Exploring Omnivore: A Powerful Open-Source Alternative for Content Aggregation

An illustration representing the concept of content aggregation and management. The image should feature a central figure symbolizing the Omnivore app, depicted as a powerful, modern tool. Around it, various streams of information (like articles, newsletters, and notes) are being efficiently organized and funneled into a streamlined workflow. Elements representing key integrations, such as Obsidian, should be subtly included, emphasizing the seamless connection between Omnivore and other productivity tools. The overall tone of the image should convey a sense of organization, efficiency, and technological advancement, suitable for a blog post about innovative content management solutions.

Streamlining Content Consumption with Omnivore – Integrations, Features, and Personal Experience

In the quest for efficient content management and aggregation, especially for those of us embedded in the tech and AI space, finding the right tools can be a game-changer. Today, I want to delve into one such tool that has recently caught my attention: Omnivore.

A Glimpse into the Landscape

Before Omnivore, there were several alternatives for content aggregation, notably Readwise. However, as a family man and solopreneur deeply involved in blockchain and AI, I sought something more aligned with my workflow, particularly integration with Obsidian, my go-to for managing information. Here’s a quick overview of what I considered:

  1. Zotero2Readwise: A Python library, perfect for bibliographic data management.
  2. Obsidian-Readwise: A TypeScript plugin syncing Readwise highlights into Obsidian.
  3. Omnivore: An open-source app with a compelling Obsidian integration.
  4. Wallabag: A self-hosted read-it-later app focusing on privacy and offline reading.
  5. Instapaper: A freemium model for saving web pages, albeit not entirely open-source.

Each of these tools offers something unique, but Omnivore stood out for its alignment with my needs.

Why Omnivore?

Omnivore appealed to me for several reasons. Its open-source nature means I can delve into its workings, ensuring it meshes seamlessly with my other tools. More importantly, its integration with Obsidian—a cornerstone of my workflow—made it an attractive option.

Personal Journey with Omnivore

My journey with Omnivore wasn’t without hiccups. Initially, I faced issues setting it up and turned to Wallabag as an interim solution. However, Wallabag’s lack of annotation features, crucial for my work, led me back to Omnivore.

After overcoming initial setup challenges and integrating it with Obsidian, the true potential of Omnivore began to shine. Its ability to pull in newsletters and possibly handle RSS feeds aligns perfectly with my need to sift through vast amounts of information efficiently.

Current State and Future Plans

I’m currently in the process of refining my use of Omnivore. The aim is to fully integrate it into my morning routine, replacing the manual trawl through news feeds with a more streamlined, Omnivore-mediated process. This change promises to enhance my competitive edge by allowing me to process and act on information more effectively.


In conclusion, for those in the AI and blockchain space, tools like Omnivore can significantly impact our daily productivity and information management. It’s not just about consuming content; it’s about integrating it into our workflow in the most efficient way possible. Omnivore, with its open-source nature and integration capabilities, is a tool that deserves attention for anyone looking to streamline their content consumption.

2024-01-09 Reading List

A collage for a blog post featuring six different themes: 1) A person doing equipment-free exercises like push-ups or squats at home. 2) An abstract representation of Bitcoin and ETFs, like golden coins with the Bitcoin symbol and documents or graphs. 3) Futuristic elements and gaming graphics inspired by a popular puzzle-platform game. 4) A computer screen displaying code and a certification badge, symbolizing tech education and certifications. 5) A juxtaposition of a corporate office setting with a home office, representing remote work and office policies. 6) Various cryptocurrencies, like coins and digital symbols, depicting the crypto market. The image should be vibrant and engaging, capturing the essence of each theme.

6 equipment free exercises for sculpting muscle over the holidays, according to a triple Olympian

So I did this to mix it up today, I’m well rested and need a pretty intense day to hit my goals, so I wanted to get a quick exercise in this morning at home instead of going to the gym. One of my goals that I’ve failed to do this year is to do five minutes of activity soon after waking. It’s good for sleep. So while I only intended to do this for five minutes, it took me well over twenty.

The target here is three sets of 20-25 reps with 30 seconds rest, but I cannot do that many pushups or pikes (yet!) so I only did half.

  • Plank ups
  • Knees to ISO squat
  • Pike ups
  • Split Squat
  • Push up
  • Side Lunge

I actually didn’t see side lunges, so I guess I cheated a bit. My plan was to go out running afterward, but it’s raining and only getting worse out there today, so I’m going to need to be creative to get it done today.

SEC Hustles to Answer Latest Bitcoin ETF Filings: Source
No comment here, will have more about the crazy week another day.

A fan-made, 7-hour Portal 2 prequel just hit Steam for free and it’s so good that I’m sad Valve stopped making Portal all over again

If you have Portal 2 it’s a no-brainer.

GitHub Certifications are generally available

I’m not sure if this is going to be a desired trait for job seekers, but if I was a few decades younger I would probably go for this. It’s definitely geared at enterprises, and seems to be similar to the CompTIA and Microsoft tracks. Exam costs are $200 each. The training materials might be worth bookmarking, it’s all on Microsoft Learn.
GitHub Actions
GitHub Administration
GitHub Advanced Security
GitHub Foundations
I would recommend any students or junior level job seekers to pick these up.

The data is in: RTO policies don’t improve employee performance or company value, but controlling bosses don’t care

The headline says it all. My partner is dealing with this right now, and I’d like to find a way to help her fight it. Collecting these notes is the first step.

8 Best Altcoins in January 2024: Reviewing the Top Altcoins Including Celestia, Solana, Sei, Corgi Ai, ApeMax, Injective, Bonk, and Arbitrum
Best Crypto to Buy Now January 8 – Injective, Stacks, Axelar
No recommendations here, DYOR, as I am.

Portkey : Control Panel for AI Apps

Adding this to the list of LLM observability tools. Hosted/OSS. Something I will be evaluating against Chainlit/LiteralAI.

LangChain v0.1.0
They’ve come a long way and were one of the first ‘agentic workload’ implemetnations. I’ve yet to really use them because of Steamship and OAI Assistants, but I should probably go back and rebuild an app using their stack for comparision’s sake.

Navigating the Ups and Downs of AI and Blockchain Development: A Real-World Saga

A futuristic workspace with advanced AI and blockchain technology, featuring screens displaying complex code and digital interfaces, amidst a modern, sleek office environment. The setting is vibrant yet professional, embodying the cutting-edge nature of AI and blockchain development.

From Technical Hurdles to Unexpected Downtimes – Lessons Learned in the Trenches of Tech Innovation

In the ever-evolving world of AI and blockchain, we often encounter a myriad of challenges, some anticipated and others completely unforeseen. Our recent experience serves as a testament to this unpredictable journey. Last Friday, we started encountering issues with our OpenAI assistant, deployed through a Chainlit app. For those unfamiliar, Chainlit is akin to Streamlit but with a Python backend and a React frontend, a blend of technological sophistication.

Our initial task seemed straightforward – uploading a PDF and several markdown files as source documents to enhance our app’s user experience. However, the challenge lay in properly displaying these citations in the UX. The complexities of this task led us down a rabbit hole of technical intricacies. We faced peculiar issues with the model’s response, especially when it came to retrieving information from the database. An instance that stood out was when a query about taking a day off on a birthday yielded no relevant results from the documents, leading to confusing model responses.

Determined to resolve this, we dived deeper, fine-tuning our approach. We realized that when no results were found, it was best to assume the documents didn’t cover the queried topic. Yet, this was just the tip of the iceberg. As we delved further, testing and tweaking, more bizarre responses emerged. The model reported technical difficulties in reading documents, yet contradictorily, it displayed search results.

Amidst this chaos, we discovered another glaring issue – the redundancy of file uploads. We had inadvertently uploaded the same file numerous times, a clear oversight in our workflow. This necessitated a thorough cleanup and recreation of our assistants, ensuring everything functioned seamlessly in our playground tests and in the app.

The real twist came when we recognized that the root of our troubles lay not within our code or approach but with OpenAI’s retrieval runs. Their models were not performing as expected, leading us to some prompt engineering adjustments. Frustrated and exhausted, we decided to pause and revisit the problem with a fresh perspective on Monday.

Monday brought its own set of surprises. Our return was greeted by a non-functional app, thanks to Chainlit Cloud being down. This downtime was a significant blow as Chainlit Cloud is integral to our data persistence layer, storing user maps and conversation databases. It’s akin to how ChatGPT displays threads of conversations. This persistence is crucial for our app’s functionality.

Rushing to the Chainlit Discord server, which I hadn’t joined until then, I learned that their cloud service was undergoing an update. This revelation was both horrifying and enlightening. In response, we disabled data persistence on our production server as a temporary fix and spent the day refactoring our code to adapt to the new Chainlit Cloud version.

This refactoring journey was not just about code. It was about understanding and integrating changes from the Chainlit cookbook, separating our custom code, and preserving the essence of third-party contributions. We faced dilemmas about merging different code histories and ensuring our customizations, especially around the assistant’s citation returns, were seamlessly integrated.

Our approach was to move our customizations into a separate file and import Chainlit’s updated cookbook as a utility. This method, while effective, brought its own set of challenges. Chainlit functions uniquely, with Python on the backend and a React frontend, a system that allows for deployment on platforms like Replit. Our goal was to customize this frontend without bloating the repository.

The downtime also highlighted a critical migration issue with Chainlit Cloud, now rebranded as Literal AI. They had changed key determinations and OAuth methods, leading to a temporary loss of our historical data. This situation underlined the importance of staying attuned to dependencies and the risks involved in relying on external services.

Despite these challenges, we emerged with valuable insights and a stronger, more resilient application. Our journey with Chainlit and OpenAI continues, marked by both triumphs and tribulations. As we progress, we remain committed to exploring and harnessing the immense potential of AI and blockchain technology, ready to tackle whatever hurdles come our way.

2024-01-08 Reading List

A stock image featuring a modern, clean desk with a computer displaying graphs and cryptocurrency logos on the screen. Beside the computer, there's a notebook with handwritten notes about AI and trading strategies. Several finance-related books are stacked neatly on the desk, along with a cup of coffee, symbolizing long hours of research. In the background, a corkboard or whiteboard is visible with various tasks and goals, emphasizing themes of freelancing and personal branding. The setting conveys a focused work environment, suitable for an entrepreneur engaged in the digital economy, crypto, and AI space.

I’ve decided that if I’m going to spend the time to go through my news feed in the morning, I should at least start posting what’s relevant to me (and hopefully, to you as well).

SEC reissues crypto ‘FOMO’ warning amid hope for spot Bitcoin ETFs

Just say no to drugs, kids.

11 Ways To Earn Money on TaskRabbit With ChatGPT

TL;DR: Basically anything that you can do on TR
– Offer Landscaping and Design Services
– Turn Car-washing into Auto Detailing
– Offer Home Design Services
– Provide Eco-friendly Cleaning Services
– Create Customized Grocery Lists and Recipes
– Get Tips and Suggestions for Baby Proofing
– Create Resumes and Help People Find Jobs
– Schedule Appointments
– Manage Social Media for Companies
– Send Marketing Emails
– Use ChatGPT to Market Your Services

This Paper Introduces LARP: An Artificial Intelligence Framework for Role-Playing Language Agents Tailored for Open-World Games

Saving for later, I have plans for stuff like this.

Project: https://miao-ai-lab.github.io/LARP/
Paper: https://arxiv.org/abs/2312.17653
GH: https://github.com/MiAO-AI-Lab/LARP

Your Personal Brand Is The Key To Building A Successful Career

This is the stuff google feeds me now to keep me pumped up about being a solopreneur.

5 Passive Income Ideas For 2024

Looks like I should take blogging/writing seriously again. I’m not buying property this year, and dividend stocks aren’t going to do it. I don’t really have the desire to start a course, but we’ll see what happens.

Investors hedge bets on Bitcoin with $50K call options before ETF decision

I have been telling people that the ETFs wont’ get their BTC for less than $60k, but given that the supposed approval will go through on Tuesday, I am willing to admit that I could be wrong. Still, part of my hedging plan for this year is to take advantage of calls and put options.
Right now I can make a $19 premium by selling BTC-Mini-12JAN2024-45000-Call option, but I’ll lose out on any rise over the strike price. I’ve done it a couple of times when I thought there was no chance of the price running high, but we’re only 4 days out and there aren’t any higher calls available unless I go further out. The ### BTC-Mini-26JAN2024-75000-Call is only going for about a dollar right now.
Generally speaking, the open interest on Derbit indicates short term consensus sentiment around the 50k mark.

Bitcoin Stock to Flow Model is Back With $532K BTC Price Prediction

Ah, PlanB, he of the eternal bull. There’s no doubt in my mind that BTC will hit this price, but if you had to pin me down I’d say this is a very long-term bet. I can’t say whether I think we’ll see that this cycle. The S2F model is a useful one regarding BTC’s design and long-term sarcity model, but I think that the 30% error in PlanB’s predictions from last cycle are too far off to make any bets. I’m still holding a profit-taking level (3.6x the Mayer Multiple), which is currently around $117k.

3 Altcoins That Could Replicate Solana’s 975% Annual Growth

Borroe Finance ($ROE): Pioneering Hassle-Free Fundraising
WEMIX (WEMIX): A Steady Growth and Diverse Ecosystem
Conflux (CFX):  Blockchain  with Hybrid Consensus and Tree-Graph Technology

AlgosOne AI Trading Solution
Put this in the bucket of things I’m willing to put money in but haven’t. Their front end is slick as hell, and it doesn’t have any schemey vibes at all. Unfortunately I’m unable to complete KYC due to non-US participation, but I’m still gonna shill my reflink:

A Messari Report: Crypto Theses for 2024

I usually read these religiously, but the report was restricted to paying member back in December. Now it’s available for freebie account holders. I’ll be putting this into a vector database for sure and using it to inform my 2024 strategy.

Notes on Culture DAO

Since getting laid off I’ve been tossing the idea of a new DAO around. I’ve been reading Iain Bank’s Culture series for the last few months, and the AI minds from the series are what brought me to it. There was a tweet asking which sci-fi series was most likely, given the advances of GPT and other LLMs, and one of the answers was the Culture. So I picked up Consider Phelbas and now I’m halfway through the series.

Building an AI assistant has been one of the projects I’ve been working on — I named mine Zephyr — it named itself, actually, but I digress. The Westworld examples of multiple independent LLM agents running around independently has fascinated me, and I want to build my own little world of these things. There are some gaming projects that are putting them to use in not only player interactions, but in other ways as well. A fishbowl perhaps, that we can peer in and play god.

Seeing as how I’ve spent the last year and a half working on the Star Atlas DAO, it was the only thing I could think to do but build a new one. The tech industry is notorious for their lack of unions, and I figured this could be a place for us to build a new group, a way for us to stick together and work toward some new network state.

And so Culture DAO. If we want to take the optimistic path in our new AI-assisted future, then the Culture is probably the best vision of utopia that I’ve seen put to media. Now of course we’ll have to be careful about how we associate ourselves with Bank’s vision — his estate had a deal with Amazon to develop a series — so we can’t take directly from it. But this idea of super-intelligent AI that function as drones, ships, and habitats as conscious individuals should be general enough.

I put some of my thoughts about the project into GPT and published this What is Culture DAO post.

Do nothing stupid

It’s been half a year since I published anything publicly. It got to the point that I was no longer comfortable talking about things that might let things slip about what was going on at work. It was just safer to not post.

I wish I could say that I kept up the journalling habit, but I didn’t really. Not typing things out the hard way, by hand. Instead, I started using Obsidian, a note-taking app, to start organizing my life. Obsidian has a number of plugins, core ones that ship with the app, and a host of open source community plugins that provide additional capabilities. So instead of these journals, I’d been dictating to Obsidian, then using Whisper to transcribe the files, at which point they could be passed as context to GPT for whatever purpose.

I became quite obsessive with using it, I would transcribe work meetings and use it to create task lists, stakeholder notes, or whatever task I needed. The dream was to have a codex of company information that could be used to build an AI assistant. Things haven’t quite come together, for reasons I’ll explain later.

Of course the subtext here is that I was laid off from Star Atlas last week – my last paycheck was yesterday so now I am officially unemployed. The news was not completely unanticipated — we hadn’t heard anything about VC funding in some time, so it was pretty clear the writing was on the wall. I’d also been bothered by economic doomer news, and had been saving up cash as much as I could these last few months. So while I’ve got my short-term financial needs in the bank, I’m a bit lost when it comes to my long-term plans.

I’ve decided to put all the worry out of my head for the next week or two — the kids start school at the end of the month, so I’m telling myself to just be a dad and enjoy the time with them. I pulled them out of day camp this week — Missus is out of town, so it’s just the three of us at the house the week. My hope is to keep them on track with their studies, and try to return to some semblance of routine once next week when they go back to camp and school. Last week was such a mess, it just didn’t make sense to pay $160 to keep them at camp while I was at home futzing about.

I’m just not ready to start looking for a new job. I don’t even know what I want to do yet, not ready to quit what I was doing. I’ve put so much mental energy into the SA DAO over these past two years that I can’t just stop and pivot. There’s still a lot of things I want to finish. I’ve got a lot invested in SA, and still believe that the team can do it. Now we know they have another year and a half runway. More than me, at this point, but all I want to do is survive. I don’t want to update my resume and go network on LinkedIn and fill out job applications that I don’t want in order to qualify for unemployment. The last two years have been a dream, and I’m not ready to wake up.

If I have to ask what I have to show for the last two years of fat times, I think I’d be hard pressed. We’ve travelled a bit: Costa Rica, Niagara Falls, Puerto Rico; I didn’t really buy a bunch of toys — maybe a bit much on video games — but we’ve actually been taking steps to declutter the house. I’d ask where it all went, but I know most of my budget lately has actually been groceries.

I’ve accumulated a fair bit of crypto, I’d have to go back and look at my cost basis before I make any claims to whether I’m up or down, but I have a sizable stash of Solana that I’m not in a hurry to part with. That’s probably the main determining factor as to why I’m not itching to look for jobs in the real world quite yet. I can’t cut my ties from Star Atlas completely — I’m locked for five years, like 99% of people invested in the DAO. So I’m not going anywhere yet.

There were about 120 people laid off last week, the contractors actually have another week per their agreement, but regardless. Last week we all moved into a new Discord server to commiserate and console each other. My initial reaction was the fact that we had a great team of people that were left out in the cold. After some initial conversations, we determined there are a good number of us that don’t need to go back to work right away. And many of us have the personal runway and desire not to return to the workforce. So while we all deal with the shock of our forced departures, we’re helping each other with resumes and LinkedIn connection, trying to help each other along, figure out who needs to work and who wants to build something new.

I really haven’t the faintest clue what I’m doing, so right now I’m sticking to the Do Nothing Stupid plan: coast on inertia, don’t risk capital unnecessarily, don’t burn bridges. I’m taking some time to get back up to speed on the latest advancements in the Solana ecosystem, and work with anyone from the old team who wants to build. I’m not sure if anyone else has the stomach to work within the SA ecosystem, but it’s what we know, and there’s nothing saying that we can’t do it. Permissionless blockchains and all that…

Exploring Nature & Technology

Morning journal entry, January 17. Time is 8.40. Girls just left for work since it’s Mrs. Work from Home Day on Tuesday. I took Younger in this morning. She wanted to ride her bike. It wasn’t too bad. It was about 44 degrees or so. So I wrote her in without too much trouble and made it back, meditated and here I am.

So I got about 20 minutes just to collect my thoughts. I had a lot I was thinking about during meditation today. I’ve got a busy day at work. Not a lot of time to work on other things this morning. So I’m not going to talk about that in particular for this entry.

Yesterday we took our neighbor/cousins, the T’s, to a state park. We packed a picnic and I got the bikes out. We took a walking trail and it was a lot harder than I thought it was going to be. We had walkie-talkies so we could stay together and not get lost. We found this one little gorge that was beautiful where we had a large tree had fallen over and it kind of made this bridge across the ravine. We played on it for a bit and then took a picture. Afterwards, we ate some more and went to the playground. I took the two little girls on a bike ride around the lake which took us about half an hour. We had to push the bikes up the hill, but got to ride down which was pretty fast and fun.

Mrs. was ready to leave so we packed it up and headed back. It was pretty fun, although I wish we would have stayed longer. I was surprisingly sore in my legs given that it wasn’t that big of a deal. I slept great last night.

I want to talk about Starseer this morning. There’s a CLI tool that integrates with the GPT that I want to try. We also took another look over the weekend at the Discord bot that we have set up in our staging server. So, it seems to be working right now. We’re going to hopefully take some make some efforts on that, but I really want to spend some time planning what I do in the ML department in the AI ML department. This sprint we’re gonna need a demo by the end of the sprint so I need to figure out like what I can show now and what we can create that we can ship and actually demo. Maybe we focus on the Whisper pipeline and integrate it with Discord to make recording easier. There’s a channel in Discord where people can go into, they can get the help commands from the channel and they can hit a transcribe command. As they start talking it will record their audio and provide them with a text document.

The house is a mess. We’ve just destroyed it after we got home last night.

I am trying to get the girls to do some more video stuff. We started recording one of their science experiment kits over the weekend and gave them cameras for the trip to the park. Just trying to figure out how we want to operate, like with all these PCs and Macs and iPhones and Androids. It’s kind of a mess.

I think I’m gonna break here and get to work on work stuff because we got 10 minutes till nine.

A Sunday Morning Reflection

It’s Sunday morning, January 15th, and not even eight o’clock yet. It’s very unusual for me to be up this early on a Sunday, let alone doing a journal entry, but I woke up pretty early this morning and just kind of laid in bed.

Saturday went well. My mom came down Friday and after the last recording we had a celebration of Star Outlaces two-year anniversary. After my mom got here, we went to pick up the girls and then we ate and drank. We played Dungeons and Dragons actually last night, or Saturday night. It was a very, very condensed version. I didn’t want to subject Nana and Missus to 16 rounds plus of the two hour Dungeons and Dragons adventures game, but we managed to just do the beholder 1-4 and it still took us an hour. It still proves my point that the Star Outlist RPG adventure is a good idea.

There was a free game called GameDec (short for Game Detective) on Epic Games. I started playing that and it’s a wonderful kind of asymmetric adventure game and real engine, I guess you would call it like 2.5d.

We did additional tests with Starseer here and GPT index over the weekend. GPT index had a new release that adds PDF reader, and there’s some other functionality. I think it does like web pages now, simple web pages, but we did manage to load up the master agreement between the union and the VA. I was able to query it and it worked actually pretty well. It did use a lot of tokens, so it’s a long document. Still, I think my API usage was less than a dollar yesterday.

We need to focus on like the command line interface, building a GUI to kind of manage some of the stuff. I’m still creating these indexes manually. I still have not been able to compose a recursive index yet, so I think maybe I can work on that today hopefully.

We went roller skating yesterday. We took our friends, whom I’ll start referring to as my cousins because Dan and I have that kind of relationship. We took the cousins down to the roller skating rink and had fun. We were mainly out there trying to figure out how to do like hockey turnarounds or whatever you call it, just trying to get comfortable spinning and stuff.

After that, Nana left about one o’clock. The kids went out to play with their cousins and we just kind of chilled. The house is kind of needs some cleaning up, but it’s not too bad. I am gonna do some work today. I’ve had some thoughts about how the indexing process in the summary. What I did do is I’m not sure I’ve talked about before making file documents from the indexer and kind of putting the original path and the original hash of the dock. I really want to focus on indexing GitHub repos and file systems. If you’re indexing a file folder, it should check to see whether it’s in a GitHub repo and if it is, it should have to get the hash of the repo.

I don’t want to call it directly from the command line, obviously, but we should be able to get some sort of information file hashes, whether it’s changed, and then maybe even have some kind of like file watching system that it automatically re-indexes data when it changes. But that use case might be a little much for what we’re doing.

The test with the union agreement proves some valid theses that we’re gonna be alright. If we can create a user interface that’s as user-friendly as GPT, we should be able to create a grievance intake form or intake process where the website actually prompts the user for information. Please guide them with questions until we’ve refined the data that the grievance form needs, and then examine the user’s problem against the master agreement.

Yeah, I mean we start there, but there’s still a lot of work to do. I want to see if I can get these transcripts through some kind of process. The one thing I’m trying to figure out is how accurately we can regurgitate a document. Like this transcript goes over 2,000 tokens, or 4,000 tokens. I mean if the transcript stays under 2,000 tokens, it’s really easy to just have that window. I don’t think that indexer is gonna be the end-all-be-all for this type of stuff. There needs to be some other kind of like workflow, kind of like task passing. Here’s an index file, it’s been transcribed, run this query on it, run this on it, and then once that’s done then you can index it.

Tomorrow is a holiday, and we’re gonna try to go back up to the state park where I almost died from over-exercising. There’s free admission to the parks, and Misses wants to do something. I talked to the cousins about coming down with us, doing some bike riding, some hiking, packing a cooler with sandwiches and snacks and stuff, and just kind of hanging out there all day in the park which I think will be fun.

The girls are going to church here in about an hour and a half. I’ll see if my brother wants to play some video games and we’ll just chill till they get back. Maybe go to brunch with Misses, and we don’t really have to do anything today. Just relax and kind of chill. It should be chill.