They depend heavily on open source..

Discord users send billions of messages daily. Users wish to search these messages. How do we index these to search by message keywords?

Let’s find out.

1. Discord utilizes Elasticsearch. Elasticsearch is a free, open search engine for textual, numerical, geographical, structured, and unstructured data. Apache Lucene powers Elasticsearch.
   

2. How does elastic search store data? It stores it as numerous key-value pairs in JSON documents.
   

3. How does elastic search index? Elastic search's index is inverted. An inverted index lists every unique word in every page and where it appears.

4. Elasticsearch indexes documents and generates an inverted index to make data searchable in near real-time. The index API adds or updates JSON documents in a given index.

1. Let's examine how discord uses Elastic Search. Elasticsearch prefers bulk indexing. Discord couldn't index real-time messages. You can't search posted messages. You want outdated messages.

6. Let's check what bulk indexing requires.
1. A temporary queue for incoming communications.
2. Indexer workers that index messages into elastic search.

1. Discord's queue is Celery. The queue is open-source. Elastic search won't run on a single server. It's clustered. Where should a message go? Where?

8. A shard allocator decides where to put the message. Nevertheless. Shattered? A shard combines elastic search and index on. So, these two form a shard which is used as a unit by discord. The elastic search itself has some shards. But this is different, so don’t get confused.

1. Now, the final part is service discovery — to discover the elastic search clusters and the hosts within that cluster. This, they do with the help of etcd another open source tool.
   

A great thing to notice here is that discord relies heavily on open source systems and their base implementations which is very different from a lot of other products.

Why breakthrough technologies must be accessible

ChatGPT has exploded. Over 1 million people have used the app, and coding sites like Stack Overflow have banned its answers. It's huge.

I wouldn't have called that as an AI researcher. ChatGPT uses the same GPT-3 technology that's been around for over two years.

More than impressive technology, ChatGPT 3 shows how access makes breakthroughs usable. OpenAI has finally made people realize the power of AI by packaging GPT-3 for normal users.

We think of Thomas Edison as the inventor of the lightbulb, not because he invented it, but because he popularized it.

Going forward, AI companies that make using AI easy will thrive.

Use-case importance

Most modern AI systems use massive language models. These language models are trained on 6,000+ years of human text.

GPT-3 ate 8 billion pages, almost every book, and Wikipedia. It created an AI that can write sea shanties and solve coding problems.

Nothing new. I began beta testing GPT-3 in 2020, but the system's basics date back further.

Tools like GPT-3 are hidden in many apps. Many of the AI writing assistants on this platform are just wrappers around GPT-3.

Lots of online utilitarian text, like restaurant menu summaries or city guides, is written by AI systems like GPT-3. You've probably read GPT-3 without knowing it.

Accessibility

Why is ChatGPT so popular if the technology is old?

ChatGPT makes the technology accessible. Free to use, people can sign up and text with the chatbot daily. ChatGPT isn't revolutionary. It does it in a way normal people can access and be amazed by.

Accessibility isn't easy. OpenAI's Sam Altman tweeted that opening ChatGPT to the public increased computing costs.

Each chat costs "low-digit cents" to process. OpenAI probably spends several hundred thousand dollars a day to keep ChatGPT running, with no immediate business case.

Academic researchers and others who developed GPT-3 couldn't afford it. Without resources to make technology accessible, it can't be used.

Retrospective

This dynamic is old. In the history of science, a researcher with a breakthrough idea was often overshadowed by an entrepreneur or visionary who made it accessible to the public.

We think of Thomas Edison as the inventor of the lightbulb. But really, Vasilij Petrov, Thomas Wright, and Joseph Swan invented the lightbulb. Edison made technology visible and accessible by electrifying public buildings, building power plants, and wiring.

Edison probably lost a ton of money on stunts like building a power plant to light JP Morgan's home, the NYSE, and several newspaper headquarters.

People wanted electric lights once they saw their benefits. By making the technology accessible and visible, Edison unlocked a hugely profitable market.

Similar things are happening in AI. ChatGPT shows that developing breakthrough technology in the lab or on B2B servers won't change the culture.

AI must engage people's imaginations to become mainstream. Before the tech impacts the world, people must play with it and see its revolutionary power.

As the field evolves, companies that make the technology widely available, even at great cost, will succeed.

OpenAI's compute fees are eye-watering. Revolutions are costly.

Deep learning-based generative art is being researched. As usual, self-learning is better. Some models, like OpenAI's DALL-E 2, require registration and can only be used online, but others can be used locally, which is usually more enjoyable for curious users. I'll demonstrate the Stable Diffusion model's operation on a standard PC.

Let’s get started.

What It Does

Stable Diffusion uses numerous components:

• A generative model trained to produce images is called a diffusion model. The model is incrementally improving the starting data, which is only random noise. The model has an image, and while it is being trained, the reversed process is being used to add noise to the image. Being able to reverse this procedure and create images from noise is where the true magic is (more details and samples can be found in the paper).

• An internal compressed representation of a latent diffusion model, which may be altered to produce the desired images, is used (more details can be found in the paper). The capacity to fine-tune the generation process is essential because producing pictures at random is not very attractive (as we can see, for instance, in Generative Adversarial Networks).

• A neural network model called CLIP (Contrastive Language-Image Pre-training) is used to translate natural language prompts into vector representations. This model, which was trained on 400,000,000 image-text pairs, enables the transformation of a text prompt into a latent space for the diffusion model in the scenario of stable diffusion (more details in that paper).

This figure shows all data flow:

The weights file size for Stable Diffusion model v1 is 4 GB and v2 is 5 GB, making the model quite huge. The v1 model was trained on 256x256 and 512x512 LAION-5B pictures on a 4,000 GPU cluster using over 150.000 NVIDIA A100 GPU hours. The open-source pre-trained model is helpful for us. And we will.

Install

Before utilizing the Python sources for Stable Diffusion v1 on GitHub, we must install Miniconda (assuming Git and Python are already installed):

wget https://repo.anaconda.com/miniconda/Miniconda3-py39_4.12.0-Linux-x86_64.sh
chmod +x Miniconda3-py39_4.12.0-Linux-x86_64.sh
./Miniconda3-py39_4.12.0-Linux-x86_64.sh
conda update -n base -c defaults conda

Install the source and prepare the environment:

git clone https://github.com/CompVis/stable-diffusion
cd stable-diffusion
conda env create -f environment.yaml
conda activate ldm
pip3 install transformers --upgrade

Download the pre-trained model weights next. HiggingFace has the newest checkpoint sd-v14.ckpt (a download is free but registration is required). Put the file in the project folder and have fun:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Almost. The installation is complete for happy users of current GPUs with 12 GB or more VRAM. RuntimeError: CUDA out of memory will occur otherwise. Two solutions exist.

Running the optimized version

Try optimizing first. After cloning the repository and enabling the environment (as previously), we can run the command:

python3 optimizedSD/optimized_txt2img.py --prompt "hello world" --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Stable Diffusion worked on my visual card with 8 GB RAM (alas, I did not behave well enough to get NVIDIA A100 for Christmas, so 8 GB GPU is the maximum I have;).

Running Stable Diffusion without GPU

If the GPU does not have enough RAM or is not CUDA-compatible, running the code on a CPU will be 20x slower but better than nothing. This unauthorized CPU-only branch from GitHub is easiest to obtain. We may easily edit the source code to use the latest version. It's strange that a pull request for that was made six months ago and still hasn't been approved, as the changes are simple. Readers can finish in 5 minutes:

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available at line 20 of ldm/models/diffusion/ddim.py ().

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available in line 20 of ldm/models/diffusion/plms.py ().

• Replace device=cuda in lines 38, 55, 83, and 142 of ldm/modules/encoders/modules.py with device=cuda if torch.cuda.is available(), otherwise cpu.

• Replace model.cuda() in scripts/txt2img.py line 28 and scripts/img2img.py line 43 with if torch.cuda.is available(): model.cuda ().

Run the script again.

Testing

Test the model. Text-to-image is the first choice. Test the command line example again:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

The slow generation takes 10 seconds on a GPU and 10 minutes on a CPU. Final image:

Hello world is dull and abstract. Try a brush-wielding hamster. Why? Because we can, and it's not as insane as Napoleon's cat. Another image:

Generating an image from a text prompt and another image is interesting. I made this picture in two minutes using the image editor (sorry, drawing wasn't my strong suit):

I can create an image from this drawing:

python3 scripts/img2img.py --prompt "A bird is sitting on a tree branch" --ckpt sd-v1-4.ckpt --init-img bird.png --strength 0.8

It was far better than my initial drawing:

I hope readers understand and experiment.

Stable Diffusion UI

Developers love the command line, but regular users may struggle. Stable Diffusion UI projects simplify image generation and installation. Simple usage:

• Unpack the ZIP after downloading it from https://github.com/cmdr2/stable-diffusion-ui/releases. Linux and Windows are compatible with Stable Diffusion UI (sorry for Mac users, but those machines are not well-suitable for heavy machine learning tasks anyway;).

• Start the script.

Done. The web browser UI makes configuring various Stable Diffusion features (upscaling, filtering, etc.) easy:

V2.1 of Stable Diffusion

I noticed the notification about releasing version 2.1 while writing this essay, and it was intriguing to test it. First, compare version 2 to version 1:

• alternative text encoding. The Contrastive LanguageImage Pre-training (CLIP) deep learning model, which was trained on a significant number of text-image pairs, is used in Stable Diffusion 1. The open-source CLIP implementation used in Stable Diffusion 2 is called OpenCLIP. It is difficult to determine whether there have been any technical advancements or if legal concerns were the main focus. However, because the training datasets for the two text encoders were different, the output results from V1 and V2 will differ for the identical text prompts.

• a new depth model that may be used to the output of image-to-image generation.

• a revolutionary upscaling technique that can quadruple the resolution of an image.

• Generally higher resolution Stable Diffusion 2 has the ability to produce both 512x512 and 768x768 pictures.

The Hugging Face website offers a free online demo of Stable Diffusion 2.1 for code testing. The process is the same as for version 1.4. Download a fresh version and activate the environment:

conda deactivate  
conda env remove -n ldm  # Use this if version 1 was previously installed
git clone https://github.com/Stability-AI/stablediffusion
cd stablediffusion
conda env create -f environment.yaml
conda activate ldm

Hugging Face offers a new weights ckpt file.

The Out of memory error prevented me from running this version on my 8 GB GPU. Version 2.1 fails on CPUs with the slow conv2d cpu not implemented for Half error (according to this GitHub issue, the CPU support for this algorithm and data type will not be added). The model can be modified from half to full precision (float16 instead of float32), however it doesn't make sense since v1 runs up to 10 minutes on the CPU and v2.1 should be much slower. The online demo results are visible. The same hamster painting with a brush prompt yielded this result:

It looks different from v1, but it functions and has a higher resolution.

The superresolution.py script can run the 4x Stable Diffusion upscaler locally (the x4-upscaler-ema.ckpt weights file should be in the same folder):

python3 scripts/gradio/superresolution.py configs/stable-diffusion/x4-upscaling.yaml x4-upscaler-ema.ckpt

This code allows the web browser UI to select the image to upscale:

The copy-paste strategy may explain why the upscaler needs a text prompt (and the Hugging Face code snippet does not have any text input as well). I got a GPU out of memory error again, although CUDA can be disabled like v1. However, processing an image for more than two hours is unlikely:

Stable Diffusion Limitations

When we use the model, it's fun to see what it can and can't do. Generative models produce abstract visuals but not photorealistic ones. This fundamentally limits The generative neural network was trained on text and image pairs, but humans have a lot of background knowledge about the world. The neural network model knows nothing. If someone asks me to draw a Chinese text, I can draw something that looks like Chinese but is actually gibberish because I never learnt it. Generative AI does too! Humans can learn new languages, but the Stable Diffusion AI model includes only language and image decoder brain components. For instance, the Stable Diffusion model will pull NO WAR banner-bearers like this:

V1:

V2.1:

The shot shows text, although the model never learned to read or write. The model's string tokenizer automatically converts letters to lowercase before generating the image, so typing NO WAR banner or no war banner is the same.

I can also ask the model to draw a gorgeous woman:

V1:

V2.1:

The first image is gorgeous but physically incorrect. A second one is better, although it has an Uncanny valley feel. BTW, v2 has a lifehack to add a negative prompt and define what we don't want on the image. Readers might try adding horrible anatomy to the gorgeous woman request.

If we ask for a cartoon attractive woman, the results are nice, but accuracy doesn't matter:

V1:

V2.1:

Another example: I ordered a model to sketch a mouse, which looks beautiful but has too many legs, ears, and fingers:

V1:

V2.1: improved but not perfect.

V1 produces a fun cartoon flying mouse if I want something more abstract:

I tried multiple times with V2.1 but only received this:

The image is OK, but the first version is closer to the request.

Stable Diffusion struggles to draw letters, fingers, etc. However, abstract images yield interesting outcomes. A rural landscape with a modern metropolis in the background turned out well:

V1:

V2.1:

Generative models help make paintings too (at least, abstract ones). I searched Google Image Search for modern art painting to see works by real artists, and this was the first image:

I typed "abstract oil painting of people dancing" and got this:

V1:

V2.1:

It's a different style, but I don't think the AI-generated graphics are worse than the human-drawn ones.

The AI model cannot think like humans. It thinks nothing. A stable diffusion model is a billion-parameter matrix trained on millions of text-image pairs. I input "robot is creating a picture with a pen" to create an image for this post. Humans understand requests immediately. I tried Stable Diffusion multiple times and got this:

This great artwork has a pen, robot, and sketch, however it was not asked. Maybe it was because the tokenizer deleted is and a words from a statement, but I tried other requests such robot painting picture with pen without success. It's harder to prompt a model than a person.

I hope Stable Diffusion's general effects are evident. Despite its limitations, it can produce beautiful photographs in some settings. Readers who want to use Stable Diffusion results should be warned. Source code examination demonstrates that Stable Diffusion images feature a concealed watermark (text StableDiffusionV1 and SDV2) encoded using the invisible-watermark Python package. It's not a secret, because the official Stable Diffusion repository's test watermark.py file contains a decoding snippet. The put watermark line in the txt2img.py source code can be removed if desired. I didn't discover this watermark on photographs made by the online Hugging Face demo. Maybe I did something incorrectly (but maybe they are just not using the txt2img script on their backend at all).

Conclusion

The Stable Diffusion model was fascinating. As I mentioned before, trying something yourself is always better than taking someone else's word, so I encourage readers to do the same (including this article as well;).

Is Generative AI a game-changer? My humble experience tells me:

• I think that place has a lot of potential. For designers and artists, generative AI can be a truly useful and innovative tool. Unfortunately, it can also pose a threat to some of them since if users can enter a text field to obtain a picture or a website logo in a matter of clicks, why would they pay more to a different party? Is it possible right now? unquestionably not yet. Images still have a very poor quality and are erroneous in minute details. And after viewing the image of the stunning woman above, models and fashion photographers may also unwind because it is highly unlikely that AI will replace them in the upcoming years.

• Today, generative AI is still in its infancy. Even 768x768 images are considered to be of a high resolution when using neural networks, which are computationally highly expensive. There isn't an AI model that can generate high-resolution photographs natively without upscaling or other methods, at least not as of the time this article was written, but it will happen eventually.

• It is still a challenge to accurately represent knowledge in neural networks (information like how many legs a cat has or the year Napoleon was born). Consequently, AI models struggle to create photorealistic photos, at least where little details are important (on the other side, when I searched Google for modern art paintings, the results are often even worse;).

• When compared to the carefully chosen images from official web pages or YouTube reviews, the average output quality of a Stable Diffusion generation process is actually less attractive because to its high degree of randomness. When using the same technique on their own, consumers will theoretically only view those images as 1% of the results.

Anyway, it's exciting to witness this area's advancement, especially because the project is open source. Google's Imagen and DALL-E 2 can also produce remarkable findings. It will be interesting to see how they progress.

New standards help NFTs reach their full potential.

NFTs lack interoperability and functionality. They have great potential but are mostly speculative. To maximize NFTs, we need flexible smart contracts.

Current requirements are too restrictive.

Most NFTs are based on ERC-721, which makes exchanging them easy. CryptoKitties, a popular online game, used the 2017 standard to demonstrate NFTs' potential.

This simple standard includes a base URI and incremental IDs for tokens. Add the tokenID to the base URI to get the token's metadata.

This let creators collect NFTs. Many NFT projects store metadata on IPFS, a distributed storage network, but others use Google Drive. NFT buyers often don't realize that if the creators delete or move the files, their NFT is just a pointer.

This isn't the standard's biggest issue. There's no way to validate NFT projects.

Creators are one of the most important aspects of art, but nothing is stored on-chain.

ERC-721 contracts only have a name and symbol.

Most of the data on OpenSea's collection pages isn't from the NFT's smart contract. It was added through a platform input field, so it's in the marketplace's database. Other websites may have different NFT information.

In five years, your NFT will be just a name, symbol, and ID.

Your NFT doesn't mention its creators. Although the smart contract has a public key, it doesn't reveal who created it.

The NFT's creators and their reputation are crucial to its value. Think digital fashion and big brands working with well-known designers when more professionals use NFTs. Don't you want them in your NFT?

Would paintings be as valuable if their artists were unknown? Would you believe it's real?

Buying directly from an on-chain artist would reduce scams. Current standards don't allow this data.

Most creator profiles live on centralized marketplaces and could disappear. Current platforms have outpaced underlying standards. The industry's standards are lagging.

For NFTs to grow beyond pointers to a monkey picture file, we may need to use new Web3-based standards.

Introducing NFTs 2.0

Fabian Vogelsteller, creator of ERC-20, developed new web3 standards. He proposed LSP7 Digital Asset and LSP8 Identifiable Digital Asset, also called NFT 2.0.

NFT and token metadata inputs are extendable. Changes to on-chain metadata inputs allow NFTs to evolve. Instead of public keys, the contract can have Universal Profile addresses attached. These profiles show creators' faces and reputations. NFTs can notify asset receivers, automating smart contracts.

LSP7 and LSP8 use ERC725Y. Using a generic data key-value store gives contracts much-needed features:

• The asset can be customized and made to stand out more by allowing for unlimited data attachment.

• Recognizing changes to the metadata

• using a hash reference for metadata rather than a URL reference

This base will allow more metadata customization and upgradeability. These guidelines are:

• Genuine and Verifiable Now, the creation of an NFT by a specific Universal Profile can be confirmed by smart contracts.

• Dynamic NFTs can update Flexible & Updatable Metadata, allowing certain things to evolve over time.

• Protected metadata Now, secure metadata that is readable by smart contracts can be added indefinitely.

• Better NFTS prevent the locking of NFTs by only being sent to Universal Profiles or a smart contract that can interact with them.

Summary

NFTS standards lack standardization and powering features, limiting the industry.

ERC-721 is the most popular NFT standard, but it only represents incremental tokenIDs without metadata or asset representation. No standard sender-receiver interaction or security measures ensure safe asset transfers.

NFT 2.0 refers to the new LSP7-DigitalAsset and LSP8-IdentifiableDigitalAsset standards.

They have new standards for flexible metadata, secure transfers, asset representation, and interactive transfer.

With NFTs 2.0 and Universal Profiles, creators could build on-chain reputations.

NFTs 2.0 could bring the industry's needed innovation if it wants to move beyond trading profile pictures for speculation.

ZERO GRACE/ZERO MALICE

Big companies and wealthy people make stupid mistakes too.

Your ancestors kept snakes and drank bad water. You (probably) don't because you've learnt from their failures via instinct+, the ultimate life-lessons streaming network in your head. Instincts foretell the future. If you approach a lion, it'll eat you. Our society's nuanced/complex decisions have surpassed instinct. Human growth depends on how we handle these issues. 80% of people believe they are above-average drivers, yet few believe they make many incorrect mistakes that make them risky. Stupidity hurts others like death. Basic Laws of Human Stupidity by Carlo Cipollas:

1. Everyone underestimates the prevalence of idiots in our society.

2. Any other trait a person may have has no bearing on how likely they are to be stupid.

3. A dumb individual is one who harms someone without benefiting themselves and may even lose money in the process.

4. Non-dumb people frequently underestimate how destructively powerful stupid people can be.

5. The most dangerous kind of person is a moron.

Professor Cippola defines stupid as bad for you and others. We underestimate the corporate world's and seemingly successful people's ability to make bad judgments that harm themselves and others. Success is an intoxication that makes you risk-aggressive and blurs your peripheral vision.

Stupid companies and decisions:

Big Dumber

Big-company bad ideas have more bulk and inertia. The world's most valuable company recently showed its board a VR headset. Jony Ive couldn't destroy Apple's terrible idea in 2015. Mr. Ive said that VR cut users off from the outer world, made them seem outdated, and lacked practical uses. Ives' design team doubted users would wear headsets for lengthy periods.

VR has cost tens of billions of dollars over a decade to prove nobody wants it. The next great SaaS startup will likely come from Florence, not Redmond or San Jose.

Apple Watch and Airpods have made the Cupertino company the world's largest jewelry maker. 10.5% of Apple's income, or $38 billion, comes from wearables in 2021. (seven times the revenue of Tiffany & Co.). Jewelry makes you more appealing and useful. Airpods and Apple Watch do both.

Headsets make you less beautiful and useful and promote isolation, loneliness, and unhappiness among American teenagers. My sons pretend they can't hear or see me when on their phones. VR headsets lack charisma.

Coinbase disclosed a plan to generate division and tension within its workplace weeks after Apple was pitched $2,000 smokes. The crypto-trading platform is piloting a program that rates staff after every interaction. If a coworker says anything you don't like, you should tell them how to improve. Everyone gets a 110-point scorecard. Coworkers should evaluate a person's rating while deciding whether to listen to them. It's ridiculous.

Organizations leverage our superpower of cooperation. This encourages non-cooperation, period. Bridgewater's founder Ray Dalio designed the approach to promote extreme transparency. Dalio has 223 billion reasons his managerial style works. There's reason to suppose only a small group of people, largely traders, will endure a granular scorecard. Bridgewater has 20% first-year turnover. Employees cry in bathrooms, and sex scandals are settled by ignoring individuals with poor believability levels. Coinbase might take solace that the stock is 80% below its initial offering price.

Poor Stupid

Fools' ledgers are valuable. More valuable are lists of foolish rich individuals.

Robinhood built a $8 billion corporation on financial ignorance. The firm's median account value is $240, and its stock has dropped 75% since last summer. Investors, customers, and society lose. Stupid. Luna published a comparable list on the blockchain, grew to $41 billion in market cap, then plummeted.

A podcast presenter is recruiting dentists and small-business owners to invest in Elon Musk's Twitter takeover. Investors pay a 7% fee and 10% of the upside for the chance to buy Twitter at a 35% premium to the current price. The proposal legitimizes CNBC's Trade Like Chuck advertising (Chuck made $4,600 into $460,000 in two years). This is stupid because it adds to the Twitter deal's desperation. Mr. Musk made an impression when he urged his lawyers to develop a legal rip-cord (There are bots on the platform!) to abandon the share purchase arrangement (for less than they are being marketed by the podcaster). Rolls-Royce may pay for this list of the dumb affluent because it includes potential Cullinan buyers.

Worst company? Flowcarbon, founded by WeWork founder Adam Neumann, operates at the convergence of carbon and crypto to democratize access to offsets and safeguard the earth's natural carbon sinks. Can I get an ayahuasca Big Gulp?

Neumann raised $70 million with their yogababble drink. More than half of the consideration came from selling GNT. Goddess Nature Token. I hope the company gets an S-1. Or I'll start a decentralized AI Meta Renewable NFTs company. My Community Based Ebitda coin will fund the company. Possible.

Stupidity inside oneself

This weekend, I was in NYC with my boys. My 14-year-old disappeared. He's realized I'm not cool and is mad I let the charade continue. When out with his dad, he likes to stroll home alone and depart before me. Friends told me hell would return, but I was surprised by how fast the eye roll came.

Not so with my 11-year-old. We went to The Edge, a Hudson Yards observation platform where you can see the city from 100 storeys up for $38. This is hell's seventh ring. Leaning into your boys' interests is key to engaging them (dad tip). Neither loves Crossfit, WW2 history, or antitrust law.

We take selfies on the Thrilling Glass Floor he spots. Dad, there's a bar! Coke? I nod, he rushes to the bar, stops, runs back for money, and sprints back. Sitting on stone seats, drinking Atlanta Champagne, he turns at me and asks, Isn't this amazing? I'll never reach paradise.

Later that night, the lads are asleep and I've had two Zacapas and Cokes. I SMS some friends about my day and how I feel about sons/fatherhood/etc. How I did. They responded and approached. The next morning, I'm sober, have distance from my son, and feel ashamed by my texts. Less likely to impulsively share my emotions with others. Stupid again.