not be!) the infrastructure managers, they’re going to go insane. There should be some model for taking care of the actual PE people to make sure we’re not shoveling too much on them.
Summarized by the 🤖
"I'm quitting my job at Amazon because in-person work isn't enough"
Source: Business Insider
The author writes about how they have decided to quit their job at Amazon despite liking the company and their job because they feel that in-person work isn't enough for them anymore.
The author explains that they were happy with their job when they first started because they enjoyed working with their colleagues in person and they felt that they were able to build strong relationships with them. However, with the pandemic forcing people to work remotely, the author has found that they no longer feel as connected to their colleagues as they used to.
The author also mentions that they have been feeling unfulfilled with their job lately, despite the fact that they like their team and manager. They feel that they are not being challenged enough and that their work is becoming repetitive.
Although the author acknowledges that they are fortunate to have a job at a company like Amazon, they have decided to leave in search of a job that offers them more fulfillment and allows them to work in a way that suits their needs.
The Three Word Brief"
Source: LinkedIn
The author, Bob Hoffman, writes about the power of simplicity in advertising, using the example of the "Got Milk?" campaign as a three-word brief that has become iconic.
Hoffman argues that a brief should be simple and memorable, and that too often advertisers and agencies overcomplicate things in an effort to appear clever or sophisticated.
He suggests that a three-word brief can be effective because it allows for creativity within constraints, and forces advertisers to distill their message down to its essence.
Hoffman also notes that a brief should be based on insights and understanding of the target audience, rather than just being a creative exercise. He suggests that a great brief should be able to answer the question, "Why should anyone care?"
Overall, Hoffman encourages advertisers and agencies to focus on simplicity and clarity in their briefs and messaging, rather than trying to impress with complexity or jargon.
"God Did Us a Favor by Destroying Twitter"
Source: Wired
The author argues that Twitter's recent outage, which lasted for several hours, was a welcome relief from the constant stream of negativity and outrage that dominates the platform.
The author contends that Twitter has become a place where people go to be offended and outraged, and that the platform's algorithms and design encourage this behavior.
The author also notes that Twitter is not a true representation of the world, as it amplifies extreme viewpoints and creates an illusion of consensus that does not reflect reality.
While acknowledging the usefulness of Twitter as a tool for communication and information sharing, the author suggests that the platform's negative aspects outweigh its benefits.
The author concludes that perhaps the outage was a blessing in disguise, and that it serves as a reminder that we should be mindful of the impact that social media has on our mental health and well-being.
"Three Things with Caitlin Stasey"
Source: The Guardian
Australian actress Caitlin Stasey shares three things that are important to her, including her feminist website, herself as a person, and her new role in a TV series. In the interview, Stasey discusses her passion for promoting women's voices and her website, which features stories and essays written by women. She also talks about her own journey to self-acceptance and the importance of mental health. Finally, she reflects on her new role in a TV series and the challenges and opportunities it presents.
What does she say about mental health?
In the interview, Caitlin Stasey emphasizes the importance of mental health and shares her own journey to self-acceptance. She discusses the need for open conversations about mental health and the importance of seeking help when needed. Specifically, she states: "It's really important to talk about your mental health and to be open and transparent about what's going on in your life. There's a lot of shame attached to it, and I think it's really important to break down that stigma."
“Why it’s simpler for AIs to write about art than for robots to do your laundry”
The author notes that while writing about art and culture may seem like a uniquely human activity, it can be broken down into patterns and structures that can be analyzed and replicated by AI systems.
According to the article, it’s hard for robots to do laundry because it requires fine motor skills and human-like dexterity. Unlike other household tasks that robots can perform, such as vacuuming or mopping, doing laundry requires the ability to manipulate and fold fabric in a way that is difficult to replicate with current robotics technology. Additionally, laundry often involves a variety of different fabrics and textures, which can make it even more challenging for robots to handle.
Explanation
I want to create an email newsletter that is not curated and just contains most everything I read or come across. I'll use ChatGPT to summarize a lot of the articles Axios style. It will be a lot of links and content. Come up with suggestions for names and descriptions. Names I can think of are things like: The Firehose, The Flow, something related to Heraclitus's river flowing stuff.
🤖 Here are a few suggestions for names and descriptions for your email newsletter:
Name: The FluxStream Description: Dive into a constantly changing stream of knowledge as The FluxStream delivers a flood of summarized articles and links, all inspired by the ever-flowing wisdom of Heraclitus.
Name: The Knowledge Rapids Description: The Knowledge Rapids brings you a swift current of information, summarizing an array of articles from diverse sources, akin to Heraclitus's river of constant change.
Name: The InfoCascade Description: Immerse yourself in a cascading waterfall of information with The InfoCascade, offering summaries of everything we read or come across to quench your thirst for knowledge.
Name: The River of Reads Description: Embark on a journey through The River of Reads, where you'll navigate a vast and flowing selection of summarized articles, links, and content that never cease to evolve.
Name: The Wisdom Torrent Description: The Wisdom Torrent unleashes a deluge of summarized articles, links, and content to keep you informed and enlightened, inspired by the ever-flowing river of Heraclitus's teachings.
Name: The Current of Curiosity Description: Let The Current of Curiosity sweep you away with a continuous flow of summarized articles and links, feeding your inquisitive mind with an unfiltered stream of knowledge.
Name: The Incessant Tide Description: Ride The Incessant Tide, where you'll experience an unrelenting wave of summarized articles, links, and content, inspired by the ever-changing nature of Hera
What are possible meanings of the phrase "cached out"?
🤖 The phrase "cached out" is a play on words that combines "cashed out" and "cache." It could have various meanings depending on the context, some of which might include:
In the context of technology or computing: If someone has "cached out," it might mean they have cleared or emptied a cache, which is a temporary storage area used to speed up data retrieval. This could refer to clearing a web browser cache or any other type of cache in a computing system.
In colloquial language: The phrase could be used metaphorically to describe someone who has used up their mental or physical resources, similar to clearing a cache. In this context, "cached out" might mean that someone is exhausted, burnt out, or has reached their limit in terms of energy or ideas.
In the context of geocaching: Geocaching is an outdoor recreational activity where participants use a GPS receiver or mobile device to hide and seek containers, called "geocaches" or "caches," at specific locations marked by coordinates. "Cached out" in this context could mean that a participant has found all the geocaches they intended to find or that they are done geocaching for the day.
Remember that "cached out" is not a common phrase and might not be immediately understood by everyone. It is a creative and playful combination of terms that could be used in specific contexts or for a unique effect.
Colophon
A lot of the above are summaries I’ve had ChatGPT4 do. The chat history listing is down now, so I don’t have all of them.
I mean, how bad could it be? I blew $15 on a shitty movie that came with popcorn and ran only two hours.