Steve Jobs once tried to buy Dropbox for reportedly a 9 figure sum, but was turned down by the two founders. Jobs met with Drew Houston and his partner, Arash Ferdowsi, for a meeting at Cupertino, and Houston said that no, that they were not interested in selling.
Jobs smiled warmly as he told them he was going after their market. “ He said we were a feature, not a product,” says Houston.
That same year, Steve Jobs unvelied iCloud at that year’s keynote, as the way to solve the problem of file syncing across all your Apple devices.
And with that swift blow, killed any chance of Dropbox ever becoming an Apple product.
LLMS ARE A FEATURE, NOT A PRODUCT
The way I see it, LLMs will never be a product, they will just integrate into our daily lives, and we are already there, we are already using them as a feature. In VScode and other developer tools, for example, you call an LLM to help with your code. Same thing with ChatGPT, you use them as part of your iPhone, as part of Siri.
LOCAL LLMS ARE THE FUTURE
What’s stopping Apple and Google from making their own local LLMs and letting people freely use them? They have the resources, people and know-how to do it. This of course will crater the current LLM market and send OpenAI, Anthropic, and the rest of the pack into the ground.
At the moment I’m writing this, I’m using the free version of OpenAI’s ChatGPT on my phone. How much do you think it costs me to use it? Zero, nada, zilch.
I signed up for a free account on OpenRouter, which provides access to a bunch of LLMs, including the latest OpenAI models released a couple days ago; it’s free, and has some daily limits on how much do you use the models, but if you want to use it more often, you can add US$10 in credits to your account to increase daily limits on free models.
I started using those, seamlessly switching between them, testing Google, OpenAI, DeepSeek, Mistral and others, with no problem at all, and also switching between paid and free models.
LOCAL FIRST
On iOS, apps like Enclave and OpenCat, among others, let you download and use local LLMs on your phone. So you can use a LLM for the unbeatable price of free. [^1]
STEAL UNDERPANTS → ??????? → PROFIT
I don’t know about you, but the current LLM market looks like that famous South Park underpants scene, where a US$300 billion company like OpenAI doesn’t have point #2 defined.
When I switched between models I realized that I was using them as a feature, not as a product.
Basically Fitt’s Law is one of the most important principles of human-computer interaction, which states that bigger targets are easier to hit than smaller ones.
And Apple doesn’t give a damn about this because they just threw it out the damn window.
Before iOS 26, you had the bottom of the screen, and the navigation bar was convenient not because of being on the bottom, but because these targets were big enough for you to hit them. Now, they are tiny compared to previous iOS versions.
HOW MUCH WORSE DO YOU WANT IT? YES
Now there’s a huge gap between the bottom of your phone and the bottom navigation bar, now called Toolbar. Which creates a lot of problems for elderly people, people with bad eyesight, people with disabilities, people with bad motor skills, and in general, for anyone who doesn’t have perfect hand-eye coordination.
I tried disabling transparency across the whole iOS, but that didn’t work.
MACOS AND WINDOWS
If you have a Mac or Windows computer, you will notice that at the bottom of the screen is the dock or taskbar respectively, this area is easy to get to. Why? Because of Fitt’s Law!
You can throw your mouse at this target and hit it, no matter how strong or weak your hand is, you will hit it 100% of the time when you reach the bottom of the screen. The elements in this area are buttons with an infinite bottom.
MACOS
On macOS, at the top of the screen is the Mac menu bar, which, also thanks to Fitt’s Law, is easy to hit. Throw your mouse and you will hit with a 100% success rate. The elements in this area are buttons with an infinite top.
Apple has been lately become less boring but not less annoying. For the wrong reasons, they not only decided to update the UI, the “look & feel” of the OS, but they also decided to change the way things work, and honestly I hate it.
On top of the visual updates, AKA liquid ass, they added NEW interactions like the expanding-search feature.
I was ok with the search input field at the bottom in Safari, I think it is ergonomic and easy to reach, but now that they updated other apps like Music, App Store, and everywhere else, it’s just too much of things not working as expected.
**ACROSS ALL APPS**
SEARCH
Search is crap.
If you tap the search icon, it doesn’t activate, it just changes to a search field, you need to tap it *again* to activate it.
BUTTONS
Buttons changing size and shape.
Why?
Why do the buttons need to move from where they were?
SAFARI
I swear to God they moved some of the elements a couple pixels in the Safari input field, as I find myself accidentally tapping the microphone button a couple times a day.
APPLE MUSIC
If you use the Music app, don’t tap the Checkmark, or it will delete your albums.
That someone at Apple approved using that icon for this action is dumb.
**WHAT TO DO INSTEAD: SLOW AND STEADY**
If you are going to make such dramatic changes, start small, get the user accustomed to it, and then start making changes.
Doing two things at the same time, updating the UI and changing the way things work is a recipe for disaster.
On the debate about Perplexity using stealth techniques to crawl websites:
Who would you rather believe? A company with decades of experience, or a newcomer trying to catch up?
The stock has been flat the whole year, and only moved this past week thanks to the US$100 billion investment announcement by the president of the United States.
Chart: Year to date performance of Apple (AAPL) stock
THE MAGNIFICENT SEVEN
Now let’s zoom out and compare it against the rest of the market.
Chart: Year to date performance of Apple (AAPL) stock, compared to other tech stocks
As we can see, Apple has underperformed if we compare it to Oracle (ORCL), Broadcom (AVGO), Nvidia (NVDA), Meta Platforms (META), Amazon (AMZN), Alphabet (GOOGL), and even Microsoft (MSFT) has performed better than Apple.
Microsoft has performed better, let that sink in for a second. The company run by Satya Nadella, which shamelessly, without any shame or regard for the user, has put ADS(!) in Windows 11, is doing better than the company focused on privacy and user experience.
And also Meta outperformed it, which we know is run by Mark Zuckerberg and the less we say about them the better.
If you bought shares by any of these companies you’d be up.
Figma, the collaborative design app, debuted on July 31st under the ticker FIG.
As of today, it (very) quickly went up, and then lost its gains.
Its use of AI to create designs is a big deal. A huge deal. It automates and streamlines a lot of the work that user experience designers do, facilitating collaboration and making it much much faster to work by using text prompts. I’ve tried it and honestly it’s pretty cool, allowing to create even full-fledged prototypes. It even allows you to use your own design system library so the AI can generate designs that look like your brand and style.
Neat little AI tool that helps you compose better messages across all your apps. You start writing something and Cotypist will suggest phrases, sentences, and short paragraphs based on your writing style.
This year Apple (AAPL) stock value hasn’t been that great for shareholders which bought at the start of the year, barely scratching the price level on February… and — this is the important part —, not regaining ever since.
THE COLD HARD TRUTH
As of August 1st, Apple stock is down 17% compared to the start of the year (year-to-date).
If you bought US$100 worth of Apple stock on January, today you’d have US$83.
Three weeks in, it had already fallen 11% before recovering, before continuing to fall at the end of the month. February painted a slightly better picture, gaining again the lost ground in the previous month, with signs of losing traction again at the end of the month.
In March, Apple stock suffered a great drop, falling even below the levels reached at the end of January
THE FALL BELOW US$200
April was brutal for the stock market, stocks started to lose ground in march, and with the tariff statements from the US government the whole stock market saw red for several weeks, and Apple fell below the US$200 mental mark to US$172 dollars per share.
Here is where the stock paints a very different picture. While Microsoft, NVDA and others have been releasing AI features, they have also seen their stock value rise, while Apple’s AI Intelligence has been delayed, removed from launch, or not even talked about, and Wall Street has noticed this.
The stock has been floating since around US$200, traction is basically nowhere to be seen, and it is not even climbing like the other companies in the so called “Magnificent Seven” have done, reaching new highs. But not Apple.
ARE INVESTORS WORRIED ABOUT APPLE’S (LACK OF) AI STRATEGY?
Is it a winter for Apple, or is it something bigger? Services is a huge cash cow, but aside from that, Apple as of lately has been boring — now, don’t get me wrong, this is fine for me, I love their products and the M chip is great —, but as a company, its public performance hasn’t been that great. Apple is not the only one facing regulatory issues (but those have been happening all the time), specially in Europe.
If you want to see a company doing some great public stunts just look at Tesla.
And let’s not talk about the UI changes coming up for iOS 26, their latest beta is quite something, change for change’s sake.
This week, Aeropostale (AEO) ▲ exploded because of some girl wearing jeans in a TV ad.
Sydney Sweeney lately has been everywhere, acting, selling expensive soap (at over $1k a pop), to modeling, and now pumping stocks. I must disclose I’m one of the few people who enjoyed seeing Madame Web, and now that I think about it, I have watched not one, but three of her recent movies, including her rom com and the nun one, which honestly sucked.
It has been revealed that OpenAI is the customer that has printing free money to Oracle.
When the news came out that they were using Orale’s infrastructure, the Oracle (ORCL) ▲ stock exploded, and Larry Ellison became even more rich, maybe the richest man on the whole universe, at least temporarily.
How much did OpenAI is spending on Oracle’s infrastructure PER YEAR? Thirty BILLION dollars! That’s a lot of zeroes even for these companies. I’m pretty sure that Oracle’s infrastructure is a mix of the usual: Curl (open source), Linux (open source) and some app like Syncthing (also open source).
In my next life I’ll wrap open source in a box and sell it to devs.
Tesla (TSLA) ▼ and Chipotle (CMG) are down ▼
Chipotle fell down almost -15% these past couple days, and the market hasn’t been kind to Elon’s company. Nor him.
After being caught red handed deleting tweets, and before that getting denied electric vehile subsidies, the market is responding by slowly dropping Tesla’s stock. It’s been rough and while the stock has been trading way above its real value, the market still thinks it should trade as high as possible.
Opendoor (OPEN) ▼ is the latest victim of Reddit’s r/wallstreetbets. The subreddit had seen lots of recent activity promoting the stock, people showing screenshots of Robin Hood with their buy orders… until if reached a critical mass large enough to move the needle.
Historically, Opendoor had been trading barely above $0.50 (that’s fifty cents), and these past couple days climbed and climbed until it reached a crazy high of $4.97… before falling down to $2, and now it’s been staying there for the last couple days.
If you bought it at $0.50, I hope you are happy. If you bought it at $2 dollars, let me know if you will apply diamond hands and hold on to it.
Newegg (NEGG) ▲ is now being pumped. Avoid buying it since it is falling down, just go to Yahoo Financeand zoom out to see how it has been trading. It’s down -96.58% from its all time high of $2,155 in April 2010. After precipitously falling the following months, in August 2010 it was trading at $556.80. In 2014 it went up again, and never recovered after that.
So, buy if you are willing to wait 10 years for it to recover.
In other news, coming soon to Google Graveyard®: Opal, one of the latest attempts of Google trying to kill any good will it still has left. I stopped caring about any of their projects a long time ago, and avoid them like multi-level marketing scammers. They recently started, and stopped selling domains, and also killed their shortened URL service, which might have cost them peanuts compared to their size. They should just avoid creating any new products, and not because I will never use them, but because plenty of people will do use them, only to be afflicted and experience hardship when they put all their eggs in this ADHD company.
Chart: Year to date performance of Apple (AAPL) stock, compared to other tech stocks