Réduire les déchets seraient un moyen de régler le problème

What we are trying to do is reach theoretical saturation, the point where additional research doesn’t give us new insights.

The image pictures it well: https://swipefile.com/from-data-to-impact and creates a consistent pattern.

A exercice of thought:
Raw data points is random and inconclusive. Information makes them consistent. Knowledge (or Findings) describe emerging patterns in data but aren't actionable. Insights reveal opportunities in this knowledge. Wisdom emerges from insights. Impact is reached while taking action with these insights and wisdom.

Hindsights are reflections of past actions and outcomes.
Foresights are informed projections, insights with extrapolation.

Here’s what it then looks like in real life:
Data ↓
Six users were looking for ”Money transfer” in “Payments”, and 4 users discovered the feature in their personal dashboard.
Finding ↓
60% of users struggled to find the “Money transfer” feature on a dashboard, often confusing it with the “Payments” section.
Insight ↓
Navigation doesn’t match users’ mental models for money transfers, causing confusion and delays. We recommend renaming sections or reorganizing the dashboard to prioritize “Transfer Money”. It could make task completion more intuitive and efficient.
Hindsight ↓
After renaming the section to “Transfer Money” and moving it to the main dashboard, task success increased by 12%. User confusion dropped in follow-up tests. It proved to be an effective solution.
Foresight ↓
As our financial products become more complex, users will expect simpler task-oriented navigation (e.g., “Send Money”, “Pay Bills“) instead of categories like “Payments”. We should evolve the dashboard towards action-driven IA to meet user expectation.

Cela ne sent pas bon pour les futurs investissements déraisonnable.

Comme pour toute bulle d'investissement, seule les applications éprouvées continueront d'être financées.

One of the figure who coined the term Enshittification. He writes a lot

Des notes réfléchies desquelles j'ai créé certains shaare plus récent

The books were once criticized similarly to AI now.

Saying no to AI is a luxury too. You have to be able to.

10 topics to consider while creating a file format.

Je suis auto-entrepreneur. Je veux changer mon adresse. C’est une immondice sans nom.

suite aux retours: https://lehollandaisvolant.net/?mode=links&id=20250521213749

Thus, the general expectation is that AI implies, at the very least, software that consistently and reliably outperforms a human expert at any task in any given field it claims to be proficient in.

Les LLM ne sont pas assez efficace sur des tâches spécifiques et consomment trop de ressources. En ce moment, il y a aussi le retour d'autres types d'algorithmes. Ces algorithmes peuvent être associés mais n'ont pas besoin de LLM.

En revanche, les SLM et les réseaux de neurones vont continuer à être utilisé, puisqu'ils ont des utilités spécifiques.

There are cycle in web development. For example, Server Rendering -> SPA -> MPA -> Server Rendering

When JS tools are written in other programming languages, it increases the barrier to contribute in these tools. The contributors need to know a second language.

A rewrite in another language is already a rewrite, which means things will be better built.

Node performance is often slow, but it works without cache... Let's try export NODE_COMPILE_CACHE=1 first.

Moreover JS is easily debuggable.

I fundamentally disagree with you. If a piece of technology allows for such mistakes, then it is at fault, not people

It has to be accessible but the software limiting the user
"So far, we’ve been lucky. When technologies try to protect you from yourself too much, they tend to die out in a few years"

In order to avoid user mistakes, what should the email client do?

Email is powerful beyond any other medium precisely because it does not try to wrest control from its users.

As always: frontend focus provide qualitative resources.

They have a partnership with Tonkotsu this time :)

it’s amusing to consider how complicated modern software systems are that the developers themselves don’t know everything about them

Against the experience of SPAs

Some things you have to consider with SPAs:

What happens when users refresh the page?
What happens when users click the back button?
What happens when users click the back button twice?
What happens when users click the back button twice, the forward button once, and then the back button again?
What happens when users try to open a link in a new tab?
What happens when users users copy the link from the address bar and send it to a friend?
Where does the page focus go when it navigates?

But SPAs make sense in some cases though.