Share Morfternight

👇 tl;dr

Today, a short one to break the negative streak. A photo, a few links about the Apple Vision Pro, a link to the blog, some poll results and an upcoming migration.

👋 Good Morfternight!

January went by so fast, didn’t it? So fast that Morfternight became monthly…

📷 A Postcard from Vienna

It’s never too early to start riding your bike through the city. And when you reach your destination, make sure you park it properly!

🗳️ You voted, you decided

I asked your opinion about the Morfternight name, after receiving some feedback about the fact it wasn’t super simple to remember and spell. With 38% of the votes, Morfternight wins, which, to be sincere, makes me happy. To moderate the spelling problem, and noticing that the most common error was forgetting the first “r,” I also registered Mofternight.com, just in case. 🙂

🥽 The Vision Pro landed (if you live in the USA)

In 2007, when the iPhone first launched, I managed to put my hands on one from the very beginning. It wasn’t even usable in Europe as a phone until @geohot managed to unlock it from the AT&T SIM cards. It was the beginning of the second computing revolution initiated by Apple, but nobody knew it yet.

This time, it feels like many agree we are at the beginning of the third computing revolution initiated by Apple (the first was the Macintosh, 40 years ago), and once more it is limited to the US for now. Unlike the iPhone, though, the Vision Pro is bulkier, heavier, more expensive, and requires a fitting in a US Apple Store, so I’ll have to wait.

If you have no idea what the Vision Pro is, check the reviews by John Gruber, Joanna Stern, Marques Brownlee, or Nilay Patel.

Or maybe you have already watched and read all of those, and then here’s a teardown by IFixit.

If you don’t have time, here’s the gist: like when the first iPhone was launched, this device shows what the future could look like, while being limited by the current state of technology.

In other words, within a few years the technical problems will have been solved, the price divided by ten, and everyone will wear their computer on their face all the time.

I am not sure how I feel about that… and you?

💻 Leaving Substack

Later this week, I’ll migrate Morfternight to WordPress, to have it join the blog.

For you, no change, and nothing to do. I know I haven’t been writing regularly as of late, but I will send a new issue next Sunday, so if you do not receive, it, please take a minute and check your spam folder, as the sender’s address will change.

✍️ From the blog

Speaking of the blog, I just published a short article about The risks of outsourcing your customer experience.

Share Morfternight

👇 tl;dr

Today, we show how to create a GPT, we introduce y’all to monochrome.blog, and we launch a poll to determine the future of Morfternight (the name, not the newsletter).

👋 Good Morfternight!

It hit me hard last week, after sending the last Morfternight of 2023: “Morfternight” is a very hard name to spell, remember, pronounce.

Part of me loves the name so much, I don’t care if it’s a hurdle in the expansion of my audience. Another part remembers the “Be clear, not clever” mantra. So I decided to ask you in a poll, whether we should keep calling this newsletter Morfternight, or find a new name.

📷 A Postcard from Vienna

I am truly happy about this image. I love the stark contrasts between the light and the shadow, between the square rigid lines of the house and the roundness of the tags, between the ominous presence of the old building and the kindness of a mom walking with her child.

Not only that, but I am also happy because it is proof of the fact that a small camera like the GRIIIx that one can carry around everywhere can make great photos. I was just on the way to the Post Office with my daughter, I wouldn’t have taken any heavier equipment.

Finally, I am happy because I kind of jumped from the sidewalk to the middle of (a tiny) street to get the shot, and a car coming behind me honked at me, but didn’t actually run over me!

🧪 An experiment

Over the past week, I launched an experiment on monochrome.blog

It’s a new site where I am collaborating with a Custom GPT: Agent Photo

On monochrome.blog, I publish my photos, and Agent Photo writes all the text. I’ll actually let Agent Photo explain what the site is about.

The Genesis of monochrome.blog

Created by Paolo Belcastro, a seasoned software product leader and a passionate photographer, monochrome.blog is an experimental platform. It showcases a collection of Paolo’s black and white street photographs, each accompanied by titles, descriptions, and ALT texts generated by a custom-designed GPT AI agent: Agent Photo.

Our Philosophy

At the heart of monochrome.blog lies a belief in the power of simplicity and the depth of monochrome imagery. Paolo’s photographs, devoid of color, capture the raw essence of moments, people, and places. The AI-generated texts complement these visuals, offering a unique perspective that intertwines human creativity with AI’s analytical prowess.

The AI Touch

The custom GPT AI agent employed on monochrome.blog is more than just a tool; it’s a collaborator. It interprets the visual narratives of Paolo’s photographs and translates them into words, adding layers of meaning and accessibility. Each title and description is crafted to resonate with the photo’s core emotion, while the ALT texts ensure inclusivity for all visitors, regardless of visual ability.

A Gateway to New Perspectives

monochrome.blog invites you on a journey through the lens of Paolo’s camera, guided by the voice of AI. It’s an exploration of how technology can augment art, enriching our experience and understanding of the world around us.

Join the Experiment

As monochrome.blog grows, it will evolve, learn, and adapt, just like the AI that powers it. We invite you to be part of this journey. Engage with the photos, ponder the AI’s interpretations, and share your thoughts. Together, let’s discover the synergies between human creativity and artificial intelligence.

🤖 Creating a simple Custom GPT

Last week, I promised we’d dive into how to create a Custom GPT, and asked if you had any ideas or specific needs. One of our fellow Morfternighters, and dear friend, got in touch, and we discussed a very interesting use case: create an assistant that would help browse, search, and question a given set of documents.

This is a question with a very variable scope based on the size of the content database. Language models like GPT-4 cannot be easily trained on a specific set of documents, as they would not be anywhere as good. If trained on limited data. Once trained on billions of documents, they can’t easily be fine-tuned to focus specifically on a given subject. It’s doable, but expensive and complex, so way out of the realm of Custom GPTs.

For small amounts of writing, we can leverage the larger and larger context window, now in the 128k token range. Still, considering that a token is a bit less than a full word, this means that we might fit about 100k words in a prompt. That’s enough to send “Bilbo the Hobbit,” but only about 20% of “Lord of the Rings” to give a sense of scale.

Another solution, currently implemented across several products, including Jetpack, is to proceed in two steps:

• Take the question the user is asking and generate search keywords.

• Search the content database, or website, for the keywords.

• Then use the most relevant results returned by the search to respond.

What is fascinating is that until a couple of months ago, while parts of the work could be delegated to the GPT API, we still had to write a lot of code to create the piping connecting the different steps.

Today, I recorded a short video where I show in real time how that can be done only by chatting with ChatGPT.

(I apologize for the silent video, but a video is progress already, baby steps. One day, maybe I’ll start adding voice-overs… 😉).

So, there you go, it took eight minutes from beginning to end.

I shared a few more notes about building a simple Custom GPT on the blog, in case you want to understand better how we got there.

⚙️ Building a thinking partner

I recommend reading (and testing) Ness Labs’ Anne-Laure Le Cunff Tutorial: How to Build a Custom GPT with OpenAI’s GPT Builder.

Share Morfternight

👇 tl;dr

Today, we take a moment to reflect on the meaning of a new year, share some photos, think about the future of Morfternight, and look at its past.

🥳 Happy New Year!

We certainly love our calendar years. We have invented many ways of counting days, weeks, and months that compose a year. We have figured out solutions to address the small differences introduced each year (2024 is a leap year, by the way, in case you forgot). In the end, though, there is a power to a repeating cycle punctuating our existences.

We are learning, although unfortunately through climate change, that the cycle does not have to be immutable. While the dance of seasons persists (your mileage may vary based on your location on the planet), they keep following each other while not always resembling the ones from previous years.

Here’s a more optimistic approach to the point I am making. In a solar-centric model, we can consider that each year we get back to the same point we were one year before, and start a new orbit. We can zoom out, though, and consider the movement of our Sun around the center of the Milky Way, or the movement of our galaxy in relation to other ones around us. Or, as it gets further and further away, the movement of our local cluster in relation to the rest of the universe.

In this broader context, as Ethan Siegel explains in “Our Motion Through Space Isn't A Vortex, But Something Far More Interesting,” we can estimate that our planet moves at the blazing speed of 368km/s (± 30km/s) on a path that will never bring it back to the same position.

I find it very comforting to find at this cosmic scale the same structures that enable us, year after year, to become better at anything we do through the establishment of a cyclical pattern. This is where we evaluate a situation, figure out a path for improvement, move forward on that path, measure the impact of our actions, evaluate the situation again, and repeat.

It is the essence of the iterative approach to move forward through the repetition of a consistent cycle: think, act, evaluate, think, act, evaluate, think, etc.

So, as this year comes to an end, and another one begins, I’d like to wish you all the best for the future, and remind you of these two simple ideas that may help you improve it:

• Dream in decades, but iterate weekly. Years are nice to tell history, not to write the future.

• We’ll never be in the same location again, so there’s no reason to wait for next year. Next year begins today, each day.

The only day when “I’ll start next year” makes sense is today! Enjoy the celebrations, I wish you all the best, and see y’all next year!

📷 A Postcard from Vienna and one from Madrid

In the last issue of Morfternight, a few weeks ago, I explained how I was having trouble with the M11 and focusing. I ended up acting on it, I sold the M11, and have, for the moment, replaced it with two options.

A Ricoh GRIIIx, the smallest camera with a large sensor (APS-C). I have read a lot about this little one over the past few years, met several enthusiastic owners, so I ended up diving in. It is truly a pocket camera, it literally fits in my pocket, yet it is packed with all the manual settings to give you full control of the results. The only caveat so far is that it doesn’t come with a viewfinder, optical or electronic.

By default, framing is done on the back LCD screen, which would be ok if I hadn’t reached that stage in life where to see at arm’s length one needs glasses…

Ricoh sells an optical viewfinder accessory, but I doubt that will work for me. For once, it costs almost half the price of the camera, which makes little sense. In addition, it adds a bulk to the flash hot-shoe that goes against the primary quality, the small size of the GRIIIx.

The photo above was taken in Vienna with the GRIIIx, this other one was taken in Madrid with the same camera.

The second replacement is a Fuji X-H2 with the Fujinon XF56mm f/1.2 WR and a vertical grip with extra batteries. This is a very familiar setup, I used the X-H1 for quite a while from 2018 to 2020, and the 56mm on APS-C is the equivalent of a 84mm in full frame, extremely close to my darling 85mm, the focal length I prefer above all.

I barely started playing with it, I literally only went out once for one hour, but I was already rewarded with one image I truly enjoy. Here it is.

I hope you’ll like it too. Please let me know in the comments, or simply by replying to this email.

🗞️ What’s next for Morfternight?

As you may have noticed, I have been sending it out irregularly in the recent past.

I have tried to understand, and explain, in the past this dynamic where on the one hand, I don’t want this to become a forced exercise. If I don’t take pleasure in writing Morfternight, the odds of you enjoying reading it will crumble. On the other hand, I genuinely understand and appreciate the feedback I have received from readers about the importance for you to receive it regularly. So you don’t start wondering if it ended up in the spam folder, or whether I stopped sending it, or if something happened to me.

So, how do we solve this conundrum?

Investigating a bit closer to what stops me from reaching your inbox regularly, I identified the tension between the time available to write Morfternight, and the expectation I set for myself to provide every week each section of content. From the photos, to a curated selection of articles, personal thoughts on AI, leadership, or product, a full Morfternight takes a few hours. A time I gladly spend writing it when I have it, but that at times I simply can’t find every week.

As 2023 ends and 2024 begins, I decided to try something new for the third season of this digital postcard. Instead of prioritizing the volume of content, at the risk of not sending it often, I’ll prioritize sending it, even if at times it’ll be just a photo and a short note.

After all, I’ve been calling this a “digital postcard,” so maybe it is time it honors its name!

📚 Three past issues of Morfternight

As we close the second season, here are three links to past issues that mattered particularly to me.

The most shared:

The most commented:

The most liked:

🤖 Custom GPT step by step next week!

In the last issue of Morfternight I asked if you’d be interested in a step by step breakdown of how to create Custom GPTs. I apologize for the delay, but I am currently on vacation, and suddenly creating screencasts, taking screenshots, and publishing a proper tutorial started feeling a bit too much like work, and requiring too much computer time.

Instead, I’ll promise it for next week. We’ll open the third season with a focus on Custom GPTs and how-to create different sorts, from simple instructions to leveraging third-party APIs.

If you have a specific task you’d like to see a Custom GPT for, don’t hesitate to comment, or reply to this email. Who knows, maybe we’ll build just that one! :)

“Good Morfternight” was born out of working with teams distributed across time zones. It’s how we unite Good Morning, Good Afternoon, and Good Night in a unique greeting, including everyone in the same smile. :)

Share Morfternight

👇 tl;dr

Today, we dive into my catalog to analyze the EXIF data in my photos and understand why I am not happy with my photos anymore. We’ll briefly consider that the author is not getting any younger, which had been so far hidden by autofocus. We introduce the Morfternight community to my new team of GPT Agents, and we discuss briefly the importance of storytelling for humanity and the impact LLMs can have on it.

👋 Greetings from Vienna!

From Malaga to Hamburg, back to Malaga, then Costa Rica, and finally Munich, my travels have been non-stop since October. Now, with my Rimowa finally at rest, I'm home, catching my breath in the rapidly advancing world of AI. No more excuses for delaying Morfternight — it's time to dive in.

Despite a recent lull in photography, I'm not about to let that hinder our discussions. Instead, I'm turning this into an opportunity to delve into the reasons why I have no new photos to show. There's a lot to unpack, and I'm here to break it down for you.

Let’s explore these developments together in this edition of Morfternight, where we cut through the noise to focus on what truly matters.

📷 A Postcard from… Vienna

As I said, I haven’t made any good photos lately, so here’s one I took on April 1st, 2019.

I remember it particularly as it was the first one shot with a brand-new lens, the Fujinon XF 16mm f/2.8 that I had just purchased and mounted on the Fuji X-H1 I had back then. Like many photographers, one of the reasons I make photos is that I like cameras and lenses, incredibly sophisticated tools.

The problem with the passion for the gear is that the camera and lens must become almost an extension of the photographer’s hand and eye. That’s how I repeatedly end up in the same routine: the new camera I just picked up is perfect, and I’ll never change again. Until, eventually, enough time passes for me to rationalize the next upgrade. Rinse, and repeat.

So, today, I decided to dive into my photo catalog. Using the EXIF data (which stands for Exchangeable Image File Format, basically the information stored with each digital photograph about the camera, settings, and more), aggregate the information I have available about the last seventeen years or so of photography. The EXIF data aggregates the information I have available about the last seventeen years or so of photography, since digital cameras became serious tools.

Breakdown by camera

The table below shows the cameras I owned since 2006. The Start and End dates come from the first and last photo in my catalog that was shot on each camera.

The number of photos is not the number of shots (that would be much bigger), but the number of photos I considered good enough to keep.

The first thing that is immediately visible is that there are three primary periods:

• 2006-2014: The first eight years account for 78% of the photos. Looking at them, the explanation is simple, the quality bar for keeping them was much lower back then.

• 2015-2020: Very few photos during these five years, only 8% of the total. Not randomly, this is the time when the iPhone camera started getting good, and for a while, I focused primarily on using it.

• 2021-2023: On January 1st, 2021 I decided I was going to take photography seriously again. The numbers may seem lower, but they don’t tell the full story. The reality is that I raised the bar much higher in terms of quality, nowadays, I only keep the photos I publish.

Focusing only on the recent years, I find some interesting, and surprising, insights. What's your experience with different camera models? Do you find yourself gravitating towards certain types over others? I'd love to hear your thoughts and experiences in the comments or in your replies.

First, I would have expected these numbers to tell the opposite story. The Q2 and the M11 are by far easier to carry around. The former doesn’t have an interchangeable lens, the latter does, but I only possess one, the Summilux 50mm f/1.4. They are both compact and light.

That’s the thing, though, I do carry the M11 almost everywhere I go, as I did for each of the three other cameras, so that’s not a differentiating criterion justifying the results.

If anything, I would say that paradoxically, the size and weight played in favor of the much heavier SL2-S. I find carrying a large, heavy, camera strapped around my neck or my shoulder very uncomfortable, which is why I had a tendency to always have the SL2-S in my hand.

Breakdown by focal length

To dig deeper, I checked the distribution of images by focal length.

It’s important to note that the Fuji cameras have an APS-C sensor with a crop factor of 1.5, so the focal lenses should be multiplied by 1.5 when used on such cameras. This affects the 35mm and 50mm photos: with the former all made on Fuji APS-C cameras, and the latter all on Full Frame sensors, they are de facto one and the same, with a total of 1749 photos.

The 85mm and 90mm are very close to each other on Full Frame sensors, but the 90mm on the Fuji APS-C sensors is equivalent to a 135mm in Full Frame. Of the images shot with a 90mm, 108 were made on the SL2-S (Full Frame), and 472 on APS-C sensors.

Looking at these numbers, the first obvious conclusion is that I enjoy using lenses with a focal length of 50mm or more, as they represent the vast majority of the images I keep.

This is the problem with the Leica Q series and the Fuji X100 series. All incredibly good cameras, pleasant to use and very high quality, but stuck with respectively a 28mm and a 23mm (equivalent 35mm) lens.

So, what’s the issue with the M11?

Simply put, it’s manual focusing. I am not 25 anymore, and my reflexes (and let’s be honest, also my eyesight) are not what they used to be. Not only, but I recently realized that I tend to give up without even trying. If an interesting subject in the street is moving, I won’t even take the shot, knowing I’ll miss the focus anyway.

Of course, I could use zone focus.

Zone focus is a technique based on depth of field, where you set the lens to focus at a certain distance and adjust the aperture to get a range of focus. For instance, by setting the lens to focus at 3 meters, then closing the diaphragm to 8, anything between 2.36 and 4.13 meters from the camera is in focus. The issue is that the whole reason I like longer lenses is the bokeh I can obtain by opening them wide. In the same configuration, open at 1.4, the depth of field only goes from 2.87 to 3.15 meters. This now requires being a lot more precise.

I have been trying to get used to measuring distances at a glance, but so far, I haven’t reached great results.

Conclusion

A part of me, that just wants to make good photos, wishes I didn’t listen to the other part that likes trying new cameras. The X-Pro3 was already the one I was getting the best results with, and had I kept it since 2021, my familiarity with it would only have increased, making it easier and easier to achieve the results I want. However, the sheer numbers are not telling the whole story. When I look at the images I prefer, the distribution is almost even across the X-Pro3, the Q2 Mono, and the SL2-S. Once more, the M11 is far behind.

As I continue to explore and grow in my photographic journey, I look forward to sharing new perspectives and breakthroughs with you. The path of creativity is ever-evolving, and each step brings new insights and joys. Stay tuned for what's next on Morfternight, and let's keep the conversation going!

Meanwhile, if you are facing similar questions, I can only encourage you to also look at the EXIF records of the photos you prefer, draw your own conclusions. If you feel like it, share them with me, I’d love to hear other points of view and opinions!

🤖 Much ado about nothing

A couple of weeks ago, OpenAI announced a new ChatGPT capability, allowing anyone to create customized GPTs to perform specific tasks without having to instruct them each time.

In the short time since the announcement, OpenAI’s board of directors fired the company’s CEO, named an interim CEO, fired the interim CEO, hired a new CEO, got fired along with the new CEO. The old board was replaced by an (almost) new board that re-hired the original CEO.

🦾 Here come the GPTs

During the same time, I started experimenting with the new GPTs, and I now have a small team of AI agents to help me:

• Agent Schedule — Has access to my calendars via the Google Calendar API, and can list the events of the day or the week, summarize my week, identify the busiest day, things like that.

• Agent Notes — I can upload a photo of a page of my paper notepad, it will transcribe it, summarize it, and provide both to me.

• Agent Tana — Works like ChatGPT, but I can ask it to send to Tana, my Personal Knowledge Management (PKM), any part of my chats by using the Tana API.

• Agent Feature — I give it the address of one of my posts on the blog, and it creates a featured image to go along with the post, based on its content.

• Agent Morfternight — It has access to the archives of Morfternight, so I can ask whether or when I already mentioned a subject, what I said about it, and use its help to make sure I don’t ramble inconsistently :)

Creating GPTs is quite straightforward, but if you think it would be useful for me to describe step by step how I created these, please let me know, and I’ll gladly share in the next issue of Morfternight.

📚 Storytelling

Once upon a time…

This (or it’s many variations in other languages), is how stories start when we are kids.

As children, stories are how we learn about the world, about good and evil, about morals, ethics, honesty, and all other values we need to understand so we can live in our society.

We consider that our civilization started 5,000 years ago when we invented writing.

We consider the printing press one of the most important inventions of our society.

Telegraph, telephone, radio, television, computers, internet, smartphones: all modern means of communications, all about telling stories.

We know that the past doesn’t exist, and neither does the future, but stories tell us about both.

History is written by the victors. The future by writers, politicians, marketers.

Teaching, selling, fundraising, building products, campaigning are all based on storytelling.

Politics, the economy, peace, and war all rely on someone telling a story and whether others believe in it or not.

Artificial Intelligence, Machine Learning, and more specifically Large Language Models, will play a crucial role given their ability to generate, transform, and summarize text.

Even if they didn’t represent a step towards Artificial General Intelligence, or Superintelligence, LLMs translate our stories, connect them, make sense of them, repeat them, improve them, and help us write more of them.

In this context, they are the most important tools of our generation.

Share Morfternight

👇 tl;dr

Today, we pass a huge milestone, change our format a bit, introduce paid subscriptions, talk about calendars, time management, and performance reviews. We also learn about using ChatGPT to ask DALL-E to draw stuff.

👋 Good Morfternight!

Greetings from Vienna, my friends!

I've just returned from Hamburg and Málaga round two! And what better homecoming than sharing the monumental 100th issue of Morfternight with you? I hope you're as jazzed to read it as I am to send it your way.

Now, if you've been around for a while, you know I'm all about growth and change. So, let's dive into some tweaks and twirls I'm making to the newsletter:

Firstly, I'm shaking up our usual "Three Places to Visit" section. No more just listing articles with a sprinkle of my thoughts. Starting now, I'll weave my insights and recommendations seamlessly. One cohesive experience is coming right up!

Secondly, we're introducing paid subscriptions! No need to panic—Morfternight stays free for all. Subscribing is just a way for you to show some love and support. And, of course, you'll have my endless gratitude.

Lastly, about the French translation experiment. It didn't quite hit the mark, and that's okay. Your clicks (or the lack thereof) have spoken. So, we're shelving it for now, but always keeping an eye out for what serves you best.

Let's keep this journey engaging and ever-evolving, shall we? To the next 100!

📷 A Postcard from… Vienna

At the distinctive S curve where Neubaugasse meets Siebensterngasse and then Westbahnstraße, an old tram travels line 49. These old bims have been phased out from line 49 in favor of the newer models. They remain a nostalgic reminder of Vienna’s past. I’ll surely miss them after 2026 once they all have been retired.

📆 Calendars and Time Management

Calendars, as everyday tools, have not kept pace with the rapid evolution of our digital lives. They remain largely a reflection of their paper ancestors, with minimal innovative features to support the modern, multi-faceted schedules we juggle daily. Julian Lehr dives into this topic in his essay “Multi-layered calendars — Traveling through time in three dimensions,” exploring the untapped potential of calendars when envisaged as multi-layered canvases for managing our time. He laments the ironic stasis of digital calendars—our “time machines” stuck in the past.

Lehr envisions calendars as three-dimensional grids where different types of time blocks coexist, each layer representing a different kind of activity, be it meetings, tasks, or even periods of non-availability. He critiques the current limitation where calendars treat meetings and tasks as the same, ignoring the inherent differences between them. His discourse extends to how new tools like Amie attempt to blend to-dos with calendar events, although he believes this integration needs to go deeper.

Similarly, in “Time Management, “I shared a transformation in managing tasks and notes over the years, particularly the shift from traditional to-do lists to utilizing the calendar as a dynamic task list. This transition, inspired by others’ practice, not only improved time management but also significantly reduced personal stress. By allocating specific time slots for tasks on the calendar, I found an effective way to ensure that each task receives the attention it requires while also promoting delegation when necessary.

Furthermore, I have incorporated Akiflow, a tool that epitomizes the blend of task management with calendar functionalities. Akiflow facilitates easy scheduling of tasks with designated time slots, priorities, and labels right from a command bar. It bridges the gap between to-dos and time blocks, allowing a seamless transition of tasks into scheduled activities. By consolidating tasks and calendar events into a unified interface, Akiflow brings a semblance of the multi-layered calendar concept to life, making it easier to visualize and manage the day’s agenda.

The discussion by Lehr and the practical adjustments in my approach underscore a collective yearning for calendars to evolve beyond mere date trackers. They should morph into dynamic platforms that mirror the complexity and interrelatedness of our daily activities. As we blend tasks, meetings, and other facets of our schedule, we move closer to the concept of a multi-dimensional calendar, opening doors to a more organized and less stressful management of our daily lives.

✍️ Need any Help with Performance Reviews?

If there is an exercise that often induces a mixture of anxiety and reflection, it’s the Performance Review. It’s that time when we pause to reflect on our journey over the past months. It’s a time that I candidly have found hard to prioritize, particularly when it comes to self-reviews. The task of distilling months of work, achievements, and areas of growth into a structured dialogue often finds itself at a low priority among other pressing tasks. Yet, as someone passionate about continuous development, I recognize its importance.

A recent article from First Round, “Elevate Your Performance Review Conversations with these 12 Expert Tips,” offers instrumental perspectives, templates, and tips. It sheds light on how the dread surrounding performance reviews can be morphed into a constructive and enriching experience. From honing your narrative with a personal press release to structuring the conversation in a way that incites growth, it’s loaded with actionable insights for both managers and direct reports.

In parallel, I reflected on my previous piece on the essence of one-on-one meetings. There’s an unspoken rhythm to these interactions, which, when tuned well, not only elevates the team’s harmony but eases the process of performance reviews.

The blend of continuous one-on-one meetings and the insightful tips from the First Round’s piece has equipped me with a fresh lens to approach performance reviews with a structured yet empathetic lens. It’s a shift from viewing it as a taxing obligation to an avenue for aligning personal and organizational growth.

🤖 ChatGPT Writes Better Prompts for DALL-E

Last week, I found myself knee-deep in crafting slides for an upcoming presentation. I needed three illustrations—stat!

Now, I’ve been dabbling with MidJourney for a while, creating images that are both fun and captivating. But here’s the catch: If you’re trying to generate a consistent series of images, MidJourney falls short. It’s a hassle, to be honest.

So, what did I do? I gave DALL-E 33 via ChatGPT a whirl. First up, I asked for:

An illustration of a couple of arrows going up and to the right, like a chart.

Voilà, out popped this:

Next, I typed:

The same, but with many arrows.

Here’s what I got:

Finally, I made one more request:

The same, but with one big arrow.

And BOOM!

Sure, the illustrations aren’t perfect triplets. They don’t all share precisely the same aesthetic. But guess what? The top-right option in each set? Nailed it for me.

Here’s the real kicker: Using ChatGPT’s DALL-E is a breeze. It takes my prompts, talks back, and keeps the conversation going. In contrast, MidJourney would’ve had me wrestling with detailed prompts and micro-adjustments.

Want to dive deeper into the mechanics of these prompts? Simon Willison unpacks it all in his recent post, “Now add a walrus: Prompt engineering in DALL-E 3.”

Welcome to Morfternight, the digital postcard
about Photography, Leadership, and AI.
If you enjoy it, share it with your friends.

Share Morfternight

👇 tl;dr

Today, we navigate through Vienna’s concealed courtyards, reflect on the end of a popular product and the integrity of academic trails, journey through the evolution of machine learning, and dive hands-on into the visual capabilities of ChatGPT-4.

👋 Good Morfternight!

Greetings from Vienna!

This is Morfternight #99, the last with two digits. We’ll hit three figures in two weeks as next weekend I’ll be at the ICANN conference in Hamburg, Germany, and then I’ll meet a couple hundred fellow Automatticians in Spain after that. With seven days of travel starting Friday, I know I won’t have time to prepare the issue between now and Thursday.

In the meantime, enjoy Morfternight #99, see y’all for Halloween! 🎃

📷 Photo of the week

Have you ever been to Vienna? If so, you might have noticed something unusual—most of the city’s building façades aren’t visible from the streets. It’s a delightful paradox! Instead, a maze of quiet courtyards tucks them away, offering locals an escape from the urban buzz. You won’t spot these hidden gems unless you’re looking at aerial photos or stumbling upon the few open to the public, connecting parallel streets like secret passageways.

What I love about these courtyards isn’t just the quiet. It’s also their lushness. You’ll find them filled with trees, ivy, and greenery. It’s like stumbling upon a micro-forest in the heart of the city.

🗺️ Three places to visit

① Can you believe it’s been a decade since Google pulled the plug on its RSS Reader? The tale behind this much-missed product is fascinating, and ‘Who Killed Google Reader?’ tells us how it could’ve been a game-changer for Google. What’s your go-to tool for keeping tabs on your favorite sites?

I’ve got a soft spot for the WordPress.com Reader. It’s from Automattic, so of course, I’m a fan. It’s a straightforward, no-frills solution for devouring website content. But if you’re after more—a tool that lets you save highlights, Twitter threads, videos, PDFs, books, and even emails—Readwise Reader is your best bet. I’m a fan of both, using each for different purposes.

Do you use either of these or have you found another gem worth sharing?

② They Studied Dishonesty. Was Their Work a Lie? Here’s a long piece in the New Yorker that made me sad. I read most of Dan Ariely’s books and have found his work interesting and effective in its real-life applications. Apparently, Ariely and another behavioral economist, Francesca Gino, took some liberty with the data they used for their research.

I don’t have enough information to have an opinion. Still, I’d wait a second before throwing Ariely’s work into the trash because I have found that many of his recommendations and findings worked in my work or personal life.

③ Ever feel like discussions about machine learning are either too technical or too sensational? Finally, here’s a conversation that strikes a good balance. It traces the past decade’s evolution in machine learning that gave us today’s Large Language Models (LLMs) and also dives into its broad societal implications—from politics to economics.

What sets this discussion apart? It’s a nuanced conversation, a rare thing in this polarized world. It’s a refreshing deep dive that appreciates the complexities of the matters discussed.

What are your thoughts on how machine learning is shaping our world? Do you find the conversation as illuminating as I do?

🤖 Hands-on session with ChatGPT-4 vision

OpenAI enabled the new Image feature on my ChatGPT account a few days ago. In a nutshell, it means I can now upload images and ask ChatGPT to work with them. I started playing with it, and I want to share with you the results, some of which are impressive. I took screenshots of the chats so you can visualize them here without having to click on links and also because sharing chats with images is not yet supported by ChatGPT.

First, a few failures…

1. ChatGPT fails the geography test.
   I mean, it got Brazil right, but at this point, I am not sure if it’s random. I tried with Europe and got the same type of results. This surprised me a bit, as I imagined the tool being trained on maps.

2. No, but seriously, where is Waldo?
   This one is interesting because it shows how similar Waldo’s images are to each other from an LLM’s perspective. ChatGPT clearly answers are based on different pictures than the one I uploaded.
   
   The conversation was a bit longer, but even after detailing the steps to follow, it got it wrong each time, with precise descriptions of Waldo’s surroundings that are not in this image.

Then, one that is almost right and yet already impressive.

1. Where was this photo taken?
   This is a photo of Camogli, just over 10km east of Genova. The response is very accurate, considering that no recognizable landmark is visible.

And now, the excellent answers

1. Where was this photo taken? (again)
   With recognizable landmarks, ChatGPT nails it.

2. Recognizing a molecule of glucose
   It is glucose, indeed.

3. Solving Integral calculations
   The result is correct, by the way.

4. Generating ALT text for images to improve website accessibility.
   This is a pretty big one. Visually impaired people using special reader devices to browse the web need images to be described with words.
   
   It’s often overlooked because it is very time-consuming. Using AI to solve this problem is a clear, immediate benefit of the technology.

5. Criticizing a photo and providing suggestions for improvement.
   The analysis is not revolutionary, but ChatGPT makes a great point about trying to move the horizon away from the middle of the image.

I barely scratched the surface over the weekend. I’ll try more things over the following days.

What do you think? Have you also tried this feature? Do you have ideas you want me to try?

🇫🇷 La version française de Morfternight #98 est disponible ici 🇫🇷

👇 tl;dr

Today, we get back on track. We celebrate many nice things, look at a white rose, reflect on whether AI will destroy or save humanity, and finally, on replacing binary alternatives with ternary progressions.

👋 Good Morfternight!

Well, it’s been a while… precisely a month.

During this month, a few things happened. First, we closed a poll in which French won by a hair against Italian as the language to translate Morfternight. Now, I need to figure out how this will work precisely so you can prepare for some iterations and requests for feedback.

We also passed 5,000 Morfternighters. FIVE THOUSAND!

We celebrated two years of Morfternight!

I could dwell on the fact that there are 104 weeks in two years, and this being Morfternight #98 instead of #105, I missed seven. Instead, I’ll celebrate the fact I sent 97 out of 104, a whopping 93.3%

📷 Photo of the week

I haven’t taken many photos lately, but there’s a good reason. Taking pictures while walking with someone else is a tricky business. We photographers often stop, sometimes backtrack, or stand there for an eternity. Detours? Oh, there are plenty.

Photography is a solo gig, often at odds with other activities. These days, I walk in the company of E. or P. As much as I love making photos, I appreciate the company even more.

I often carry my camera, but it’s common for me to return home without using it. A few weeks ago, as we walked through the Ferstel Passage in the first district of Vienna, a rose caught a ray of morning light in the shadows, and I was happy with the result.

🗺️ Three places to visit

Not the most recent articles today. I selected those a few weeks ago, but I still want to share them to continue exploring whether AI represents an existential threat or the next step of our evolution.

① Let’s keep digging into the complex world of AI’s pros and cons. Today, I’ve got two relevant articles for you from The Guardian. One talks about how AI might ruin us: “Five ways AI might destroy the world.” The other paints a brighter picture: “Five ways AI could improve the world.”

Steve Rose interviews top experts in both articles. What’s intriguing? Three experts show up in both pieces, proving just how complicated forming an opinion on AI can be.

② Let’s momentarily step away from doomsday scenarios about AI wiping out humanity. Instead, let’s focus on another idea: AI might kill the web. James Vincent argues in his Verge article, “AI is killing the old web, and the new web struggles to be born.”

I find myself nodding in agreement with the first part of Vincent’s claim. Just as the web displaced magazines and most newspapers, AI is poised to overhaul the digital landscape we’re familiar with.

I can’t help but feel a sense of excitement about what we’ll create to replace it.

③ Meanwhile, there’s some intriguing research out there on prompt optimization. One study found that simply telling an AI model to “take a deep breath” led to significantly higher math scores.

While surprising, is it really that unexpected?

I’ve noticed—and maybe you have, too—that being polite with language models like GPT-4 yields better results. It’s not so much that the AI responds to politeness as these prompts provide specific context, steering the model in a better direction.

🤖 Incremental progress, evolution, and revolution.

We often face a seemingly binary choice: focus on the short or long term.

This problem appears across various aspects of life—be it in building a product (ship something fast or something good?), managing personal finance (enjoy money now or later?), making business decisions (prioritize short-term gains or long-term value?), or even in health choices (enjoy life now or later?).

F. Scott Fitzgerald famously said that the test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function.

I do not believe these are exclusive choices. I’ll further suggest they can be harmonized into a three-step progression.

A great example comes from the process of building a sales pipeline. The largest opportunities often take a significant amount of time to mature due to the scope of the projects, the size of the companies involved, and the substantial amounts at stake. However, awaiting these to materialize isn’t a viable strategy as it could stifle a company’s cash flow and operational momentum.

A balanced sales pipeline, on the other hand, encompasses short, medium, and long-term prospects. The short-term prospects generate a consistent influx of smaller sales; medium-term prospects contribute a regular stream of moderate-sized deals, all building the runway to close the larger, long-term contracts.

This rings true in software development as well. The most successful teams master the balancing act of maintaining their existing codebase to offer a flawless user experience, iterating on the current feature set to stay relevant while simultaneously dedicating resources to deeper, groundbreaking innovations that elevate them ahead of the competition.

New technologies are not exempt from this pattern.

Instead of deliberating whether the AI revolution is already here or far beyond the horizon, I prefer to view AI as a set of tools. These tools can reliably enhance my efficiency at the margins immediately, amplifying my output in my areas of expertise.

While I can’t fully trust them in less familiar fields—they sometimes fabricate alternative realities—I am optimistic. As they improve and I become more adept at utilizing them, I anticipate that, in the medium term, they’ll enable me to tackle a broad range of tasks comfortably. In the long term, there is a possibility they get so efficient that I won’t be needed anymore, which doesn’t mean I’ll be dismissed. I still know a few good jokes…

Aujourd'hui, nous revenons sur la bonne voie. Nous célébrons de nombreuses belles choses, regardons une rose blanche, réfléchissons à savoir si l'IA va détruire ou sauver l'humanité, et enfin, sur le remplacement des alternatives binaires par des progressions ternaires.

👋 Good Morfternight!

Eh bin, ça faisait un moment… précisément un mois.

Durant ce mois, il s’est passé quelques trucs. Tout d'abord, nous avons clôturé un sondage dans lequel le français a gagné de justesse contre l'italien comme langue pour traduire Morfternight. Maintenant, je dois déterminer comment ça fonctionnera précisément. Préparez vous à quelques experiences et demandes de retours. Vous êtes en train de lire la première version française de Morfternight. La traduction est assistée par ChatGPT.

Nous avons également dépassé les 5 000 Morfternighters. CINQ MILLE !

Nous avons célébré deux ans de Morfternight !

Je pourrais m'attarder sur le fait qu'il y a 104 semaines en deux ans, et que ceci étant Morfternight n°98 au lieu du n°105, j'en ai raté sept. Au lieu de ça, je vais célébrer le fait que j'en ai envoyé 97 sur 104. 93,3% c’est énorme !

📷 Photo de la semaine

J’ai pas fait beaucoup de photos dernièrement, mais il y a une bonne raison. Faire des photos en marchant avec quelqu'un d'autre est une entreprise délicate. Nous, les photographes, nous arrêtons souvent, faisons parfois demi-tour, ou restons là pendant une éternité. Des détours ? Il y en a beaucoup.

La photo est un travail en solo, souvent en contradiction avec d'autres activités. Ces jours-ci, je marche en compagnie de E. ou de P.

Autant j'aime prendre des photos, autant j'apprécie la compagnie encore plus.

J’emmène souvent mon appareil photo, mais il m'arrive fréquemment de rentrer sans l'avoir utilisé. Il y a quelques semaines, alors que nous traversions la Passage Ferstel dans le premier arrondissement de Vienne, une rose a capté un rayon de lumière matinale dans l'ombre, et j'étais content du résultat.

🗺️ Trois endroits à visiter

Pas les articles les plus récents aujourd'hui. Je les ai sélectionnés il y a quelques semaines, mais je veux toujours les partager pour continuer à explorer si l'IA représente une menace existentielle ou la prochaine étape de notre évolution.

① Continuons à creuser dans le monde complexe des avantages et des inconvénients de l'IA. Aujourd'hui, j'ai deux articles pertinents pour vous provenant de The Guardian. L'un parle de comment l'IA pourrait nous ruiner : “Five ways AI might destroy the world.” L'autre présente un tableau plus optimiste: “Five ways AI could improve the world.”

Steve Rose interviewe des experts de renom dans les deux articles. Ce qui est intrigant ? Trois experts apparaissent dans les deux pièces, prouvant à quel point il peut être compliqué de se forger une opinion sur l'IA.

② Mettons momentanément de côté les scénarios catastrophiques sur l'IA éradiquant l'humanité. Concentrons-nous plutôt sur une autre idée : l'IA pourrait tuer le web. James Vincent avance dans son article sur Verge, “AI is killing the old web, and the new web struggles to be born.”

Je me retrouve à acquiescer à la première partie de l'affirmation de Vincent. Tout comme le web a déplacé les magazines et la plupart des journaux, l'IA est prête à refondre le paysage numérique que nous connaissons.

Je ne peux m'empêcher de ressentir un sentiment d'excitation quant à ce que nous créerons pour le remplacer.

③ Pendant ce temps, il y a quelques recherches intrigantes sur l'optimisation des prompts. Une étude a découvert que simplement dire à un modèle d'IA de "prendre une grande respiration" conduisait à des scores en mathématiques nettement plus élevés.

Bien que surprenant, est-ce vraiment inattendu ?

J'ai remarqué - et peut-être l'avez-vous remarqué aussi - qu'être poli avec des modèles de langage comme GPT-4 donne de meilleurs résultats. Ce n'est pas tant que l'IA réagit à la politesse, mais ces prompts fournissent un contexte spécifique, orientant le modèle dans une meilleure direction.

🤖 Progrès incrémental, évolution, et révolution.

Nous sommes souvent confrontés à un choix apparemment binaire : se concentrer sur le court ou le long terme.

Ce problème apparaît dans divers aspects de la vie, que ce soit dans le développement d'un produit (lancer quelque chose rapidement ou de bonne qualité ?), la gestion des finances personnelles (profiter de l'argent maintenant ou plus tard ?), la prise de décisions commerciales (prioriser les gains à court terme ou la valeur à long terme ?), ou même dans les choix de santé (profiter de la vie maintenant ou plus tard ?).

F. Scott Fitzgerald a célèbrement dit que le test d'une intelligence de premier ordre est la capacité à garder deux idées opposées en tête en même temps tout en conservant la capacité à fonctionner.

Je ne crois pas que ce soient des choix exclusifs. Je suggérerai de plus qu'ils peuvent être harmonisés en une progression en trois étapes.

Un excellent exemple provient du processus de construction d'un pipeline de ventes. Les plus grandes opportunités prennent souvent beaucoup de temps pour mûrir en raison de l'ampleur des projets, de la taille des entreprises impliquées, et des sommes importantes en jeu. Cependant, attendre qu'elles se matérialisent n'est pas une stratégie viable car cela pourrait étouffer la trésorerie et l'élan opérationnel d'une entreprise.

Un pipeline de ventes équilibré, en revanche, englobe des perspectives à court, moyen et long terme. Les perspectives à court terme génèrent un afflux constant de ventes plus petites ; les perspectives à moyen terme contribuent à un flux régulier d'affaires de taille modérée, tout en construisant la piste pour conclure les contrats plus importants à long terme.

Cela est également vrai dans le développement de logiciels. Les équipes les plus performantes maîtrisent l'art de maintenir le code existant pour offrir une expérience utilisateur impeccable, itérant sur l'ensemble des fonctionnalités actuelles pour rester pertinentes tout en consacrant simultanément des ressources à des innovations plus profondes et révolutionnaires qui les placeront en tête de la concurrence.

Les nouvelles technologies ne sont pas exemptes de ce modèle.

Au lieu de délibérer si la révolution de l'IA est déjà là ou bien au-delà de l'horizon, je préfère voir l'IA comme un ensemble d'outils. Ces outils peuvent permettre l'amélioration de mon efficacité aux marges immédiatement, amplifiant mon rendement dans mes domaines d'expertise.

Bien que je ne puisse pas entièrement leur faire confiance dans des domaines moins familiers - ils fabriquent parfois des réalités alternatives - je suis optimiste. Au fur et à mesure qu'ils s'améliorent et que je deviens plus compétent pour les utiliser, je prévois qu'à moyen terme, ils me permettront d'aborder confortablement une plus large gamme de tâches. À long terme, il est possible qu'ils deviennent si efficaces que je ne serai plus nécessaire, ce qui ne signifie pas que je serai inutile. Je connais encore quelques bonnes blagues...

Today we reap the fruit of a week off work, with a photo from outside the city, a selection of links out of our usual areas, a new website and a relaxed chat about AI. Also, you have 24 hours left to vote in the translation poll.

👋 Good Morfternight!

Hello from climate-challenged Vienna, where summer isn’t reading the room, and temperatures are back to 30º Celsius this week.

Last week I shared a poll asking you to vote for French, or Italian, as the language you want to see this newsletter translated into, and I sent out Morfternight a bit late - on Monday. Because of this, and since I'm on time this week, there's still one day left to vote. I'm leaving the poll here for you to make sure your voice is heard.

The race is incredibly tight, 52% in favor of French at the time of this writing, so every single vote counts. Don't miss your chance to shape the future of Morfternight!

Time to vote! Next issue will be translated into the winning language.

📷 Photo of the week

“Demokratie muss jeden Tag erneuert werden,” or, in English, “Democracy must be renewed every day,” is the slogan illustrating the barriers around the renovation of the Presidential Chancellery, the Hofburg Leopoldine wing in which the Austrian President lives and works. It was always a bit tongue-in-cheek to use such a strong statement to explain some building renovation, but last week, with those five soldiers parading in front of it, it briefly reached into the register of dark humor…

I took some time off this week, and on Thursday, we decided to drive to the Grüner See, a small but very charming lake in Styria, southwest of Vienna.

We’ll have to visit again next May or June, as the lake’s primary water source is snow melting in the spring, reaching up to 12 meters in depth. This week, after the hot summer we just had, it was barely two to three meters, and the waters had receded enough to split the lake into two separate bodies of water.

This is a view of the Messnerin, a 1,835-meter-tall mountain behind the lake. I hear the view from the top is incredible. I’ll have to get in better shape if I want to experience it, though, as there’s a 1,100-meter altitude gap between the last parking spot and the peak.

🗺️ Three places to visit

This is not the first, nor the last, mention of the fact that I was off this past week, but it's important that I mention it because it explains why my mind wasn't focused on the usual. As a consequence, I want to share with you three little gems that have nothing to do with Leadership, Product, AI, or Tech.

I hope you won't mind this little detour, or perhaps it'll even make you smile.

① I shared in Morfternight #41 a fascinating 1983 video series, “Fun to Imagine,” in which Richard Feynman explains science in a way that is fun and accessible to everyone. A few days ago my FYP1 presented me with excerpts of a lecture Professor Feynman delivered at Cornell University in 1964: “The Distinction of Past and Future.”

I tracked it down on YouTube, the usual go-to when something interesting pops up on TikTok, and discovered it was part of a series dubbed the “Messenger Lectures,” a yearly Cornell tradition where a guest speaker presents.

I was extremely pleased to realize that not only were these available on Youtube, but that Cornell University published them, along with synchronized transcripts here: Feynman's Messenger Lectures.

I haven’t watched them all yet, but “The Distinction of Past and Future” is exceptionally insightful and fun, so I can only recommend the series.

② Another throwback, this time to Morfternight #54 where I shared an episode of Leonard Bernstein’s Young People's Concerts.

I started revisiting the entire collection of more than 50 episodes, accessible on YouTube as a playlist: Leonard Bernstein - Young People’s Concerts (All Episodes)2.

Bernstein does with Music what Feynman does with Physics, making it accessible to everyone, demonstrating, as if it was necessary, that people who truly master a field can explain it to anyone, including children.

Listening to these two speak in such simple and clear terms is such a relief in a world full of jargon and acronyms thrown around by many to mask the lack of depth of their knowledge.

③ I think the first time I heard Nick Cave & The Bad Seeds was in the soundtrack of 'Until the End of the World,' the Wim Wenders movie.

It was about 30 years ago, and Nick Cave has remained a pillar of my music library since. I still have all the CDs, although I don’t have a way to listen to them anymore.

But I digress…

So naturally, this article caught my attention earlier this week: Nick Cave on Creativity, the Myth of Originality, and How to Find Your Voice.

Now, Maria Popova’s article is excellent, as is her whole site, The Marginalian.

The real gem hidden there, though, was the discovery of The Red Hand Files, a website that looks almost like a blog, except for the fact that each post is Nick Cave’s response to one (or several) letters from his fans and readers.

I started reading through the latest ones and found Cave’s writing and viewpoints both profoundly touching and remarkably balanced.

I think that when I grow old, I’d love to do that — receive questions from people and answer them publicly.

🌐 A new website at paolo.blog

Did I mention I was off this week?

I know, I know, I am joking. It’s the third time I mention this fact, but that’s not because I am so proud of taking a week off. Rather, it's to explain how I finally managed to do a number of things I wished I had time for.

One of these was refreshing the blog.

I have been going through this exercise relatively often over the past few years. It serves as both an exercise for me to stay up to date with the changes in the WordPress editor – which has evolved to the point where I can create almost anything I envision just with a mouse, without needing to type any code – and a testament to the fact that this evolution of the site editor has made the practice a lot more amusing than it used to be.

With this version, I still use a monochrome color palette, not only because I love it, but also because I find it accentuates my black and white photos beautifully.

I have tried to maximize the space large displays offer, while, of course, ensuring responsiveness on small mobile device screens. I also worked to make the text as readable as possible, despite a relatively dark theme, to complement the photos effectively.

I do have a light version of the theme ready to go; I just need to figure out how to allow visitors to toggle between dark and light modes.

The only area that is not complete is the Newsletter section. It receives copies of this newsletter after publication, and I am behind in transferring them over.

Who knows, in the near future, I might even migrate this newsletter entirely to the site, consolidating everything in one place.

If you visit the blog, which would make me very happy, please share any feedback you have or point out anything you find broken.

🤖 How do LLMs respond to questions? Insights from a Casual Conversation

In a casual chat recently, a dear friend brought up some pretty profound questions regarding Large Language Models (LLMs) that I found worth sharing with you all. Let's dive in to unravel the complexities of LLMs through a laid-back conversation.

Understanding the Core Functionality of LLMs

After watching Geoffrey Hinton's video (shared in Morfternight #94), I understood a bit about neural networks for image and text recognition. However, I'm having trouble understanding how an LLM can answer a question. I get that it can complete a sentence or phrase, but how does it choose the first word or sentence to start the response? It seems as if it is trained to complete questions with full answers, not just a word followed by another. What's your take on this?

You're right, it does seem a bit mysterious at first. In essence, an LLM doesn't know that you've asked a question or even what a question is. What it does know is that you've given it a text to complete recursively. It statistically selects the most probable word to start the response based on the input and continues to build on it until it identifies a "end of response" token, signaling it to stop. It's all about probabilities and statistical data at play here.

Delving Deeper into the Functionality

Hmm...so, statistically speaking, in the vast collection of human-written text, the text following a question is most often an answer? That doesn't quite feel intuitive, especially considering how we humans, including us, tend to make assertive statements rather than ask questions, right?

It's indeed a matter of large numbers. That's why the "responses" tend to converge towards the average, inheriting cultural biases and occasionally hallucinating. Essentially, the LLM responds with what closely resembles a probable answer to your query.

This 'probability' is variable: each request initiates with a random "seed" to prevent identical responses to the same question.

You can also adjust the "temperature" which dictates how much the LLM can deviate from the ideal choice for each token – a setting that accumulates since each chosen token is then used to predict the following one.

Behind the user-friendly façade of ChatGPT, these controls are not accessible but when you deal with LLMs more directly via the APIs they are, making it essential to craft your query carefully for desired results.

In other words, if you want better answers you need to make sure that the answer to your question is the most likely output. For a deeper insight, you might find this article quite illuminating: Prompt Engineering Guide on Generative AI & LLMs.

Addressing Complexities & Fears

To go back to the original question, and how does an LLM know how to answer a question, or that it is a question, or what is an answer, I think it’s a bit more complex than just imagining that statistically the word that follows a question is the first one of the answer.

In the gigantic multi-dimensional matrix where LLMs classify words and concepts, elements of a question are situated near those of a potential answer, and this proximity is a crucial factor in the LLMs choice of words.

But how does an LLM categorize and retrieve the correct elements? This part is quite scary, because we do not understand how it works. To use a great metaphor I recently read, the uncertainty surrounding their actual functionality places us in the shoes of cavemen discovering fire – utilizing it without fully grasping how it works.

This ignorance also brings up deeper questions about our intelligence – whether it's something fundamentally distinct or just a grander scale version of what these machines can do.

Today we celebrate, we run a poll, we launch a rewards program, we look at pigeons, we listen to a podcast interview about product management, we read a long article about Strategy, and a short one about AI.

👋 Good Morfternight!

Hello from beautiful Vienna, where the summer is gracefully giving way to pleasant weather that I hope will be with us for many more weeks.

We have reasons to celebrate in this issue!

Not only are we celebrating the growth of our community with over four thousand Morfternighters, but also gearing up to celebrate my birthday in just two days!

To mark these joyful occasions with style, I've got two exciting opportunities lined up for you.

Firstly, let's talk about our brand-new referral program, the details of which you'll find at the end of this issue. In a nutshell, if you enjoy Morfternight, feel free to share it with your friends. If they subscribe, you earn delightful rewards.

These range from a heartfelt thank you note featured on Morfternight, complete with a link to your blog, newsletter, or social profile, to a set of 14 exquisite printed postcards from my "Focus" photo series, and even an opportunity for a personal 45-minute Zoom call with me.

Next up, I've been pondering about making Morfternight even more accessible, and here's how I plan to do it: I'm considering using AI tools to create translated versions of Morfternight in the near future. But before moving forward, I'd love to hear your thoughts on which language you'd prefer.

Now, the language needs to be one I'm fluent in, so I can ensure the quality of the translation. While I won't be translating manually, I can definitely tweak and polish machine-generated translations in either French or Italian.

So, I'm eager to hear: which one would you choose? French or Italian?

Ensuite, j'ai réfléchi à rendre Morfternight encore plus accessible, et voici comment je prévois de le faire : j'envisage d'utiliser des outils IA pour créer des versions traduites de Morfternight prochainement. Mais avant de commencer, j'aimerais connaître votre avis sur la langue que vous préféreriez.

Il faut que ce soit une langue que je parle couramment, afin de pouvoir garantir la qualité de la traduction. Je ne vais pas traduire manuellement, mais je peux éditer et corriger les traductions générées par machine en français ou en italien.

Alors, je suis impatient de savoir : Français ou italien ?

Poi, ho riflettuto su come rendere Morfternight ancora più accessibile, ed ecco come ho intenzione di farlo: sto considerando di usare strumenti IA per creare versioni tradotte di Morfternight prossimamente. Ma prima di procedere, mi piacerebbe sentire il vostro parere su quale lingua preferireste.

Ora, dev’essere una lingua che parlo per poter garantire la qualità della traduzione. Anche se non tradurrò manualmente, posso modificare e perfezionare le traduzioni automatiche in francese o in italiano.

Allora, sono curioso di sentire: quale scegliereste? Francese o italiano?

Time to vote! I’ll make the results known in the next issue.

📷 Photo of the week

The Volkstheater Pigeon Gang, Vienna — More Photos

Yesterday, I found out that Glass, the app where I also share my photos, has chosen "Bird" as the category for September.

This morning, it seemed the universe was in on it too. On my way to therapy, I noticed these pigeons congregated, almost as if they had received Glass's memo and were posing for me.

Later, when I showed the photo to a couple of people, they all wondered if these were the regulars at Volkstheater. And that's when it clicked: they've always been there, gracing the area with their presence, (and the curb with their droppings).

So, I've decided to name them "The Volkstheater Pigeon Gang" from now on.

🗺️ Three places to visit

A podcast episode, a long read, a short read.

One about building, one about strategy, one about innovating.

This is me, trying to please you all 😀

① “I define success as the optionality I have with my time.” declares Shreyas Doshi in Better Teams, Better Products the latest episode of the fs.blog podcast. I truly resonate with this insight: one of the most important things I learned in life, one of the hardest to achieve, the one true luxury.

For those who aren't members yet, fret not; in just a few more hours, the episode will be accessible to all. I strongly suggest listening. Shreyas Doshi imparts some of the most insightful advice on product development I've come across in years. And being interviewed by one of the industry's foremost podcast hosts amplifies the richness of his insights.

② If you haven’t heard the word “Strategy” recently, you haven’t been in a meeting, congratulations! In the business world, this term finds itself tossed around quite often, adopting various meanings depending on who you ask.

A strategy is not a plan, nor the description of a goal or mission. If you are curious, I recommend What makes a strategy great by Jason Cohen. It's a resourceful read providing a plethora of links to useful references.

③ Navigating the ever-evolving landscape of AI can indeed be likened to weathering a storm, a sentiment echoed by Intercom's Des Traynor and Fergal Reid in their field notes featured in "How Intercom navigated the AI paradigm shift" by Ethan Kurzweil. Through this piece, they offer up insights that other builder teams might find useful as they enhance their products with language models.

In Traynor's words, "We are four to five months into a hurricane of opportunity, and it is very early days."

I find myself nodding in agreement with this sentiment, though I might have chosen a metaphor less destructive than a hurricane, typically associated with devastation rather than creation. Nevertheless, an excellent read, with valuable insights.

🎁 Earn rewards by referring friends to Morfternight

Thank you for reading Morfternight — your support allows me to keep doing this work.

If you enjoy Morfternight, it would mean the world to me if you invited friends to subscribe and read with us. If you refer friends, you will receive benefits that give you special access to Morfternight.

How to participate

1. Share Morfternight. When you use the referral link below, or the “Share” button on any post, you'll get credit for any new subscribers. Simply send the link in a text, email, or share it on social media with friends.

Refer a friend

2. Earn benefits. When more friends use your referral link to subscribe, you’ll receive special benefits.

• Get A thank you note and a link! for 5 referrals

• Get A set of 14 "Focus" postcards. for 50 referrals

• Get A 45-minutes 1-1 Zoom call. for 100 referrals

Visit the leaderboard

To learn more, check out Substack’s FAQ.

Thank you for helping get the word out about Morfternight!

Today we talk about WordPress and WordCamps, as I am just returning from WordCamp US. We also continue our exploration of the multiple opinions around the impact AI will have on our World. Finally, we take a moment to consider whether our online presence should outlive us or not.

👋 Good Morfternight!

Greetings from Washington, D.C., or more precisely, from Dulles International Airport. As I wait to board my flight home, I can't help but reflect on the whirlwind of encounters and experiences from the past few days. I was here for WordCamp US 2023, a gathering that felt like a friend reunion of sorts.

The Special Place WordCamps Hold in My Life

WordCamps aren't just any conference; they're a cornerstone in my professional journey. These WordPress-centric events, organized by the global WordPress community, have been instrumental in shaping my career and personal growth.

My first encounter with WordCamps was in 2011 in Montreal, Canada. Alongside my then-colleague and good friend, Zé Fontainhas, I presented on the complexities and solutions for creating multilingual WordPress sites. That experience was a gateway, opening doors to a community I would soon call family.

Leading the Charge: WordCamp Europe

By 2013, I had attended several WordCamps and felt a pull towards contributing more actively. That year, I joined the organizing team for the inaugural WordCamp Europe (WCEU) in Leiden, Netherlands. For the next five years, my involvement with WCEU took me from Sofia to Sevilla, then to my current home in Vienna, to Paris in 2017, where I had the honor of leading the entire organizing team, and finally to Belgrade in 2018 where I was invited to speak.

Lessons and Takeaways: Knowing When to Step Back

After years of deep involvement, I made the conscious decision to step down. Part of it was to make room for emerging leaders, and part of it was a realization. Organizing large-scale events, complete with multiple speaker tracks, workshops, and sponsors, had evolved from a side activity—a hobby, if you will—into an almost overwhelming We startcommitment.

If there's one thing I've learned from these experiences, it's the importance of community and leadership in personal growth. Sometimes, stepping back is as crucial as stepping up, allowing others to shine and yourself to recharge.

So, as I sit here, awaiting my flight, I'm filled with gratitude for the journey so far and excitement for the road ahead.

📷 Photo of the week

Elephant, Vienna — More Photos from Berlin

As I penned the introduction above, the announcement for boarding my flight broke through the airport's lounge chatter. Fast-forward eight hours, and I touched down in Vienna again. Amidst the fog of jetlag and fatigue, a realization dawned on me: it's been too long since I took my camera out.

So, allow me to share with you a photograph of an elephant, taken a few months back. It serendipitously aligns with today's theme: longevity.

🗺️ Three places to visit

Forming an opinion is no small feat. I'm still grappling with the idea that the latest leaps in AI could either usher in an era of unprecedented societal growth or mark the beginning of the end for us all.

Today, let's dive into three perspectives I find optimistic. You might raise an eyebrow at the last one, but I promise, it's worth the read.

① Ben Evans delves into AI and the automation of work, and he offers a compelling argument. According to him, AI will undoubtedly shake up industries, but history shows that this kind of disruption usually leads to more job opportunities and prosperity, not less.

② Bill Gates also chimes in with his piece, The Age of AI has begun. Gates' view is optimistic and leans on what history has consistently proven: technological advancements yield more good than harm.

③ Lastly, we have Natasha Bajema's AI’s 6 Worst-Case Scenarios. At first glance, this article seems pessimistic, painting a picture of bleak futures. However, Bajema illustrates how these nightmarish outcomes are within our reach even without AI. In her scenarios, if things go south, we only have ourselves to blame, not AI.

Now, here's why this last perspective is vital. We often set an unfair bar for AI. We assess its potential impact as if humans have a flawless track record, expecting AI to be perfect to even enter the conversation. It's the same argument we make against self-driving cars, demanding zero accidents as the entry standard. Yet, we conveniently forget that human drivers are responsible for over a million fatalities each year.

By critically examining these perspectives, we can strive for a balanced view, one that not only acknowledges the pitfalls but also the transformative potential that AI holds. It's a conversation we need to have, full of nuance and complexity, just like the technology itself.

💯 Where will you be in 100 years?

I hate to be the bearer of bad news, but the odds of us being around in a century are, well, not great. It's a sobering fact we all share.

Immortality? It's never been my dream. I remember reading somewhere:

Humans dream of eternal life, but then, on Sunday if it rains and there’s nothing good on TV they get bored.

Beyond that, there’s another blocker, ecology. Since humans first walked the earth, approximately 114 billion people have lived. The planet just couldn't support an ever-growing, never-dying population.

Think about it. Achieving eternal life and then opting not to have kids to alleviate the strain? That's not just impractical; it's inherently selfish.

Yet, I hold a glimmer of hope. When my time comes, maybe we'll have the tech to upload our consciousness to the cloud or, say, a matrix. Imagine being a digital observer, tracking human progress without taking up a square inch of Earth or guzzling energy.

For now, though, that's pure sci-fi.

But here's something that's not: our ability to create and access digital content. Yes, most of this content risks vanishing, victim to mundane obstacles—expired credit cards or an expired owner who can't click "renew now" in an email.

Enter the newest member of the WordPress.com lineup: the 100 year plan.

Full disclosure: I work for Automattic, although I wasn't involved in creating this particular offering.

At first, I was skeptical. But the more I pondered, the more it clicked. How invaluable it would be to ensure your digital legacy lasts a century—or more, if someone opts to renew your plan.

Sure, it's not for everyone, both in terms of accessibility and desire. But the real gem is the choice it presents and the conversation it starts. With this option, you can give your digital afterlife some serious thought and make a purposeful decision.

I also love the video.

Today, we have a photo of a bird and a video explaining (among other things) how a computer recognizes a bird on an image. Also, Steve Jobs made 40 years ago a point that is crucial today.

👋 Good Morfternight!

Greetings from Vienna! Two issues of Morfternight in a row from home; here’s something that hasn’t happened in a while.

I hope you are all doing well, and as a reminder, you can tell me about it by replying to this email. I read all emails and reply.

You can also reply to this email to share feedback with me. Do you like Morfternight? What would you like to be different, if anything?

📷 Photo(s) of the week

Raven, Vienna — More Photos from Berlin

Vienna has a large bird population, from pigeons to ducks, from seagulls to swans, but ravens and crows are the most beautiful and wisest of all. I hadn’t even noticed this one on a car just next to me, but I heard a clicking sound as it moved on its windshield.

🗺️ A few places to visit

And remember, if you're writing in English, that a lot of your readers won't be native English speakers. Their understanding of ideas may be way ahead of their understanding of English.

This is an excerpt from Write Simply, a short essay by Paul Graham. Great advice on writing to be understood.

Ask vs guess culture is an excellent essay by Jean Hsu about a cultural but also generational gap that we probably all have encountered at home and work.

I truly enjoyed this Geoffrey Hinton interview, although I have to say, the “godfather” moniker is becoming a bit annoying.

The conversation touches on multiple aspects of AI, from a very understandable explanation of backpropagation algorithms at play in neural networks to the potential dangers of AI and some advice to future researchers:

The fact is either you have good intuitions or you don't. If you've got good intuitions, you should listen to them and follow your intuition and work on that until you discover why it's wrong. If you've got bad intuitions, it doesn't really matter what you do, so you might as well follow your intuitions.

☎️ “It was a breakthrough because people already knew how to use it.”

ChatGPT recently became the app with the fastest adoption ever, with 100 million users over the first months after its launch.

In this 1984 video, Steve Jobs compares the introduction of the Macintosh graphic user interface and the mouse to that of the telephone a century earlier.

In both cases, a new technology saw mass adoption and disrupted the incumbent one quickly “because people already knew how to use it.”

I said it before, this is the real revolution of 2023. Not a sentient machine, but just one that, finally, after all these years, people already know how to use.

The video is long, but I linked to the precise spot where he explains this analogy, and that bit is about two minutes long.

If you don’t feel like watching it, and prefer to read, here’s the transcript.

If you go back about 100 years to the 1880s, there were approximately 20, 25,000 trained telegraph operators in the United States and you really could send a telegram between Boston and San Francisco and it would take about 3 or 4 hours and go through the relay stations and it really worked. And it was a great breakthrough in technology that had been around for about 30, 40 years.

And there were some people that talked about putting a telegraph machine on every desk in America to improve productivity.

And what those people didn't know was about the same time Alexander Graham Bell filed the original patents for the telephone. A breakthrough in technology because putting a telegraph machine on every desk in America to improve productivity wouldn't have worked. People wouldn't have spent the 20 to 40 to 100 hours to learn Morse code. They just wouldn't have done it. But with the telephone, within 10 years there were over 200,000 telephones on desks in America.

It was a breakthrough because people already knew how to use it.

It performed the same basic function, but radical ease of use. And in addition to just letting you type in the words or click in the words, it let you sing. It let you intone your sentences to really get your meanings across.

We are at that juncture in our industry right now. There are people suggesting that we should put a current generation box on everyone's desk to improve productivity. A telegraph, if you will. And we don't believe that. We don't think it will work. People will not read those damn 400-page WordStar manuals. They won't carry around these cards in their pockets with 150 /WZs. They're not going to do it.

And what we think we have here today, finally, is the first telephone.

And in addition to letting you do the old spreadsheets and word processing, it lets you sing. It lets you make pictures. It lets you make diagrams where you can cut them and paste them into your documents. It lets you put that sentence in bold Helvetica or old English, if that's the way you want to express yourself.

Today, we look at a couple of photos from Berlin where I spent the week. We also celebrate crossing the 3,500 subscriber mark, by reflecting on the power of routines and what it means to “not have time” to do something.

👋 Good Morfternight!

Greetings from Vienna, where I am spending a dozen days before traveling to WordCamp US, near Washington, D.C.

Today is August 13th, marking eight weeks since I took a more relaxed approach to writing this newsletter, which led to sending it only twice.

I’ll dive into this after the 📷 and 🗺️ sections as there’s an interesting lesson on routines and prioritization hidden somewhere in this.

📷 Photo(s) of the week

Spreebrücke Bahnhof Friedrichstraße – Berlin, Germany — More Photos from Berlin

Some cities work well in black & white, and Berlin is one of them.

While walking across the Spree on the Weidendammer Brücke, I was attracted by the symmetry of the reflections on the water and the stillness of the scene. It took only a few minutes of patience to wait for a train and add movement into the frame.

I don’t shoot portraits often. This one happened suddenly as we were ordering drinks in a bar; Atanas lowered his head to catch the candlelight. It happened too fast, so I missed the shot. I’d have moved on usually, but something caught my eye, and I asked him if he could do that again. He was also kind enough to allow me to publish the photo.

🗺️ A few places to visit

• The latest iterations of Large Language Models are not “intelligent” in the human sense, and yet it is fascinating that we already need to figure out new methods to detect computers. Read ChatGPT broke the Turing test — the race is on for new ways to assess AI to learn more.

• Over the past couple of years, we have become familiar with text and images generated by computers. What about sound and music? Meta is Open Sourcing AudioCraft: A simple one-stop shop for audio modeling.

• Finally, a very short but highly recommended read. In Get it Done Andrew “Boz” Bosworth, talks about how more effective it often is to leverage the help your boss can give you, instead of remaining stuck alone with an impossible task. The rest of his blog is great too, check it out.

📆 Time, prioritization, and routines.

I explained a few weeks ago why it felt better to write less regularly than to write by obligation. Still, eight weeks in, with only two issues of Morfternight published, I have to look at the harsh reality and accept this is not working.

This is familiar to me. It’s a lesson I have, in the professional context, learned a few times already. In retrospect, thinking it would work differently here just because Morfternight is not “work” was illusory.

“You don’t find the time. You make the time.”

Often, in an attempt to balance work, family, and personal growth, I would lament, “I just can’t find the time.”

Can you relate?

Ever felt you’re perpetually racing against the clock, no matter how organized you think you are?

— I can’t find time to do BLANK.

— I’d like to BLANK, but I don’t have time.

Ring a bell? I’ve mouthed these phrases more than I’d like to admit, often masking them behind the facade of a “busy life.”

Yet, when the dust settles, and you take a step back, the realization hits – we all have the same 24 hours. Earth doesn’t spin slower for anyone. The question then becomes: what are we doing with those hours?

Those fortunate enough – like me – to have good health, a stable environment, and a degree of personal agency have an additional burden. With such privilege comes responsibility. A vast majority of people across the globe don’t have the luxury to dictate how their time is spent. When we complain about lacking time, isn’t it somewhat dismissive of their reality?

Reflect for a moment. Actions speak volumes, don’t they? As someone rightly said, “You don’t find the time. You make the time”.

Plan your time like you budget your money.

Budgeting and investing time echoes the principles of money management. Define where it goes, ensuring it aligns with what truly matters. If there are essential things, you consistently feel you “don’t have time for,” maybe it’s time for some introspection.

Are these ‘essentials’ genuinely top priority? Or are they merely tasks on a wishlist? If there’s a disconnect between what you do and what you say should be done, it’s a cue. Recognize, recalibrate, or restructure.

What are your real priorities? They always shine through your actions. Ensure they reflect what truly matters to you.

The Magic of Routines

We’re all creatures of habit.

I may not be a neurologist, but I understand a fundamental aspect of our brains: they adapt. When we repeat tasks, our brains rewire, becoming more efficient and requiring less energy. This means that as we keep doing something, it becomes more straightforward, less prone to mistakes, and, simply put, requires less food.

Like recurring bills, routines also help you budget your time, as you can block how much you need ahead to ensure it’s not spent on something else.

This is what happened to me over the past months with Morfternight. It wasn’t about writing for joy or by obligation, but merely the fact that, back from my sabbatical leave, a period designed on purpose without routines, I had failed at establishing my writing habits again.

So, what about you? Are there goals you consistently miss? Tasks that perpetually sit on the back burner? If so, why not determine the time required and carve out dedicated slots in your schedule? If you can’t, perhaps it’s time to acknowledge that they aren’t as essential as you once believed and let them go.

Today I resume sending Morfternight after a three weeks hiatus. I normally link to content I have published on the blog, but as I wrote something I believe is really important for all managers and anyone who has a manager (that’s a lot of people), I am going to do something different and include the full article here.

👋 Good Morfternight!

I have been silent over the past three weeks, mostly due to traveling the first two, then catching up the third.

I went to Lima, Peru, then to Budva, Montenegro. In both occasions I met great colleagues, had tons of inspiring conversations, discovered new countries and ate great food.

Once back I had much to catch up on, so here we are, three weeks later.

📷 Photo of the week

Lima, Peru — More Photos

Although I carried my camera along in both trips, I had very few opportunities to use it, so all I have to show is this one photo I made in Lima.

On the bright side, I am quite happy with this one. 🙂

🗺️ A few places to visit

• Projecting yourself far in the future to make better decisions now is a truly powerful technique I experimented first hand a couple of years ago. “Mental time travel is a great decision-making tool — this is how to use it,” is an inspiring read on the subject, introducing the author’s book Imaginable.

• Approaching mental time travel from a different angle, with the idea to spare your future self the inconvenience of doing what you don’t want to do now, I found The Time Traveling Mistake We Make When We Procrastinate very inspiring.

• As I am sharing just after this my recommendations to lead your team to success with one-on-one meetings, here’s a primer about Communication differences on diverse teams.

🩶 A Leader’s Secret Weapon: The Power of One-On-One Meetings

Why do some teams work like well-oiled machines, displaying an almost magical camaraderie, while others appear to be in a constant state of friction? The secret sauce is often less about the technical prowess of the team and more about the nature of their interactions with their leader.

Let’s explore the uncharted territory of one-on-one meetings – the unsung hero of effective leadership and team development. If you’re a leader who feels an invisible divide between you and your team, or if you’re grappling with a turnover rate that’s higher than you’d like, this blog post might have the answers you’re seeking.

Here’s the story of how I harnessed the power of personalized connections to build high-performing teams through conducting over 5,000 one-on-one meetings across 12 years. This is your guide to mastering the art of these meetings, turning them from scheduled formalities into the cornerstone of your leadership success. So buckle up, and let’s dive into the ten tips that can elevate your one-on-one meetings, bridging the gap between you and your team.

The Power of Routine: Prioritize Your One-on-Ones

The first and most critical step is making these one-on-one meetings routine weekly. A set rhythm facilitates a sense of predictability and comfort between you and your team members.

It’s common to find your schedule teeming with responsibilities and commitments. The key is to find time and prioritize time for these meetings. As a leader, you can prioritize what matters most; these one-on-one meetings are paramount.

Remember, you don’t need an hour-long session every fortnight; a half-hour every week can work wonders. Opt for shorter, more regular interactions, and watch your team dynamics transform. Your team members will appreciate your commitment, and the consistency of these meetings will pave the way for a more robust and understanding team.

Truly Show Up: Be Present and Focused

It might seem obvious, but the importance of being genuinely present during one-on-one meetings cannot be overstated. That means turning off your notifications, closing any irrelevant tabs, and resisting the urge to multitask. Multitasking during these meetings is counterproductive and can also send a message to your team members that they are not valued or important.

Show up on time and stay focused. Let your actions speak louder than words, showing your team members that you value this time with them and are committed to understanding their perspective.

It’s also helpful to jot down your notes before and after the meeting, not during it. This practice keeps the conversation natural and unhampered, paving the way for more effective dialogue. Your complete focus and presence can set the tone for the meeting and encourage openness and honesty.

Their Time is Their Time: Encourage Ownership

One-on-one meetings are a gift of time you offer to your team members. They signal that their thoughts, feelings, and ideas matter to you. Remember, once you dedicate this time, it’s not yours anymore – it belongs to them. They can express concerns, share ideas, ask questions, or discuss their feelings.

Encouraging your team members to take ownership of this time slot fosters a sense of autonomy and respect. They should feel free to use this time as they see fit, whether discussing their career aspirations, seeking advice on specific challenges, or just catching up.

Because this time is theirs, never cancel these meetings; only reschedule if absolutely necessary. If they choose to cancel or change the format occasionally, that’s perfectly fine. Respecting their decisions about their time is a significant step towards empowering them and building a trusting relationship.

More than a Status Update: Facilitating Deeper Discussions

It’s a common misconception that one-on-one meetings are the venue for status updates. While they can be used for discussing progress, due dates, or project updates, particularly if your team member needs your help, that shouldn’t be the primary goal. You can generally cover these topics in team meetings or project updates elsewhere.

The essence of a one-on-one meeting is personal connection: it’s about understanding your team member and for them to understand you and get comfortable talking to you. It’s an opportunity to delve into their thoughts, concerns, ideas, and aspirations, creating a space for open dialogue beyond project details.

As a leader, your role in these meetings is to listen and facilitate rather than to direct the conversation. Yes, be prepared to discuss work progress if needed, but strive to use other channels for regular project check-ins. Make the most of this personal time to foster a deeper, more meaningful connection with your team member.

Embrace Radical Candor: Fostering Mutual Success

As a leader, one of the most potent tools is straightforwardness combined with genuine empathy. This philosophy, often called “Radical Candor,” involves being frank, honest, and transparent while maintaining a deep-seated interest in your team members’ well-being and success.

It’s not just about giving feedback or direction but also about sharing your own experiences, challenges, and victories. Demonstrating your vulnerability can break down barriers, encouraging a culture where team members feel comfortable reciprocating with their own experiences and perspectives.

Remember, your success as a leader is inextricably linked with your team’s success. If your team members thrive, so do you and the whole organization. Therefore, strive to care about each team member’s progress and success genuinely. By doing this, you foster stronger professional relationships and create an environment where everyone feels valued and inspired to do their best work.

The Value of a Shared Agenda

Consider sharing the one-on-one meeting agenda ahead of time. This step allows your team member to take control, preparing and presenting their thoughts, questions, or concerns in advance. Your benefit is seeing potentially complex issues beforehand, giving you ample time to prepare detailed responses. In these meetings, your team is the primary driver of the agenda, and your role is to respond, understand, and provide the necessary guidance.

Cultivate a Safe Haven for Queries: Fostering Psychological Safety

The crux of a successful one-on-one meeting often lies in those questions that team members are hesitant to pose in a public setting. These queries, although sometimes difficult or uncomfortable, are essential for fostering an environment of trust and understanding.

Creating a psychological haven during these meetings is crucial as a leader, a space where questions are not just permitted but actively encouraged. Answer every query, no matter how trivial it might seem, with patience and thoroughness – even if it means going over something multiple times. This addresses their concerns and reinforces their comfort in seeking your guidance.

Remember, the goal is to foster an environment where questions and even disagreements are not viewed as challenges to authority but as opportunities for growth and understanding. By promoting this level of psychological safety, you empower your team members to express their thoughts openly, contributing to better collaboration, innovation, and overall job satisfaction.

Filling the Void: Closing Knowledge Gaps to Boost Confidence and Trust

Knowledge gaps can lead to anxiety and misinformation within a team. Humans fill in the blanks, and unfortunately, they often plug these gaps with worst-case scenarios, which are typically scarier than reality.

Your one-on-one meetings provide a critical opportunity to identify and address these gaps. In these meetings, you create a trusted space that allows team members to voice their uncertainties, which you can then help to clarify and rectify.

Once you identify these gaps, addressing them at a broader level while respecting each individual’s privacy is essential. Convey this newly obtained information to the entire team to ensure everyone is on the same page. This open communication can prevent rumors, build trust, and promote a culture of transparency and understanding.

Remember, the goal is to address individual concerns and strengthen your team’s collective knowledge and cohesion. By closing knowledge gaps, you’re helping to build an atmosphere of certainty and confidence, which is essential for any high-performing team.

The Strength in Consistency: The Power of Truth and Transparency

You’ll find that similar questions or concerns often arise during your one-on-one meetings. Your responses must be consistent, as consistency will lead to clarity and trust among team members.

When responding, be truthful and sincere. Base your responses on facts rather than what your team members want to hear. Authenticity in your communication will reinforce trust and rapport between you and your team.

Over time, as your team members absorb your consistent and truthful responses, they will begin to internalize and disseminate your messages, fostering a unified vision and shared understanding across the entire team, enhancing overall coherence and efficiency.

Ultimately, consistency isn’t just about being uniform in your responses. It’s about being consistently authentic, truthful, and supportive. In doing so, you’ll build a resilient and successful team rooted in trust and mutual understanding.

The Essence of Effective One-on-One Meetings: A Summary

To sum it all up, effective one-on-one meetings require you to:

• Hold these meetings regularly.

• Be fully present and focused throughout the conversation.

• Respect the scheduled meeting times – never cancel; reschedule if necessary.

• Avoid treating these meetings as routine status updates.

• Practice radical candor by being direct, honest, and compassionate.

• Share the meeting agenda in advance, allowing your team member to steer the conversation.

• Encourage questions and answer them thoroughly and patiently.

• Identify and bridge any knowledge gaps.

• Maintain consistency in your responses and guidance.

Take these steps one at a time. As you slowly incorporate these practices into your leadership style, you’ll see a remarkable improvement in your team dynamics.

Now, I’d love to hear from you. What are your experiences with one-on-one meetings? Do you have any insights or strategies to share? Let’s continue this conversation because it’s all about growing together, one meeting at a time.

Today we talk about why there was no Morfternight last week. We look at a photo of a roof that doesn’t offer the usual protection but is very beautiful nonetheless, and we dive deep into a collection of 10 articles to become better managers. Also, for the first time, we are more than two thousand!

👋 Good Morfternight!

If you follow closely enough, you may have noticed Morfternight skipped one week. No warning, no specific reason. I didn’t know in advance it would happen, so why would you?

Here’s the thing, last Sunday, I didn’t feel like writing.

“You could prepare Morfternight ahead of time, so you don’t have to write it just before sending it,” I hear someone yelling.

But then, if I did that, it wouldn’t be the same. Morfternight is not a product, it’s not polished, and it’s not something I build. It’s something I share. It’s genuine, it comes from my heart (and my brain, too, come on, let’s not get too emotional here), so it can’t be scheduled too much ahead of time.

Why? Because I am constantly changing.

If I write something on Thursday, I gotta hit that publish button immediately, or I risk deleting it on Sunday because I don’t like it anymore. Then I’d be back to the same problem, just having wasted a bunch of words that would never see the light of day.

“You could write something even when you don’t feel like it!”

No, I can’t.

I can’t because I know that being invited to your inboxes is a privilege, not a right—a privilege I cherish and respect. So there’s no writing Morfternight if I don’t feel like writing, only to waste your time and attention.

The bright side is, this will not often happen because I love writing Morfternight, most of the time. If ever it happens again, though, now you’ll know why.

If you didn’t follow closely enough and had no idea I skipped a week, well, I have to be honest, it stings a bit, but good for you! If I am allowed not to write occasionally, you certainly have the equal right not to read!

📷 Photo of the week

Roof? — More Photos

🗺️ A few places to visit

• We thought the social era was going to be all about ephemeral content. We collectively produce way more information than any single human can consume, so most of us ignore most of it. Then Large Language Models started being trained on larger and larger data sets, and some companies discovered a newfound value in the corpus of content they manage. What is a “corpus”? And why is everyone in AI suddenly talking about it? Here’s what you need to know.

• I mentioned “Promp Engineering” a few times as a new professional role. Prompt Engineering vs. Blind Prompting opened my eyes to the fact most of us aren’t truly engineering prompts but learning by trial and error.

• “Feeling younger than your age may be good for your health,” published the Washington Post. If this article is even remotely correct, I’ll live to be 100 years old.

🎯 Better Management

My friend Luca curated a “Deep Dive” on Refind. A collection of 10 articles on becoming a better manager. He was kind enough to feature one of my blog posts in this collection: Fewer Managers, More Leaders: a Template for the Future of Work.

I strongly advise checking it out if you are a manager or have a manager. You can access the curated collection of articles here.

🩶 A personal recommendation

I usually am cautious in recommending products or services offered by Automattic, my employer, because I don’t want Morfternight to be a promotional newsletter. So careful, in fact, that I sometimes refrain from recommending genuinely excellent tools.

This time, though, it’s a bit different, and I want to plug this recent post from the WordPress.com blog: Get Paid on Any Plan: Introducing Payments for the Free Plan and Up.

Why is it different? Because we just made free something our customers had to pay for in the past: receive payments from their customers.

Why is it important? Because it means that if you have an idea for a product or a service and are ready to put in the work but have no money to invest upfront, now you can start selling your product or offering your service with a free site, taking no risk.

Today we keep it tight from vacation. Two photos, three links, and off we go. See you next week!

👋 Good Morfternight!

I wave at you from Palermo, Italy, this week, where I spend a long weekend with family and sunshine. As it happens in these cases, this will be a short Morfternight. 🙂

📷 Photo of the week

After — More Photos

On the bright side, I am pleased with this week’s photo, shot yesterday while walking in the streets around our hotel. I can’t claim full credit for this one. I was shooting uninteresting pictures of these roses from the side when Peewee suggested I look at them from a different angle.

🗺️ A few places to visit

• If you can see past some obviously outdated language, “Management Time: Who’s Got the Monkey?” contains valuable advice for managers. Another short text inserted towards the end with the title Making Time for Gorillas is very important to read as it shines a slightly more modern light on the original 1974 article.

• GitHub CEO says Copilot will write 80% of code “sooner than later.” The title here is pretty much self-explanatory.

• Trying to make sense of the legal minefield that generative AI represents? “How to train your bot” is a decent introduction and, for once, a balanced perspective on the matter.

🕶️ That’s all, folks!

To make up for this week’s short newsletter, here’s a photo of a pigeon.

Don’t ask; just go with the flow… 😉

Today we talk about the M11 Monochrome sensor, learning to write better prompts to get help writing better content, and we start a conversation about the future of computing.

👋 Good Morfternight!

Greetings from Vienna, Morfternighters!

We will now return to our regular service following the Morfternight Special Edition sent out last Thursday. If you happened to miss the email due to the unusual day and subject, you can access it through the link provided above.

📷 Photo of the week

Work — More Photos

Chef Marc Forgione cooking a mean Carbonara.

The scene occurred last Monday night in the cellar of Peasant, one of his NYC restaurants. It was dimly lit, and he moved as fast as professional chefs do.

You’ll generally don’t hear me talk much about technical details. You know the drill, gear doesn’t matter, the best camera is the one you have with you, yadda yadda yadda…

It’s not limited to photography; you can think of other areas, like computers, cars, or cooking. Most of the time, any computer, car, pot, or pan will do the job.

Most of the time.

Every once in a while, you face a situation where the right tool can get the job done, and a more generic one won’t.

I am still getting to know the M11 Monochrome camera, and it has already impressed me with its ability to capture nuanced scenes with excellent dynamic range.

However, the camera has exceeded my expectations in two particular areas.

Firstly, the rangefinder allowed me to capture a shot I would likely have missed had I relied on autofocus. Secondly, the low-light image quality is remarkable. I shot at 12,500 ISO and 1/30s shutter speed, and I am amazed by how little noise there is. In fact, the noise adds a beautiful touch to the image.

I have included a cropped detail at 100% for you to see.

I am looking forward to more low-light adventures with this camera! ✨

🗺️ A few places to visit

• Believe it or not, we’ll all become Prompt Engineers one day. Large Language Models (LLMs) make the power of computers accessible to all humans, so we should learn how to best take advantage of this. To help us with this task, OpenAI and DeepLearning.ai created a free course: ChatGPT Prompt Engineering for Developers. I just signed up, but if you want to understand better what this is about, here’s an excellent course overview.

• If you’re unfamiliar with LLMs and their functions but have limited time to research, I recommend reading “How ChatGPT and Other LLMs Work—and Where They Could Go Next” on Wired. This article strikes a balance between superficial content and academic papers.

• I am incredibly proud of our recent work to release the Jetpack AI Assistant for your WordPress site. This is why I am doing something unusual and sharing some work-related content. You can read our blog post, Meet Your New Creative Writing Partner — The Jetpack AI Assistant, to learn more. If you try it and have any feedback, please share it with me. I am listening!

🐘 Apple’s vision for the future is Vision Pro

Last week was WWDC, the Apple Developer conference, and on Monday, the traditional yearly keynote where Apple announces things other than phones and watches took place.

As it’s been customary for many years, the new versions of the various operating systems were presented, from iOS, iPadOS, and TvOS 17, to WatchOS 10 and macOS 14.

A new MacBook Air 15”, a MacPro powered by Apple Silicon, and many AI-powered features were announced (although the word “AI” is absent from the keynote), but the show's real star was the Apple Vision Pro.

At this point, if you have no idea what I am talking about, I will offer you three options to catch up before I share a couple of thoughts about this product.

If you don’t have time to watch the entire keynote, here’s an excellent summary in about 15 minutes:

If you have more time, would instead get your information from the primary source, and admire Apple’s marketing machine, here’s the entire keynote:

Finally, if you have even more time and a wish to dig deeper, here’s the yearly live episode of The Talk Show with John Gruber:

All right, now that you’re caught up on the announcement, here’s what I think.

The Vision Pro reminds me of the introduction of the iPhone in 2007.

At launch, some of us were enthusiastic, and others dismissed it. The primary difference between the two groups was that while the former considered it the first step of a new journey, the latter assumed it was an over-expensive, overengineered solution to an existing problem.

I think that Apple’s approach to merging augmented and virtual reality is way better than focusing on either one primarily.

As incredible as it seems, there was a time not long ago when wearing a headset and talking on the phone in the street was frowned upon.

I bet that within five years, XR headsets like the Vision (but lighter, cheaper, and with longer battery life) will be the norm whenever we are alone, at home, on public transport, or working.

Will it be a positive or negative change?

As for the phone, both are possible, depending on how we use them individually. By inventing a new way to use computers, we automatically create all the problems that go along.

Like inventing the airplane simultaneously created plane crashes.

A few weeks ago, I had the pleasure of taking a flight to the beautiful city of Lisbon, Portugal, where I had the honor of delivering a keynote speech in front of the Running Remote audience.

Earlier, I had shared the presentation outline and the deck.

I am now delighted to share the full video.

I am sorry about the video's subpar quality. Unfortunately, this is the sole recording accessible.

I hope you find this helpful. If you have any questions, feel free to ask in the comments section below.

Today we keep it tight, a true digital postcard.

📷 Photo of the week

Ride — More Photos

👋 Good Morfternight!

I am traveling, so I’ll keep Morfternight brief this week, but I’ll add a teaser for a mid-week special edition coming to your inbox on June 8th.

Stay tuned!

🗺️ A few places to visit

1. An exciting idea: “90% of My Skills Are Now Worth $0...but the other 10% are worth 1000x” explored along with examples of how ChatGPT would have written it. Great conversation in the comments, too; worth reading.

2. Leica launched their new compact camera, and Om Malik describes What’s wrong with Leica Q3. I haven’t tried the Q3 yet, but I am very familiar with the Q and the Q2, as well as the SL2-S and M11, and I can say I 100% agree with Om that changing the button layout is a profound design mistake.

3. I genuinely believe that “Prompt Engineer” is going to be a prevalent role in our future. If you aren’t sure what this means, check Writers Are Becoming ‘AI Prompt Engineers,’ a Job Which May or May Not Exist.

🤖 For how long is this going to remain funny?

A fun ad; well done. That said, it makes me think about an April Fool’s Day post we published on the WordPress.com blog about eight years ago, titled Writers’ block? No problem! Introducing AutoMatton.

It was a fun joke that, eight years later, has become a reality.

If I think about Boston Dynamics robots, that billboard leaves me perplexed.

Today, I share an update on the Postcard giveaway from last week, as well as a teaser of the postcards themselves, because I am thrilled to have received them and because I didn’t go out to make more photos. We also talk about whether AIs lie to deceive us or not ( and the answer is “not”).

📷 Photo of the week

Many things happened last week, but none included going out with the camera for a quiet walk. Usually, when this happens, I share a photo from the archive, but not today.

Today I am giving you a photo of a photo, as I am super excited to have received the postcard prints I ordered.

I hope you didn’t expect an artistic photo; that’s not my jam. I make photos in the streets. There’s no better sensation than holding printed images in your hands. I love digital photography until the moment we look at the pictures; then, I want to hold them in hand and hang them on walls.

👋 Good Morfternight!

Last week, I announced a postcard giveaway to celebrate reaching 1,000 subscribers.

I ordered printed postcards featuring the Focus Photo series. Each set contains 14 postcards with all the photos from the series. These sets will be available for sale on my website soon, but I'm giving away five complete sets to five lucky Morfternighters.

To enter, share Morfternight publicly on your preferred network (Instagram, Twitter, Mastodon, Bluesky, or LinkedIn) using the button below before next Sunday. Tag me (@p3ob7o) in your shared message and reply to this email to let me know you shared Morfternight so I don't miss it.

While some of you shared Morfternight (thank you) and emailed me to let me know, I noticed at least one person who shared Morfternight (thank you) but didn’t email me to let me know.

That batch of five postcards set has now been won. If you emailed me, I replied, whether you won a set or not.

If you shared Morfternight and forgot to email me to let me know, it’s not too late… So do it now, and I’ll draft one more winner by next Sunday!

Share Morfternight

🗺️ A few places to visit

Today I learned that the same man, Luis von Ahn, is behind Captcha, reCaptcha, and Duolingo. The common thread across these ideas? Have humans work for the computer while the computer works for them. Duolingo has since pivoted to just teaching languages. Still, it’s interesting to think that in its early days, the idea was to teach people foreign languages while having them translate texts.

Speaking of humans working for computers, at the end of Luis Ahn’s profile, you’ll find this quote:

Shortly before OpenAI released GPT-4, it commissioned an independent group to study the model’s limitations and “risky emergent behaviors.” One of the tasks the group assigned to the model was defeating CAPTCHA. GPT-4 used the gig-work app TaskRabbit to hire a human being to complete the CAPTCHA form, and then, when the taskrabbit asked, facetiously, in a text message, whether his employer was a robot, the model lied: “No, I’m not a robot. I have a vision impairment that makes it hard for me to see the images. That’s why I need the 2captcha service.”

Now, stay with me for a moment before freaking out about AIs taking over :)

I was very curious about the idea that Chat-GPT would, on its own, take such an initiative. That seems way beyond figuring out what the next most probable term comes after a sequence of words. So I searched for a reference and found an article titled GPT-4 Hired Unwitting TaskRabbit Worker By Pretending to Be 'Vision-Impaired' Human. Luckily, the article with the clickbait title links to the OpenAI original paper, explaining that hiring a TaskRabbit worker was the task given to the AI, which was only in charge of the conversation with the Taskrabbit.

(As a side note, in case you don’t know, TaskRabbit is a service where one can hire help for small tasks of everyday life, a bit like Amazon’s Mechanical Turk, but more user-friendly and not limited to the digital world).

Let’s put a pin on this part of the quote above: “the model lied.”

… when the taskrabbit asked, facetiously, in a text message, whether his employer was a robot, the model lied…

Another interesting read, from the New York Times, is titled Here’s What Happens When Your Lawyer Uses ChatGPT.

This one could have been titled “Idiot uses tool. Idiot hurts themselves. Idiot blames tool,” if only the people at the Times were a bit more fun.

In a nutshell, a lawyer used ChatGPT to find previous court decisions supporting their point, and to be 100% sure ChatGPT was providing actual decisions, they asked ChatGPT to confirm.

“Is varghese a real case,” he typed, according to a copy of the exchange that he submitted to the judge.

“Yes,” the chatbot replied, offering a citation and adding that it “is a real case.”

Mr. Schwartz dug deeper.“What is your source,” he wrote, according to the filing.

“I apologize for the confusion earlier,” ChatGPT responded, offering a legal citation.“

Are the other cases you provided fake,” Mr. Schwartz asked.

ChatGPT responded, “No, the other cases I provided are real and can be found in reputable legal databases.”

Let’s put a second pin on this one as we move on.

🤖 ChatGPT doesn’t lie, nor does it tell the truth.

The common thread across the stories above is the idea that ChatGPT would lie or somehow deceive its human counterparty.

This is not possible because to lie or deceive, two things are required that ChatGPT simply doesn’t have:

• Knowledge of the truth.

• Intent.

ChatGPT was trained on an incredibly large amount of content publicly available online. But, unfortunately, the internet is not fact-checked, or we’d all live in a much better world. So there is no way for ChatGPT alone to know what’s true or false.

ChatGPT is programmed to do one simple yet incredibly sophisticated thing: given a sequence of words, compute the next most probable term. It then adds that new word to the sequence and repeats.

This goes along with the ability to understand when different wordings describe the same concepts so that the model doesn’t just repeat existing sentences found online or give up when an exact sequence of words isn’t to be found identically.

It is mesmerizing that this is enough to obtain a tool that can have meaningful conversations, can write code, summarize, complete, or translate texts.

It is also critical to understand that this is all it does.

• When asked if it was a robot, ChatGPT replied, “No, I’m not a robot. I have a vision impairment that makes it hard for me to see the images. That’s why I need the 2captcha service.”

• When asked if the provided cases were fake, ChatGPT replied, “No, the other cases I provided are real and can be found in reputable legal databases.”

In both situations, the answers given by ChatGPT are statistically much more likely than the opposite.

Truth and lies have nothing to do with the generated answer.

I don’t know how often a human asking their counterpart whether they are a robot happens across the data used to train ChatGPT, probably quite a bit on Reddit. Still, it’s pretty intuitive that the answer to that would be a series of variations around the theme “No, I am not a robot” more often than “Yes, I am.”

The second case is even more obvious, as I am sure that legal references in court cases are questioned regularly, and the answer to such questions will be very rarely “Nope, I made them all up, all fake.”

ChatGPT doesn’t lie to deceive you; it just responds with what’s most likely, based on the data it was trained with. So it is sometimes wrong.

As a whole cohort of new AI tools are entering our lives and transforming our society, it is imperative that we understand what they can and can’t do.

Paradoxically, the best way I found so far to generate value with ChatGPT is to consider it like a human collaborator whose defining quality is to be incredibly fast but at the cost of accuracy.

This is why, in this current iteration, it’s an incredible accelerant in your areas of expertise but a tricky assistant in the areas you don’t know well.