I started designing my own clothes. The insight was that I spend 80% money on suits that I wear 2 times a year, and the rest was low quality clothing I actually wore.
I flipped it, and made suits and pants that I could wear everyday.
The fast fashion stores were crap quality, my body is not a template size and I care about fabric and comfort.
The process was to learn how to sketch, to determine fabrics, colors and fit. I made pants that stay comfortable even after I eat food, I made suits that I can wear casually.
I don’t stitch myself, for that I worked with multiple workshops, until I found one that works for me.
Took me about 3 years to reach a point where all my wardrobe is designed by and for me.
There were multiple side effects on my confidence, my life, and the opportunities coming my way.
Exactly correct. In reading some highly regarded works two things occurred to me, first that the author had captured into words some fundamental aspect of the human condition. Second was that it's easier to think about something presented as a story than it is when it is presented as an alternative to how you currently think.
If you tell someone there position on some topic is wrong, they will argue with you. If you tell someone a story where the character takes the same position they have and then through experience and personal growth comes to understand how it is wrong. They can come away realizing that they might have it wrong. Great trick when it works.
some time ago i discovered Wardley maps [1] (about company's "landscape" and strategy there), and one thing that stick with me was this:
there are 4 levels/stages of development there - genesis, custom-built, product, commodity. With three transitions, made by different kind of people - Pioneers, Settlers, Town-planners. And the last one, mass-production, is about "ruthless removal of deviation".
i guess these "diminishing returns" in keeping long-time same-thing (democracy?) have something in common with the removal of deviations/variety..
> As general literacy declines, one of the consequences seems to be that words with multiple meanings or variations become compressed into a single vague meaning.
I don’t know for who the literacy is declining. But the people who yap about democracy are the educated intelligentsia.[1] So it’s the educated that narrow and widen definitions.
> Not so long ago there was a common distinction between capital-D Democracy the political system and small-d democracy the process of power or knowledge being diffused through the masses. You don’t see this idea expressed much anymore, and even the expression “making something more democratic” almost always implies a reference to the political system, not the the second sense.
That’s not true. When people talk about “democratizing X” where X is distant from the political process they mean people participation and power. Like “democratizing social media” could mean user-controlled and driven social media as opposed to everything being controlled a by corporation or something.
> This distinction is useful, because one of the biggest trends of the technological age is the capital-D version supplanting and erasing the small-d version. Almost all of the institutional “defenders of democracy” have essentially no interest in small-d democratization processes, because they are themselves in the driver’s seat in the political democratic system.
Pretty much true.
> This leads them to ignore or disregard the issues of everyday people, which leads to populism, which is of course the biggest global political trend of the last decade.
Pretty much. That there are a group of people who can ignore or disregard the issues of everyday people means (by its very premise) that there is no democracy.[2]
People who then might have tolerated that then have enough and turn to the correct political theory: elites rule the commoners. Again the premise proves the theory correct.
> This is kind of a shame for tech in particular, because for the most part technology has been a democratizing (small-d sense) force throughout history. Phones, computers, cars, on and on: all examples of expensive exclusive technologies that became democratized and accessible to everyone.
This is a bit of a vulgar[3] conception of democratization. Democracy is about power, not access to X. If a car indirectly gives you political power by being able to travel and organize then it indirectly has that effect. But if it only gives you the opportunity to commute one hour each way to your workplace then it has got nothing to do with democratization.
And if your phone just makes you addicted to social media—as the technologists on this board so smugly like to point out—then it doesn’t give you power.
> And yet that same force doesn’t seem to have been applied to the political system itself.
Democracy is about governing your own life in harmony with the rest of the people under that democracy. The political system is a big deal there. But there are other spheres of life the workplace.[4]
[1] Parochial way of referring to relatively wealthy people who set the intellectual agenda
[2] Although people can call it “liberal democracy” if they want since the Liberal in that is much more important to the system (according to its defenders) than the Democracy part
Ideas in general are difficult to express and people struggle with conveying them separately from their private ideas, personal experiences, and personal reasons for believing what they believe.
If you want to be a good interlocutor, you have to deeply absorb what the other person is thinking and sometimes even help them develop their understanding with the hope that others can do the same for you. We are all toddlers at times.
Writing is quite different as a discipline and in its effects when it's for self-discovery, self-analysis, messaging others, etc.
It's like the vast differences between relaxing alone at home, caring for your partner, hosting guests, going to work, taking kids on vacation...
What's helpful in either case is the ritual of entering that domain: you pull yourself together for work, dissolve on the couch, and put on a happy listening face for the kids. The more consistent you are in your rituals, the more quickly and deeply you enter and exit that domain.
That's how to escape mental defaults: have a huge variety of them, widely-spaced, so you can introspect their difference and play their respective lights against their shadows.
My own conclusion about burnout is that it fundamentally comes down to who controls the agenda, and how much you invest in that agenda. I've been burned out. Many years ago, early in my career. My cure: I was in the lucky position that it was a good time to spend a year and half going back to school to knock off a graduate degree that simultaneously moved my career forward, gave me a total change of scenery, and gave me some break time between leaving the job and starting classes, time that I devoted 100% to hobbies and home improvement projects. And of course, an easy-to-tell story when reentering the job market.
So on the topic of agenda... if what you are working on is your own agenda, you don't burn out. You might change the agenda by redefining goals, but in the end, you are sailing your own ship. Not only do you not burn out, it is curative. It is when you absorb someone else's agenda and make it your own to an unhealthy extent that you burn out. Always be computing that dot product between your employer's agenda vector and your own agenda vector. Don't over-invest beyond that dot-product.
By de-incentivizing M&A, and checking larger competitors to VC darlings by hanging the Damcoles sword of antitrust.
A decade ago Marc Andressen was lobbying Obama to work on this [0][1]
In Andressen's and much of his peer's eyes, most mid-late stage startups should be IPOing sooner than they actually are. And to a certain extent he isn't wrong.
Personally, I don't buy Andressen's argument - there is a reason we added added checks and balances in the IPO process.
> Could you explain this
To go public (just like any other fundraising stage), early stage ownership stakes tend to be diluted in order to attract later investors.
IPOs are a fundraising technique like any other, but the benefits tends to bias towards funds that target late stage or roadshow investors at the expense of early investors.
In the eyes of Andressen and his peers the IPO process needs to be simplified in order to make it easier for mid-stage startups to go public AND the incentive structures need to be changed so early stage investors (read VCs like A16Z) get outsized benefit.
For most funds, this really doesn't matter, but for the mega funds like A16Z, YC, Founders Fund, etc this is a make-or-break policy as most of their portfolio are mid-late stage startups that have been pushing off IPOs because they are too small for the current market, and taking acquisitions at what a number of early stage investors view as a suboptimal price - doesn't matter to the founder because they have cash, but it does to large early stage investors.
A direct listing or SPAC would be the ideal "IPO" method envisioned, but that has been cracked down on as well (and rightfully so tbh)
I dunno, I think it could with the right evolution in the interface. Imagine an interactive story app that you listen to on your commute, where voice commands back to it are the only interface (eg so it’s safe to interact with while driving).
Maybe that’s just a subset of the more general “AI companion” opportunity, but I expect you could get some really interesting experiences by calibrating the balance between the manually curated/composed parts of it and the parts that get a bit more painted-in by the LLM.
Am thinking especially of stories with conflicting timelines, unreliable narrators, etc, where you’d maybe be revisiting the same events from multiple perspectives to piece together what actually happened.
I don’t believe that’s true. Lynch uses characters as abstractions — they represent a concept, which when you understand them add up to a ‘statement’:
* Twin Peaks is an meta commentary about the lack of balance on the small screen (our desire for sex and violence). Interestingly this ended up being a meta-meta-commentary when the network forced them to reveal the killer of Laura Palmer prematurely. Laura Palmer is balance, Dale Cooper is the audience, Bob is our desire for sex & violence, …
* Mulholland Drive is about the casting couch and the destruction of the Hollywood dream for women. Rita is the casting couch, the cowboy is Hollywood, …
* Lost Highway appears to be a comment about plagiarism (namely other directors plagiarising David Lynch). But it also has a similar theme to Twin Peaks as a meta commentary about film — the shots of the road are meant to look like film, the vehicles are meant to represent movies, …
etc.
He always talks about being “true to the idea” — so all abstractions and all surreal elements must be true to the underlying concept. It’s up to us to work out what the underlying theme is (that links all of the abstractions together).
>> 100 trillion synapses (very roughly analogous to transistors?)
Not even remotely comparable
* its unlikely synapses are binary. Candidly they probably serve more than one purpose.
* transistor count is a bad proxy for other reasons. A pipeline to do floats are not going to be useful for fetch from memory. "Where" the density lies is important.
* Power: On this front transistors are a joke.
* The brain is clockless, and analog... frequency is an interesting metric
Binary systems are going to be bad at simulating complex processes. LLM's are a simulation of intelligence, like forecasting is a simulation of weather. Lorenz shows us why simulation of weather has limits, there isnt some magical math that will change those rules for ML to make the leap to "AGI"
It stands for "energy recovery ventilator" - in a nutshell it just blows fresh air into your house and exhausts stale air, but uses the heat in the exhaust air to warm up the incoming air so it's more comfortable and energy efficient compared to an open window with a fan in it.
Before you go too far that route (a multi-thousand dollar install) I recommend getting an indoor air quality meter to get a sense of what might be the problem. I recommend the Airthings View Plus, which is about $300 but is comprehensive and reliable and since (in my case) I made a multi-thousand dollar decision based on it, it was worth the investment.
In my house it was obvious there's a massive CO2 buildup, CO2 itself isn't as big a deal itself but an indicator of stale air (ie, it's not being vented out by itself) and somewhat elevated radon. The ERV isn't the "prescribed" solution for radon but since it ends up venting it out it's fine.
One last thing - for me, the connection between energy and indoor air quality was obvious in retrospect. I could wake up early and be miserable, but once I stepped outside to take my son to school I'd feel better. Or vice versa - I could be full of energy running around all day, then come home and immediately get drowsy.
It only took about 15 years to realize indoor air was the cause.
If you are a fan of the fractals but feel intimidated by neural networks, the networks used here are actually pretty simple and not so difficult to understand if you are familiar with matrix multiplication. To generate a dataset, he samples random vectors (say of size 8) as inputs, and for each vector a target output, which is a single number. The network consists of an 8x8 matrix and an 8x1 matrix, also randomly initialized.
To generate an output from an input vector, you just multiply by your 8x8 matrix (getting a new size 8 vector), apply the tanh function to each element (look up a plot of tanh - it just squeezes its inputs to be between -1 and 1), and then multiply by the 8x1 matrix, getting a single value as an output. The elements of the two matrices are the 'weights' of the neural network, and they are updated to push the output we got towards the target.
When we update our weights, we have to decide on a step size - do we make just a little tiny nudge in the right direction, or take a giant step? The plots are showing what happens if we choose different step sizes for the two matrices ("input layer learning rate" is how big of a step we take for the 8x8 matrix, and "output layer learning rate" for the 8x1 matrix).
If your steps are too big, you run into a problem. Imagine trying to find the bottom of a parabola by taking steps in the direction of downward slope - if you take a giant step, you'll pass right over the bottom and land on the opposite slope, maybe even higher than you started! This is the red region of the plots. If you take really really tiny steps, you'll be safe, but it'll take you a long time to reach the bottom. This is the dark blue section. Another way you can take a long time is to take big steps that jump from one slope to the other, but just barely small enough to end up a little lower each time (this is why there's a dark blue stripe near the boundary). The light green region is where you take goldilocks steps - big enough to find the bottom quickly, but small enough to not jump over it.
I probably have a whole personal blog of my own on the topic, dear as it is to my soul.
My two main pieces of advice: The bar is very very low, and share your burden quickly.
99 times out of 100 you are way overestimating the value of what you're delivering and people's expectations for it, and underestimating the value of time i.e. shipping quickly.
I've turned in so many things I'm not happy with and gotten a "this is great" that now I frequently just send over pseudocode, whiteboard sketches, and bullet point design docs to just get going on the feedback loop. Nobody has ever said "this is so bad we can't use any of it."
I also realized I do much better finishing other people's work than starting my own .. and so does almost everyone else. Bringing other people in overcomes "the boredom paradox" of a looming deadline - working with other people has its own challenges, but it is definitely not boring!
One specific thing I did that helped a few years ago at my precious company was I told my team, wrote in my email signature, ran a small study group, etc. On grit, procrastination, and "growth mindset" and just made a very intentional effort to tell people how I struggled with this problem.
So many people shared the problem! It really gave us a nice community and helped us (and management) recognize some of these issues, lesrn some new techniques, and get better at coaching, setting expectations, and ultimately managing the work.
So maybe last piece of advice is be open if you have these issues.
> Data Scientist's job is to launder management's intuition using quantitative methods
Ouch. This is savage, but sadly correct in many cases.
HOWEVER, to play devil's advocate here, I've also seen corporate data scientists overstate the conclusions / generalizability of their analysis. I've also seen data scientists fall prey to believing that their analysis proves would should be done, rather than what is likely to happen.
The role of an executive or decision maker is to apply a normative lens to problems. The role of the data scientist / economist / whatever is to reduce the uncertainty that an action will have the desired effect.
Just one minor tweak of what plays next can totally rip off independent artists on the platform. We have reached an era where algorithms aren't transparent, so artists like me are bewildered that for all the promoting we do on our own music, we rarely get any views and listens unless we literally spend thousands of dollars on advertising to break the visibility barrier...
For example, if I tweet a link to my own song (hosted on spotify) not only will Twitter potentially block people from seeing the link, their URL shortener may break the link to Spotify (Because the CEO doesn't want traffic leaving Twitter) and then even if the link goes to Spotify, they do a ton of things to siphon listeners that came for my music away from listening to my music, including NOT playing more of my music after the intended song plays. The net result is that hours of promotion as an artist only generates a few leads that often get ushered away from your content... It happens in many other ways for creators, artists, and even businesses without anyone being able to know that it's happening.
The future of being an independent entrepreneur is totally disrupted by social media as it slowly creates a stranglehold on the Internet. If we all don't start acknowledging this and calling out anti competitive practices and platform scams, we'll all be weeded out from being able to make our own living and we'll be forced to work for employers for minimum wages... The Future of the Internet looks grim from where I see it.
Journaling for me is an essential tool for emotional regulation and self-knowledge. This wasn't always the case because I learned in school to write for someone else. If I'm not writing for someone else then who am I writing for? I'm writing for my future self, but who is that? My future self is determined by the thoughts and experiences I am having now, so if I can understand through writing where I am now in the world then I can make better decisions about where I am going. In effect, writing is a way of programming myself. If I am to be captain of my own ship then I need to keep a log to better learn from past experiences. I've found writing is the best way to collaborate with other parts of myself as Huxley says
If you correct your mind, the rest of your life will fall into place.
- Lao Tzu
It doesn't need to be imagined, it needs to be written down
- Philip Glass
I like the feeling of words doing as they want to do and as they have to do
- Gertrude Stein
My fate cannot be mastered; it can only be collaborated with and thereby, to some extent, directed. Nor am I the captain of my soul; I am only it's noisiest passenger
- Aldous Huxley
Good composition is like a suspension bridge; each line adds strength and takes none away... Making lines run into each other is not composition. There must be motive for the connection. Get the art of controlling the observer – that is composition
- Robert Henri
Memory is the major element in cognition, in everything that we call the humanities. If you cannot remember, then you can't think and you can't imagine, and you can't write, and you can hardly read
- Harold Bloom
The characters in my novel are my own possibilities that were not realized
I agree and if anyone is curious I am part of a project to bring open source farming robots for regenerative agriculture to everyone. We’re working on building a monthly crowd funding following to sustainably fund the project while staying completely open source. We have a good prototype vehicle now and are working on a good solid kit design we can ship in the next year or two, and will work with our user community to design tools for soil prep, planting seeds and transplants, weeding, and harvesting all in a regenerative organic process. And anyone can start a business selling their own variant of the design. We believe this will drive down costs in the same way that 3D printers dropped in price by 100x when patents expired and an open community formed around the many new designs. More details in our latest update here:
For an interesting real-world case of ignoring IP laws, one might look at the history of Philips and its light bulbs:
- The Netherlands didn't have patent laws in the decades that Edison was introducing his lamps
- Philips bought up the machines of competitors that Edison drove out of business through his patent monopoly and started producing lamps en masse in the Netherlands
- more than a few researches of said former competitors joined Philips' R&D crew, presumably partially out of spite, partially because it was the only place where they could continue their research
- this combination of experts and not giving a damn about patents resulted in really quick innovation at Philips' factories, meaning their light bulbs soon were both better and cheaper than anyone else's (of course, this only lasted until the light bulb manufacturers formed a cartel to introduce manufactured obsolescence).
- eventually the Netherlands caved in to international pressure to introduce patent laws. By then Philips was so huge that they started abusing it for patent trolling of their own.
In 'Why we talk' (https://www.amazon.com/Why-We-Talk-Evolutionary-Evolution/dp...) Dessalles arguments that we have two modes of talking - the first one is visual and probabilistic, the second one is topological - that is concerned about zero one questions - like if something is inside or outside of something. The first one is about imagining situations and judging their salience (or improbability). The second is about argumenting - simplifying the information so that you can 'proove' stuff. The point is that visuals are too rich and there can be too many ways to imagine things - so it is hard to convince someone that a particular scene is wrong or right. When you simplify things to the topological level of question if something is inside something or outside - then you can use inference to arrive at some logical conclusions, with steps so obvious that it is hard to argument against them.
Programming is mostly about logic - that is why it is hard to visualise. I think this is the problem with 'visual programming languages' (like https://enso.org/ - which, disclosure, is in my investmet porfolio) - and also here. This is kind of surprising - because we naturally expect the visuals to be as rich as the logic. But we need to carefully think about what is complementary here.
> Without a model of ADHD thought processes my attempts at empathy were counter-productive.
This is why I often wonder it empathy isn't overemphasized as a tool in our collective toolbox. When teaching it, we often fail to highlight how essential a coherent model the subject's inner world is to the empath's success. Otherwise, the empath is merely fantasizing about their reactions to similar circumstances - a recipe that can lead to festering resentment.
My professional experience has taught me that proactive empathy can be more harmful to a work environment than a simple combination of humility and compassion.
A piece of that can mean realizing that a team member isn't fit for purpose and either removing or working around them.
It can also mean taking the arduous path of developing an understanding of your coworker's inner-world, as you ended up doing. Even then, I recommend modelling around capacities and thought patterns rather than emotional responses.
There's nothing more obnoxious than a coworker without a clue, or to-which you're indifferent trying to 'understand you'. Especially if they're tremendously off-base.
I disagree so strongly with this. There's a pervasive fallacy in the industry that things that aren't quantifiable, or are hard to quantify, can't be real or true. It's just wrong: they can be very true, and often such things are deeply true but whole organizations are ignoring them because they haven't found a convenient way to shoehorn them into their KPIs / OKR framework / whatever.
Yes, it is often the case that product managers have no concern about the code quality. No, the problem is not that N days aren't given for QA. If the product managers talk to the engineers, they will get a sense of the problem. If they work as colleagues they will be in this sort of discussion all the time. When a product is implemented, they'll keep talking, and they'll be aware that the major corners are being cut, and they'll understand that that puts it at risk. Then they can make better decisions to avoid this.
If they have that information and still make the wrong decisions, then they have perverse incentivizes or are behaving maliciously to the business.
And if they try to reduce the intuition and professional opinions of their engineering team to some numbers -- estimates, QA lead times, bug counts -- which are pitched over the wall, badly summarizing actual expertise -- then they will fail, badly, over time, because that is a nonsensical way to work.
Good decision-making and leadership, imo, don't need you to have found ways to quantify things, and your time is often better spent doing good work than waiting for things to be put into numbers. Quantification helps, sometimes.
"If it isn't measurable, how can you make the needed changes? " Doesn't that question sound absurd? You make the needed changes! What does measurability have to do with it?
This practice has similarities to Rudolf Steiner's Six Exercises for Basic Esoteric Development. I've been practicing at least a few of these, nearly every day for most of my adult life. I've found them to be a great help.
My 2 favorites are:
The Control of Will
Choose a simple action to perform each day at a time you select. It should be something you do not ordinarily do; it can even be a little odd. Then make it a duty to perform this action at that time each day.
Rudolf Steiner gives the example of watering a flower each day at a certain time. As you progress, additional tasks can be added at other times.
This exercise is as hard as it is simple and takes a very strong intention to complete. To start you might think of it as you think of a dentist's appointment - you do not want to be late. It can be helpful to mark your success or failure on the calendar each day. If you completely forget at the time, but remember later, do it then and try to do better the next day.
My other favorite isn't on this list, but is a related practice:
Draw the same plant or tree every day for 5 minutes. This offers incredible insight into the observation of small changes taking place as part of the change in seasons.
These days when I hear people being 'surprised' by negative feedback they weren't getting but 'everyone' knew about:
You should ask yourself why people stopped coming to you with information, and why someone else has it instead of you. Especially if it's your job to keep on top of that issue. You've alienated them in some way that means they're avoiding talking to you, from fear or contempt.
In the cases I've been involved with, I've come to realize that people don't like to look stupid, and some people accidentally or on purpose put you into that scenario a few times, and they just start avoiding you. You no longer get all the information, and the people they feel safe giving it to will include people who you already have a strained relationship with, and they may use this information as ammunition against you.
It's better that people ask you dumb questions and you teach them how to answer them or steer them toward better questions, rather than imply that's a dumb question. Most especially if any other people are going to observe the conversation. Will you get interrupted a lot? Yes. But you'll also know where all of the choke points are in the system - where everyone else is getting interrupted. If you're doing policy or architecture without knowing pain points, you're going to be the one who looks like an idiot. But nobody is gonna tell you that to your face. Or at least not until your goose is cooked.
>there are two kinds of authors: those who like to write and those who like to have written.
I like this a lot. It succinctly points to an thought that's been at the forefront of my mind lately.
It seems to align with Daniel Kahneman's work about our two versions of our "self": an experiencing self (the writer) and the remembering self (the one who has written). As you allude, sometimes those two may not align in terms of what brings them happiness or fulfillment.
It makes me think I should be working to make sure I'm the one who optimizes for the experiencing self rather than the remembering self. Besides the fact that we spend much more time "experiencing" life than "remembering" it, the latter seems too often correlated with things outside our direct control like status etc.
I've seen PG [1] say/write versions of this: "The Y Combinator founders who followed our advice succeeded. The ones who didn't, didn't."
The advice is so simple, it's hard for a lot of outsiders to believe it's worth anything. "Make something people want." "Talk to your users." "Do things that don't scale." "Keep typing and avoid dying." People hear about this and ask "You gave away 7% of your company for that?" No, you give away 7% of your company to join a network of people showing you what it really looks like to do that.
My company got into the Winter 2009 batch of YC, the same batch as Airbnb. They weren't around for many of the dinners; they spent a lot of their time away from the Bay Area doing exactly those things that PG advised, mostly in NYC, where many of their most active users were. They just did that stuff, over and over, for several years. Now they have one of the most successful companies out of Silicon Valley in the last 15 years. (I saw PG tweet a couple of years ago that he'd recently dinner with them, and Brian would still write down PG's suggestions in a notebook.)
During that batch, I was flailing about trying to find some magical trick to make our company work. I remember one office-hours session with PG, excitedly telling him some buzzword-filled story I'd dreamed up about how our company could be a brilliant success. "Just make a good website" he replied.
It took me a while to work out how the Airbnb guys were able to follow the advice so effectively whilst we and so many others got stuck in the weeds, but looking back now it's pretty obvious. They were just very comfortable in their own skin. They didn't have ego issues around needing to seem like geniuses, needing validation all the time, fearing rejection or embarrassment. "Talk to your users" was easy, as they were sociable, likeable people who put on cool parties and who were naturally able to make everyone in their company feel welcome and valued, and everything else emerged out of that.
I don’t know if this is a mental hack or not, but I found years ago that if I mentally sing the comments (that I know would upset me otherwise), it totally removes the emotional impact of other people’s negative writing.
When I was contemplating why this might be so effective, I was reminded that satire of old often involved singing to point out other peoples absurdity. When you think about how much the powerful fear humor and satire, there might be something there…
I run a free hand-curated concert calendar in NYC[0] with a friend. It's all operated manually and we still end up with 15-20 great shows every night. If this were automated or scraped, you'd have 250 events on your calendar per day.
Sure it might work if you just pick one venue, or a handful of artists. But it's not particularly helpful since you'll either be overwhelmed by spam (venue shows you're not interested in) or you'll inevitably be missing out on a lot of great events (since you're filtering on very specific parameters).
On our app, you can swipe on any show to add it to your phone's calendar instantly. 60 seconds daily or a few minutes each week and you can build your own subset calendar from our already curated calendar. It's honestly very efficient and low-tech.
I do subscribe to some remote calendars like my favorite sports team (NBA provides per-team calendar URLs). If you're in a smaller town with just a few venues, I bet you could do that too. For the privacy-conscious among us, you can always route them through Cloudflare Workers pretty easily.
I flipped it, and made suits and pants that I could wear everyday.
The fast fashion stores were crap quality, my body is not a template size and I care about fabric and comfort.
The process was to learn how to sketch, to determine fabrics, colors and fit. I made pants that stay comfortable even after I eat food, I made suits that I can wear casually.
I don’t stitch myself, for that I worked with multiple workshops, until I found one that works for me.
Took me about 3 years to reach a point where all my wardrobe is designed by and for me.
There were multiple side effects on my confidence, my life, and the opportunities coming my way.