What actually pushed me to go was three things:

1. A lucky coincidence — my university had an exchange program; 2. My mother's death — the blow that year was too heavy, subconsciously I wanted to escape; 3. France's policies — public university fees were modest, no high requirements on the family's financial situation, and you could work part‑time.

I got on the plane with two or three hundred euros in my pocket.

Toulouse, two years: quiet — and my first real encounter with the "two faces of multiculture"

Toulouse, to me, was a very quiet city. Over those two years I crashed at classmates' places, at friends' places, I lived frugally — studied during the day, worked nights in a Chinese restaurant and a French pizzeria.

Quiet didn't mean no impact. The people in Toulouse came in all shapes, and they weren't as friendly as I had pictured — I did get insulted on the street, in different languages. That was the first time I directly felt it: "multiculture" has a shadow side, too.

Its shadow side goes like this: in a mono‑cultural country, you don't have to explain yourself every day — your skin, your accent, your surname, everything automatically falls into the "one of us" bucket. In a multicultural society, every label on you becomes a visible entry point; some people get curious about it, others get impatient about it. Both attitudes exist at the same time, and neither one ever goes away.

It also has a bright side. I met my old landlord, who later became my business partner — my benefactor in France. I treat him like an older brother, and we're still working together to this day. Years later I realized: the deepest gift of multiculture is meeting people you would never have crossed paths with in a single‑culture life.

Paris, two years: inside Master 2, I saw a "garden of joy"

After two years in Toulouse, I decided to move to Paris. The reason was simple: too quiet, missing some passion, some challenge. Whatever the cost, I wanted a bigger city.

Paris is never gentle.

The challenges: rent could eat close to half of my income, the cost of living jumped up a tier. Restaurant pay was sometimes worse than in the provinces, and the bosses were tougher. Every day: classes, homework, evening shift, back home, more studying. Exhausting.

The opportunities: but Paris is far more tolerant than any small French town. It's too international, too mixed — which dilutes the "strange looks." For a Chinese student, that mattered a lot.

The bigger gift was in the studies. In Paris I did one year each at two different universities:

• First year at Université d'Évry / Paris‑Saclay: Public Administration · International Cooperation and Solidarity (2013–2014)
• Second year at Université Paris‑Est Créteil (UPEC): M2 International Local Project Management (2014–2015)

The pure theoretical level at French universities is actually not weak — quite the opposite. What really surprised me was how finely the disciplines are branched — the catalog of available paths is far broader, and far more specific, than at Chinese universities. In China the directions usually stop at coarse buckets — "business / law / computer science" — whereas here I could drill down to niche tracks like NGO governance, municipal management, international cooperation and solidarity. That fine‑grained granularity gave me things you can't get in a Chinese setting, and it directly broadened the direction my career took later.

The second year especially. The professors really cared about the cohesion between students; the whole class felt like a big international family — perfect, almost idealized. I still think those were the two most beautiful, most precious years of my entire student life.

Why Master 2 was a garden of joy — I only figured this out afterwards: it wasn't an accident. The syllabus had built‑in icebreakers, cross‑cultural collaboration exercises; the grading system also pushed people toward mixed groups — if your group was all Chinese, you wouldn't get the top mark. In other words, "fusion" wasn't left to the students' goodwill; the system itself pushed it into existence. Across those two years what I saw wasn't a bunch of people who got along by accident — it was a designed multicultural friendliness. I only realized later: the essence of France's so‑called "garden" is that it knows multiculture doesn't happen on its own, so it pushes you at the level of the system.

In that class there was a classmate named Camille who would often spontaneously ask me whether I wanted to look at her notes. A small thing, and ten years later I still remember it. A few friends from that period later ended up working with me.

If I had to find a single word for those days — a garden of joy.

Ten years of building in France later: idealized school vs. real society

From graduation to now, I've been building businesses in France for exactly ten years. Looking back, the complexity and management difficulty at the level of real society are an order of magnitude harder than the idealized tolerance you find in school.

Multiculture at school has teachers maintaining it and a system pushing it; multiculture in real life is each person deciding, on their own judgment, how to treat the next person. Where is the gap? The gap is in friction — on the street, on the metro, at a government counter, and when you become an employer facing employees from different countries, you realize there's no "teacher" grading anyone anymore; everyone has to own their own biases and their own patience. At that point multiculture isn't a garden anymore — it's something you have to actively maintain.

In real life, multiculture brings plenty of problems. But I believe it: France cannot back away from the multicultural path.

History and civilization move forward; backward movement isn't allowed. I'm not saying it has never happened — the resurgence and religious regression in some Middle Eastern countries, in my view, is exactly that. But as a long‑term trend, multiculture won't reverse.

I often compare France with the US. American multiculture looks more like a "platter" — each color stays sharp and distinct, but the pieces aren't very keen to blend. French multiculture looks more like a "solution" — on the surface it's less colorful, but once you're dissolved in, everyone is soaking in the same set of secular republican values. Both paths have their cost, but both are much harder than "mono" — and both are closer to the future than mono is.

As an immigrant, I've taken my share of unfair treatment. But on a horizontal comparison, this kind of unfairness, across all developed countries, is still relatively fair. That's a judgment I didn't feel ready to make until ten years in.

On my two daughters, and the whole "100% integration" thing

I have two daughters, both in France.

First‑generation immigrants like us carry a double pressure: cultural shock, career, family. Saying we don't want to integrate at all — that's false. Saying we're 100% integrated — that's also false; for both internal and external reasons, we don't quite get there.

But for my children's generation, I'm full of hope.

What I can do as a parent is to not pass on the fatigue of an incomplete integration; to encourage them to embrace this multicultural future, and to grow into adults who are truly international.

I want them, when they grow up, to never feel out of place just because they are "the daughters of a Chinese man," and also to never forget where they come from just because they grew up in France. Real multiculture isn't either / or — it's holding both at the same time. Their generation has that chance — far more than mine did. What I can give them is to step out of their way, and to tell them about the pitfalls my generation walked into, so they don't have to start from zero.

---

In a way, this is my full answer to the two words "multiculture":

• Before France — a developed country.
• Toulouse — multiculture has a shadow side, and a bright side.
• Paris — the idealized multiculture is a garden pushed into existence by design.
• Ten years of building later — there's friction between the ideal and the real, but the direction is right.
• Looking at my children's future — I believe.

We'd actually tried Claude earlier than that, but at the time it only worked through code mode — you had to ask questions inside the Mac Terminal. The barrier and ongoing friction were both high enough that I didn't stick with it. About a month later the Claude web version came out, and that's when it really came back onto my desktop.

Codex is a developer, Claude is a boss

After using both for a while, the difference between Codex and Claude turned out to be huge:

• Codex feels more like a developer. You have to spell the requirement out clearly. It can take you from 0 to 90% of the work, and that part is genuinely strong — but the last 10% you have to grind through over and over, because it has a hard time catching what you actually meant.
• Claude has a "thinking layer." It feels more like the boss of a web‑dev shop, or a senior engineer — closer to how a person actually reasons. That positioning drops your communication cost a notch, and the throughput ends up much higher than Codex.

As a production tool, once the efficiency gap opens up, the slower one quietly gets shelved. I'm on the 200 € / month Claude plan, and there are still moments when it feels a little tight.

My workflow: Claude dispatches, Codex grinds

My setup right now is fixed:

• Claude breaks tasks down, plans, makes judgement calls, and writes anything that needs thought
• Codex handles repetitive mechanical work — batch edits, batch runs

The result: I burn less than 5% of my Codex quota, while Claude runs at close to 100% almost all the time. In other words, I've gone from playing around to heavy dependency.

Claude and ChatGPT are different animals

People tend to lump Claude and ChatGPT together. They're really not the same thing.

ChatGPT is a Q&A tool — you ask, it answers. What it gives you is ideas. Claude is a production tool — you give it a task, and it does the work — it hands you a result.

That's a difference in kind, not in degree.

More than 12 hours a day

Since Claude moved into my daily workflow, I use it from the moment I wake up until I go to sleep. Excluding sleep and meals, I'm on it more than 12 hours a day — heavy user, no qualifier.

The effect on my routine is real. I used to keep a "normal" workday; now everything I do orbits around Claude.

I'd bet more and more people end up like me: using Claude the way we scroll a phone, to the point where not using it feels uncomfortable. The dependency these agents create is only going to get stronger.

On the valuation

Plenty of people say Anthropic is overvalued, that it's part of the AI bubble.

From my own usage, it's not a bubble. The reason is simple:

• The more you use it, the harder it is to walk away
• And this kind of user is willing to keep paying for it

The two days with Fable

One more thing made me even more confident in Anthropic's upside.

They released a model called Fable. I only got to use it for two days before it was pulled, without even really getting a feel for it. But even in those two days, the level was clear — it was among the strongest models on the market at that moment.

Putting a model of that caliber out in the open and then quietly pulling it back tells me, in reverse: the real hand Anthropic is holding is probably much bigger than what's visible from the outside.

That's exactly why I say OpenAI is in a fragile spot. Not because ChatGPT collapsed overnight, but because Anthropic can drop Fable, walk it back, and do it that calmly — what else is sitting in the other hand?

At this pace, I think it's plausible that Anthropic ends up overtaking OpenAI. Because ChatGPT and Codex are already pushed aside in my own setup — a few more days like this and, if I still haven't reopened them, I'll cancel both subscriptions.

Abstract

After two essays on NGOs, social enterprise, and the social value of entrepreneurship, this essay turns to the lawsuit between Elon Musk, OpenAI, Sam Altman, and related parties. The dispute is not only a Silicon Valley feud. It raises a deeper question: what happens when an organization founded as a nonprofit for the benefit of humanity becomes one of the most valuable technology companies in the world?

My view is balanced. OpenAI clearly creates social value. A high-tech company that improves productivity, changes how people write, learn, code, research, and work cannot be reduced to a mere profit machine. A legal, useful, innovative company already contributes to society.

At the same time, Musk’s criticism cannot simply be dismissed. If an organization receives donations, talent, legitimacy, and public trust under a nonprofit mission, then later becomes a commercial giant that may enrich employees, managers, and investors, that transition requires serious limits and compensation. Otherwise society creates a dangerous precedent: establish a nonprofit, raise money and trust, then convert that trust into private wealth.

1. A company can be a form of social value

A company does not begin to have social value only when it gives money to charity. It creates value by employing people, paying salaries, paying taxes, training talent, serving customers, building useful products, and improving efficiency.

This is why I have always admired companies such as Microsoft and Apple. They did not merely become rich. They changed how humanity works, communicates, writes, calculates, creates, and organizes information. Steve Jobs showed that an entrepreneur can turn a technological vision into everyday life for hundreds of millions of people.

OpenAI now belongs to this category of transformation. ChatGPT, Codex, and multimodal AI tools are changing writing, programming, education, translation, research, and enterprise workflows. Whatever one thinks of OpenAI’s structure, its social impact is already real.

2. Why I chose the company path

Before turning thirty, I hesitated between politics, NGOs, and business. My studies helped me understand the value of public institutions and nonprofit work. But I also saw their limits: funding dependency, public-image constraints, slow procedures, and limited scale.

Business attracted me because it allows a person to choose a problem, organize resources, take risk, create a product, serve customers, and produce measurable impact. It is not only a matter of personal freedom. It is a matter of personal value creation.

Government, NGOs, and companies should not be trapped inside rigid definitions. They all revolve around the same core: social value. Government uses law and public authority. NGOs use mission and trust. Companies use products, services, employment, efficiency, capital, and markets.

3. The core of the OpenAI dispute

According to OpenAI’s public structure page, OpenAI was founded in 2015 as a nonprofit with the mission of ensuring that AGI benefits all humanity. In 2019, it created a for-profit subsidiary. Under the structure announced in October 2025, OpenAI Foundation controls OpenAI Group PBC, a public benefit corporation. The Foundation holds about 26% of the equity, Microsoft about 27%, and the rest is held by current and former employees and investors.

From OpenAI’s perspective, this evolution is understandable. Frontier AI requires immense capital, computing infrastructure, rare talent, and global deployment capacity. A pure nonprofit structure may not be able to sustain that mission.

But from the perspective of charitable trust, Musk’s objection is serious. If early contributors gave money, reputation, and resources because they believed in a nonprofit mission, the later transformation into a commercial giant cannot be treated as a purely technical corporate change.

4. What justice should protect

I am not a lawyer, but I think the court must hold two principles together.

First, it should not destroy a company whose business already creates major technological and social value. OpenAI is now a global infrastructure-level actor, and a disproportionate remedy could create enormous uncertainty.

Second, it should not normalize the conversion of charitable trust into private wealth. If nonprofits can be used as incubators for future public companies without serious limits, philanthropy itself will be damaged.

The best solution may not be to return OpenAI to 2015. A better path would be stronger governance: real Foundation control, greater transparency, mission audits, limits on excessive personal enrichment, clear treatment of nonprofit-era assets, and some form of compensation or value return to the original charitable mission.

Conclusion

OpenAI should not simply be destroyed, because it has created real technological and social value. But Musk’s criticism should not be laughed away, because it touches a fundamental issue: can charitable trust be converted into private wealth without limits or repair?

My position is therefore clear: recognize the social value of enterprise, but protect philanthropic trust. Business is not a sin. Profit is not a sin. But when a company is born from a promise made in the name of humanity, it must remain accountable to that promise.

References

• OpenAI: Our structure
• AP News: Musk v. OpenAI trial
• TechCrunch: OpenAI completes recapitalization
• TechCrunch: restructuring and IPO questions

Abstract

This essay continues the previous reflection on philanthropy and social enterprise. It is based on a 2014-2015 internship report written after working in business development at Les Joyeux Recycleurs in Paris. The central question is simple: can a commercial company create social value while remaining fully commercial?

The answer is yes, if the social value is built into the business model itself. Les Joyeux Recycleurs was not a traditional NGO. It sold a service, developed customers, organized collection routes, measured waste, managed relationships, and had to survive as a company. Yet the service it sold made office recycling simpler, more visible, and more practical for companies.

This experience shaped my later focus on electronics recycling. A company already creates social value by employing people, paying taxes, training workers, serving clients, and organizing useful services. But when its business model reduces waste, extends product life, improves resource circulation, and helps other companies act responsibly, it inherits the spirit of social enterprise even if its legal form is purely commercial.

1. Social value is not limited to nonprofits

We often divide society into three separate worlds: government provides public services, nonprofits solve social problems, and companies make money. Reality is more complex. Every company has social consequences. It creates jobs, pays taxes, trains people, organizes work, serves customers, and affects the environment.

The question is not whether a company has social impact. It always does. The real question is whether that impact is designed to be positive.

Les Joyeux Recycleurs showed me that a company can sell a real service while helping solve a collective problem. Office recycling is not only logistics. It allows client companies to turn environmental responsibility into daily practice.

2. Making recycling easy

The starting point is practical. Many employees and companies say they want to recycle, but recycling often feels complicated. Office waste includes paper, cups, cans, bottles, coffee capsules, cartridges, batteries, bulbs, pens, and many other materials. Each stream requires a different instruction, container, collection process, and recycling partner.

Les Joyeux Recycleurs turned that complexity into a service: colorful recycling boxes, regular collection, reporting, communication materials, and customer relationship management. The company did not merely tell clients that they should recycle. It made recycling easy to do.

That is the value of entrepreneurship: seeing why a good intention does not become action, then building a service that removes the friction.

3. Why this is a social enterprise model

Les Joyeux Recycleurs was a business because it sold a service, controlled costs, found customers, and needed recurring revenue. It also had a social and environmental dimension because its service helped companies reduce waste, engage employees, and make corporate responsibility operational.

Its model combined three layers:

• commercial value: clients paid for a service;
• environmental value: office waste entered better sorting and recycling channels;
• social value: recycling became visible, collective, and positive inside the workplace.

The internship report also highlighted partnerships with social actors such as Ares Atelier, showing that recycling can connect environmental action, employment inclusion, and local networks.

4. A company already carries social responsibility

A company does not start creating social value only when it makes donations. A well-run company already contributes by creating jobs, paying salaries, paying taxes, training people, serving customers, and building useful systems.

But this basic contribution becomes stronger when the business model itself reduces a social or environmental problem. This is why electronics recycling matters to me.

5. Why electronics recycling

Electronic waste is not ordinary waste. It involves metals, batteries, plastics, data security, repair, reuse, compliance, and resource efficiency. According to the Global E-waste Monitor 2024, the world generated around 62 million tonnes of e-waste in 2022, while formally documented recycling remains limited.

Every company now uses computers, phones, monitors, cameras, servers, routers, and connected devices. When these products age, they create asset waste, storage problems, data risks, environmental risks, and management costs.

Electronics recycling therefore needs to become a professional service: collection, sorting, diagnosis, repair, reuse, data erasure, traceability, and responsible treatment.

6. Inheriting the spirit of Les Joyeux Recycleurs

My company may be a normal commercial company in legal form. That does not prevent it from inheriting the spirit of social enterprise. The key is not the label. The key is the model.

If the business reduces electronic waste, extends product life, helps companies manage equipment responsibly, creates jobs, pays taxes, and builds reliable processes, then social value is not outside the business. It is inside the business.

Les Joyeux Recycleurs taught me that environmental responsibility must be made simple, measurable, and repeatable. Electronics recycling follows the same logic: transform an invisible problem into a useful service.

Conclusion

A true social enterprise is not necessarily a company that refuses profit. It is a company that earns revenue by solving a real problem. Profit then becomes a condition for continuity, not a contradiction.

Les Joyeux Recycleurs showed me that a company can turn environmental responsibility into everyday practice. Electronics recycling extends that lesson into the digital age. It treats used devices not as useless waste, but as assets, materials, risks, and responsibilities that need professional management.

The company I want to build does not place social responsibility beside commerce. It tries to make commerce itself part of the solution.

References

• Les Joyeux Recycleurs
• Vie-publique: French Social and Solidarity Economy law dossier
• ADEME: electrical and electronic equipment data 2023
• European Commission: Waste from Electrical and Electronic Equipment
• Global E-waste Monitor 2024

Abstract

This essay is based on the central argument of a Master 1 thesis in International Cooperation and Solidarity. The original thesis examined Chinese foundations, the China Youth Development Foundation, Project Hope, nonprofit marketization, information systems, and social enterprise. Its deepest starting point, however, was not technical. It began with a personal experience after the 2008 Wenchuan earthquake: volunteering in the aftermath of disaster revealed that the value of life is not measured only by wealth or power, but by what one is able to give and make possible for others.

The core argument is that Chinese philanthropy has moved from administrative mobilization toward a more complex model involving social mission, public trust, professional governance, market-based tools, information systems, and sustainable social innovation. The future of nonprofit organizations will not be decided only by how much money they raise, but by how well they design projects, manage data, protect trust, evaluate impact, and build long-term capacity.

1. The starting point: from compassion to capacity

The Wenchuan experience changed a basic question: what is life for? In the face of disaster, abstract ambition becomes less convincing. What matters is whether people can reduce suffering, restore dignity, and help others regain the ability to act.

This is why philanthropy cannot be reduced to donation. Real philanthropy is organized responsibility. It turns emotion into structure, sympathy into service, and temporary help into long-term capacity.

2. Chinese foundations between state, society, and market

Chinese foundations developed in a particular historical context. In the early phase, administrative resources and government credibility helped foundations mobilize money and public participation. This was useful because social trust and independent philanthropic capital were still limited.

But the same mechanism also created dependence. If a foundation relies too heavily on administrative power, it may lose independence, innovation, and responsiveness to beneficiaries.

The 2004 Regulation on the Administration of Foundations gave the sector a clearer legal basis. The revised Charity Law, effective in 2024, further reflects the institutionalization of charity in China. The sector has therefore moved from moral enthusiasm toward a more regulated and professional field.

Chinese foundations can be understood through three mechanisms: administrative mobilization, social mission, and market-oriented self-sustainability. A mature nonprofit should be able to work with government, society, and market tools without being captured by any single one of them.

3. Why Project Hope succeeded

Project Hope succeeded because it connected three powerful themes: education, poverty, and children. Education represents the future, poverty reveals injustice, and children embody collective hope. By combining these themes, the project made a large social problem understandable and actionable for the public.

The success of Project Hope also shows the importance of nonprofit marketing. Marketing in this context is not manipulation. It is the ability to translate a social problem into a trustworthy public action.

Public information from the China Youth Development Foundation states that by the end of 2022, Project Hope had received more than RMB 21 billion in cumulative donations, supported more than six million students from disadvantaged families, and helped build more than 21,000 Hope primary schools. This makes Project Hope one of the most influential philanthropic projects in modern China.

4. The danger of success: why the mission must evolve

A successful project can become a trap if it prevents an organization from seeing new needs. Project Hope answered a historical question: how can children in poor areas gain access to basic education? But as public education policy and student support systems developed, the problem changed.

Youth development today involves mental health, vocational education, urban integration, employment, digital skills, legal awareness, social participation, and rural-urban inequality. A youth foundation cannot remain limited to building schools or funding tuition. It must help young people build durable life capacity.

5. Marketization of NGOs: efficiency without losing mission

The marketization of NGOs does not mean turning nonprofit organizations into ordinary businesses. It means using professional tools to improve public value: needs analysis, project design, donor relationship management, communication, evaluation, transparency, and data governance.

Nonprofits are often more complex than companies. In business, the payer and the user are usually close to each other. In philanthropy, donors pay, beneficiaries receive, organizations implement, the public monitors, and government regulates. This complexity requires more professional management, not less.

6. Information systems: trust needs evidence

Trust cannot rest on moral language alone. It needs evidence. A modern foundation must track donations, beneficiaries, project implementation, feedback, financial transparency, and long-term outcomes.

For a large foundation, information systems are not technical decoration. They are governance infrastructure. Without data, a nonprofit cannot know what works, what fails, what should be scaled, and what should be stopped.

7. Social enterprise: a path toward sustainable social innovation

Social enterprise tries to overcome the permanent dependency on donations. It combines economic viability with social purpose. It sells products or services, but the main goal is not profit maximization. The main goal is to solve a social problem sustainably.

For China, social enterprise may become an important bridge between government, market, and civil society. It is especially relevant for problems such as youth employment, vocational education, rural development, mental health, disability inclusion, and the digital divide.

Conclusion

The central idea of the original thesis remains powerful: philanthropy must move from emotional charity to institutional capacity. It is not enough to want to do good. Nonprofits must become organizations capable of understanding problems, designing effective interventions, managing resources, earning trust, and lasting over time.

The true goal of philanthropy is not moral display. It is to help people regain a future.

References

• Regulation on the Administration of Foundations
• Decision revising China's Charity Law
• China Youth Development Foundation public UPR document
• Peter F. Drucker, Managing the Non-Profit Organization.
• Philip Kotler and Alan R. Andreasen, Strategic Marketing for Nonprofit Organizations.
• Muhammad Yunus, Building Social Business.

Introduction

If we reduce the European Union to a trading arrangement, an administrative machine, or a slow-moving bureaucracy, we miss its real historical importance. The EU matters because it is one of the modern world’s rare large-scale experiments in shared political order: sovereign states, long shaped by rivalry and war, chose to place part of their authority inside common institutions in order to escape the old cycle of fear, balancing, and destruction.

That is why the EU should not be treated as a purely European technical story. It is one of humanity’s most serious attempts to answer a broader question: can large political communities cooperate beyond the nation-state without collapsing into empire, domination, or permanent geopolitical competition?

I. What the European Union is — and what it is not

Before discussing its history, we need to define it clearly.

• The European Union is a treaty-based political and legal union with 27 member states.
• The euro area is the group of states using the euro as official currency. According to the European Central Bank, as of 2026 the euro is used in 21 of the EU’s 27 member states.[1]
• The Schengen area is the free-movement space without systematic internal border checks. According to the European Commission, it includes 29 countries in 2026, including 4 non-EU states.[2]
• Europe is the wider geographical and civilizational space, which is larger than the EU.

The EU is therefore not a normal international organization, but neither is it a completed federal state. It is a treaty-based union built around shared sovereignty, common law, multi-level institutions, and differentiated integration.

Core features

• shared sovereignty rather than total isolation of national power;
• a common legal order with concrete effects;
• dual legitimacy through both member states and citizens;
• differentiated integration;
• rule-based coordination through law, courts, budgets, markets, and common policies.

II. The origin of the Union: a political answer to the ruins of war

The deepest origin of the EU lies in post-1945 Europe. Europe had been the center of modern industry, science, finance, constitutional politics, and global power, but it had also become the center of modern self-destruction: imperial competition, extreme nationalism, total war, and mass killing. In that context, Franco-German reconciliation was not merely diplomatic; it was civilizational.

The Schuman Declaration of 1950 offered a decisive answer: peace should no longer depend only on promises or conferences, but on institutions that make war structurally less likely. The European Coal and Steel Community of 1951 took the industries most closely linked to armament and placed them within a shared framework.[3]

The Treaty of Rome in 1957 expanded integration into a wider economic project. Over time, Europe built the common market, widened the role of European law, expanded the authority of the Parliament, opened internal borders, created a common currency, and ultimately formalized the European Union in the Maastricht Treaty of 1992. The Lisbon Treaty, in force from 2009, further stabilized the institutional architecture.[4]

III. A timeline of European integration

1950  Schuman Declaration
1951  European Coal and Steel Community
1957  Treaty of Rome
1985  Schengen Agreement
1992  Maastricht Treaty
1999  Euro launched as accounting currency
2002  Euro banknotes and coins enter circulation
2004  Eastern enlargement
2009  Lisbon Treaty enters into force
2016  Brexit referendum
2020  United Kingdom formally leaves the EU
2026  EU at 27, euro area at 21, Schengen at 29

IV. Through which channels does the EU govern?

It is not accurate to say that the EU “governs Europe” in the same way a centralized state governs its territory. More accurately, the EU organizes common governance through several channels.

1. Institutions

• The European Council sets strategic direction.
• The European Commission proposes and executes many common policies.
• The European Parliament represents citizens.
• The Council of the European Union represents national governments and co-legislates with Parliament.
• The Court of Justice of the European Union guarantees uniform interpretation of EU law.
• The European Central Bank manages monetary policy in the euro area.

2. Law and legislation

One of the EU’s distinctive strengths is that it does not remain at the level of declaration. In many fields, the Commission proposes, Parliament and Council adopt legislation under the ordinary legislative procedure, and the Court gives the system coherence through legal interpretation.[5]

3. The single market

The single market is one of the Union’s most powerful organizing tools. It is not merely free trade. It is a rule-based economic space structured to make the movement of goods, services, capital, and people more coherent across borders.

4. The euro

The euro is one of the boldest projects in European integration. But it also reveals the Union’s complexity. The euro area is not the whole EU, and monetary union has not automatically produced full fiscal or political union.

5. Schengen

Schengen transformed daily European life by making movement for work, study, tourism, and family life much easier. But Schengen and the EU are not identical. Some non-EU countries participate in Schengen, while the path into Schengen for EU members has not always been simultaneous.

6. Budget, common policies, and regulatory power

The EU also acts through budgetary instruments and long-term policy frameworks: agriculture, cohesion funds, research, competition law, trade policy, digital regulation, environmental policy, consumer protection, and educational exchange.

V. What has the EU achieved so far?

• durable peace among member states;
• one of the world’s largest integrated markets;
• a common currency across 21 member states;
• a Schengen area of 29 countries;
• global rule-shaping capacity in trade, privacy, competition, and standards;
• the gradual integration of Central and Eastern Europe into a more stable legal and political framework.

VI. Three national paths: France, Germany, and the United Kingdom

France

France has often understood Europe not merely as a market but as a political and strategic space. It has tended to emphasize public power, common ambition, and a civilizational meaning of Europe.

Germany

Germany’s European path is deeply marked by the postwar settlement. European integration became a framework for stability, legal credibility, restraint, and reinsertion into a trusted political order.

United Kingdom

The United Kingdom contributed greatly to Europe, but often from a greater distance. Its approach was more market-oriented, parliamentary, and wary of deep pooled sovereignty.

These three paths help explain why the EU is structurally complex: it must constantly accommodate different understandings of law, sovereignty, markets, and political belonging.

VII. Simplified institutional diagram

European Council
  ↓ overall direction
European Commission ── proposes ──┐
                                   ├── European Parliament + Council of the EU: co-legislation
Court of Justice ── interpretation ┘
ECB ── monetary policy for the euro area
Member states ── implementation and domestic adaptation

VIII. The limits and tensions of the Union

No serious discussion of the EU should romanticize it. Its weaknesses are real.

• democratic distance or democratic deficit;
• economic and political asymmetries among member states;
• tension between monetary integration and fiscal solidarity;
• persistent questions of identity, migration, borders, and belonging.

IX. Was Brexit a step backward?

My answer remains yes: in historical direction, Brexit was largely a step backward. It brought back the imagination of solitary sovereignty at precisely the moment when large political problems had already become transnational.

Britain gained symbolic autonomy, but it lost part of its structural influence over the making of European rules. At the same time, Brexit revealed real weaknesses inside the Union itself: democratic distance, territorial imbalance, and identity anxiety.

X. Conclusion

The European Union is one of the boldest attempts in modern history to organize political coexistence beyond the narrow frame of isolated national sovereignty. It remains unfinished, imperfect, and tension-filled. Yet it remains historically precious because it demonstrates that large-scale cooperation can be built through law, institutions, markets, and negotiated power-sharing.

XI. Treaties and legal sources

• Schuman Declaration
• Treaty on European Union (TEU)
• Treaty on the Functioning of the European Union (TFEU)
• EU countries
• EU institutions and bodies
• European Parliament legislative powers
• The ECB and the euro
• The Schengen area

XII. Notes

• [1] The ECB indicates that, as of 2026, the euro is used in 21 EU member states.
• [2] The European Commission states that the Schengen area includes 29 countries in 2026.
• [3] The originality of the Schuman Declaration lies in pooling the strategic industrial resources that had made war possible.
• [4] European integration did not advance in a straight line; it was shaped by crises, enlargements, treaty revisions, and political compromises.
• [5] The ordinary legislative procedure reflects the Union’s dual legitimacy: citizens and member states both participate.

XIII. References

• Dinan, Desmond. Europe Recast: A History of European Union.
• Milward, Alan S. The European Rescue of the Nation-State.
• Judt, Tony. Postwar.
• van Middelaar, Luuk. The Passage to Europe.

Final judgment

The EU should not be judged as a perfect machine. It should be judged as one of the boldest modern attempts to organize common political life beyond strict national sovereignty.

Camera NAS Recorder is a free Windows desktop app for recording USB or built-in cameras as MP4 files, saving them to a protected local cache first, and syncing finished recordings to a QNAP NAS when available.

Why I built this app

I did not build this tool because camera recording products do not exist. I built it because the real operational problem inside the company was still not solved well.

We have many places that need cameras, and we also bought multiple QVR-style camera devices. In practice, those devices store footage separately, their storage capacity is inconsistent, and their stability is uneven. Over time, that creates very practical problems: recordings are fragmented, playback is inconvenient, management becomes expensive, and it is hard to bring everything into one reliable long-term storage system.

What I actually needed was not another heavy management platform. I needed one core thing to work well first: move all camera recordings toward a more stable, larger-capacity, centrally managed place.

That is why Camera NAS Recorder is intentionally simple in its design:

• record locally first
• avoid losing footage when the network is unstable
• sync only when the target storage is available
• gradually consolidate scattered camera recordings into one place

For me, the value of this app is not to create a flashy surveillance system. It is to solve a long-standing operational problem in the most direct and stable way possible.

Why the old setup became hard to manage

Many teams start with a simple approach: each device stores its own footage. That works at first, but it becomes harder to maintain as the number of cameras grows.

Recordings become fragmented. Capacity is uneven across devices. Stability varies from one point to another. Expansion becomes messy because you are not scaling one system, you are just adding more isolated storage islands.

The bigger issue is management overhead. Playback, troubleshooting, retention planning, and backup discipline all become more manual than they should be.

Why I chose local cache plus NAS sync

I did not start by trying to build a massive surveillance platform. I started by solving the most important reliability problem first: create recordings consistently, then consolidate them consistently.

Local cache matters because footage lands on the Windows machine immediately. If the network is unstable or the NAS is temporarily unavailable, recording does not stop at the worst possible moment.

NAS sync matters because final storage becomes centralized instead of scattered. That makes capacity planning, backup strategy, playback, and future expansion much easier to manage.

This is a deliberate tradeoff:

• the local layer absorbs instability
• the NAS layer provides centralized storage
• the app connects the two in a practical workflow

For small and mid-sized teams, that is often a more realistic path than rolling out a heavyweight VMS from day one.

What it does

• Live camera view before recording
• Continuous MP4 recording
• Motion-only recording when movement is detected
• Date and time overlay in the top-left corner
• Local cache protection when the network is unstable
• Optional automatic sync to QNAP NAS
• Playback from the built-in recordings list
• Auto-start after Windows sign-in

Who it is for

This app is built for simple Windows-based camera recording setups: connect a camera, keep a local history of recordings, and sync to NAS only when you want to.

It is especially suitable when:

• you already have multiple Windows-connected cameras
• you want stable recording first, centralized storage second
• you need something more unified than isolated device storage
• you do not want to deploy a complex and expensive centralized platform yet
• you care more about reliability and consolidation than about a huge management suite

When it may not be the right fit

If your starting point is a full enterprise video management platform, this tool is not designed to replace that entire category.

It may be the wrong fit if you need:

• large-scale orchestration across hundreds or thousands of streams
• centralized permissions, complex audit flows, and role hierarchies
• cluster-level failover and high-availability architecture
• advanced analytics, alerting pipelines, and rule engines
• an organization-wide video operations control plane from day one

My goal is not to compete with those systems directly. My goal is to solve the more basic and more common problem first: record reliably, then consolidate cleanly.

Downloads

• Download Windows Installer (.exe)
• Download ZIP Package
• Open product page

Basic setup

1. Download and install the app. 2. Open Camera NAS Recorder. 3. Add a camera in the Cameras tab. 4. Use Live View to confirm the picture. 5. If you want NAS sync, enter your QNAP share path and credentials. 6. Click Start Recording.

If FFmpeg does not list the camera directly, you can still try index:0, index:1, or index:2.

Disclaimer

This is free software provided as-is, without any express or implied warranty.

The author and publisher accept no responsibility for:

• data loss, missing recordings, or failed sync operations
• device compatibility problems
• issues caused by Windows, drivers, NAS devices, network conditions, or third-party environments
• any direct, indirect, incidental, or consequential damages

Please test, validate, and maintain your own backups before using it in any important setting.

Public release notes

The current public release ships with an English interface and does not include any personal NAS path, username, password, or preconfigured cameras by default.

Many developers first encounter Codex as if it were just a stronger autocomplete tool, or a chatbot that happens to write code. Both descriptions are incomplete. The more useful framing is this: Codex is a coding agent built for real software tasks. It does not only suggest the next line. It can read a codebase, identify the right area to work in, modify files, run commands, check results, and keep iterating toward a clear goal.

1. What Codex actually is

In its current form, Codex is positioned as a tool for real engineering work rather than isolated code snippets. It can appear in a web workflow, an IDE, a terminal, or an API-driven internal toolchain. Its defining strength is not just speed of generation, but the ability to advance a technical objective with context and constraints.

Typical examples include:

• exploring an unfamiliar repository and finding the right entry points
• fixing a reproducible bug and validating the outcome
• refactoring legacy code while preserving a public interface
• handling multiple bounded tasks in parallel and summarizing the result

2. How it differs from ordinary AI coding tools

Traditional coding assistance is often optimized for the next line, the next function, or a quick answer to a pasted error. Codex can do those things too, but it becomes much more valuable when the work looks like a coherent unit of delivery.

That means the question changes from: “What code should come next?” to: “How do we move this task to a reliable finished state?”

3. What Codex is best at

Codex tends to perform best when the task is well-bounded. Strong use cases include:

• implementing a clearly defined feature
• fixing a known bug
• adding or improving tests
• refactoring code with explicit guardrails
• understanding a new codebase quickly
• updating technical documentation
• performing a focused code review

When the product problem itself is still vague, Codex will naturally be less stable because the work is underdefined.

4. Why some teams love it and others feel underwhelmed

The biggest difference is usually not the model, but the prompt structure and the surrounding workflow. A vague request like “optimize this project” leaves too much ambiguity: performance, architecture, UX, maintainability, deployment, or SEO?

A much better pattern is:

• Goal: fix the login flow that sometimes drops users back to the homepage
• Context: inspect auth/, middleware.ts, and callback routes
• Constraints: do not change the schema or public API
• Done when: the bug no longer reproduces and the relevant checks pass

That kind of framing often changes the result completely.

5. The most useful ways to work with Codex

1. Let it inspect before it edits

On harder tasks, a quick repository read is often more valuable than immediate code generation. Asking it to map the architecture, data flow, dependencies, and risks first usually leads to better changes.

2. Always provide Goal, Context, Constraints, and Done when

This is one of the most practical habits you can build. A small amount of structure dramatically improves reliability.

3. Ask for verification, not just code changes

A better workflow is not “change the code and stop,” but:

• run relevant tests when possible
• provide manual verification steps when tests are absent
• explain root cause and regression risk

That turns the output from a raw patch into something much closer to a deliverable.

4. Put recurring project rules into AGENTS.md

If you repeatedly work with Codex in the same repositories, codifying the working rules saves time and reduces drift. Useful contents include project structure, commands, sensitive files, review expectations, deployment notes, and rollback guidance.

5. Connect external context instead of copy-pasting everything

In real engineering work, crucial context often lives outside the codebase: tickets, docs, databases, APIs, observability tools, or operational runbooks. Reliable context wiring usually beats manual copy-paste.

6. Keep one thread focused on one coherent unit of work

A login bug, a UI redesign, a deployment issue, and a documentation cleanup should not all compete inside the same thread if you want clean reasoning and consistent execution.

7. Start with a medium reasoning setting

For many interactive tasks, a medium reasoning setting is a strong default because it balances speed and depth. Raise it only when the problem is meaningfully more complex or ambiguous.

8. Be careful with internet and execution permissions

The more action authority Codex receives, the more it should be treated like a real execution agent. Domains, methods, dependency installation, and secret access should stay intentionally scoped.

6. The most common mistakes

• giving a vague request
• forgetting to define constraints
• failing to define what done means
• treating it like a search engine instead of a task agent
• opening permissions too broadly too early

7. Who gets the most value from Codex

Codex is especially useful for:

• developers moving across multiple repositories
• technical founders balancing product and execution
• engineering leads who review, refactor, migrate, and ship changes
• teams trying to systematize repetitive development work

8. Summary

> Codex is not just a tool that writes code. It is an engineering agent that can move a real software task forward inside an actual working context.

The clearer the boundaries, context, and finish line, the more reliable it becomes.

9. Official references

• Codex Get Started
• Codex Cloud
• Codex Best Practices
• GPT-5.3-Codex

For most teams, the best place to begin is not exhaustive theory. It is a small, clearly bounded task with explicit constraints and verification. That is usually where Codex becomes concrete and useful very quickly.

When people first open Codex, they often get hit by a wall of unfamiliar terms: thread, skill, app, automation, worktree, commit, Git, MCP, and more. If you are not a programmer, that can feel like an entirely different language.

The good news is that the most useful ideas are much simpler than they look. For a non-programmer, the real goal is not to learn programming syntax first. The real goal is to understand how to turn Codex into a practical workspace that helps you move projects forward.

1. The most useful mental model

Start with this translation:

• project: a thing you are building, usually a folder
• workspace: the folder Codex is currently focused on
• thread: a task track with memory and context
• skill: a specialized way of handling a certain kind of task
• app: a connection to an external service or tool
• automation: a repeated task that runs on a schedule
• commit: a saved checkpoint of the project
• worktree: a parallel copy of the same project

Once you see the interface through that lens, it becomes much less intimidating.

2. What a thread really means

A thread is not just a chat log. It is really the memory of a task.

If you use the same thread for fixing a website, summarizing a book, and debugging a server, the context starts to blur. That is why one coherent goal should usually live in one thread.

Simple rule:

• one major goal = one thread
• when the topic clearly changes, start a new thread

3. What a skill is

A skill is a specialized way of doing a certain type of work.

Examples might include:

• analyzing screenshots
• organizing Gmail
• deploying a website
• consulting specific documentation

For you, the practical meaning is simple: skills make certain task types more reliable and more efficient.

4. What an app is

An app is more like a connection to an outside system.

A simple way to remember the difference is:

• skill = how to do the work
• app = where or through what system the work is connected

For example, Gmail is an app connection. Inbox triage is a task pattern or skill.

5. What automation means

An automation is a repeated task that you hand over to the system.

Examples:

• check every morning that a website is online
• summarize messages and drafts every week
• run a VPS health check every month

The safest order is almost always:

1. do the task manually once 2. make sure the process is right 3. automate it only after that

6. What Playground is, and whether everything should live there

Not everything should live there. Playground is best thought of as an experimental desk: tests, prototypes, temporary ideas.

Long term, real projects are easier to manage when they each live in their own folder. Otherwise files, history, and context all get mixed together.

A healthier structure usually looks like this:

• one dedicated folder per important project
• a separate area for experiments
• different websites kept separate from each other

7. What the top buttons mean

Commit

A commit is a save point. Think of it as a project checkpoint.

Good moments to commit:

• after a stable fix
• after a feature is finished
• after a stage has been validated

A commit is not a launch and not a deployment. It simply means: “this version is worth keeping so I can return to it later.”

Worktree

A worktree is a parallel copy of the same project. It is useful when you want to try a major change without disturbing the current version.

If you are new, you do not need to use worktrees right away. Threads and commits already get you very far.

8. The settings, explained in plain language

General

This is where everyday behavior usually lives: language, notifications, speed, detail level, sleep prevention, and similar options.

The most useful ones for non-programmers are often:

• interface language
• how much execution detail the thread shows
• preventing the computer from sleeping during long runs
• execution speed

Appearance

This changes how the app looks, not how the work itself functions.

Configuration

This often affects model behavior, execution style, or permissions. It is worth changing carefully.

Personalization

This is one of the most valuable sections. It is where you can store durable working rules, such as:

• my personal website articles should not be signed
• explain things in simple language
• warn me before risky actions

MCP

Think of this as external context connectors. You do not need it immediately, but it can become very powerful later.

Git

This is project history management. Even without coding knowledge, one idea matters most: Git helps you avoid losing a good state of the project.

Environment

This is more technical and usually best left alone unless you know why you are changing it.

Worktrees

Management for parallel project copies.

Archived threads

Old conversations to review, not necessarily the best place to continue new tasks.

Extra: what the buttons under the input box mean

The row of controls under the input box becomes much easier once you translate it into everyday language.

+

This is usually the place where you add context, such as:

• a file
• an image
• a reference document
• something else Codex should look at for the current task

GPT-5.4

This is the model currently being used.

In simple terms:

• a stronger model is often more reliable
• a lighter model is often faster

If you do not want to overthink it, keeping the default primary model is usually fine.

中

This is usually the reasoning level, not the language.

A simple way to think about it is:

• low: faster, for simpler tasks
• medium: balanced
• high: slower, but better for harder tasks

For everyday project work, medium is often a very solid default.

本地

This means the task is running against your local machine and your current workspace.

完全访问权限

This is the current permission level. Full access usually means Codex can read and modify files and run more commands locally. That is powerful, but it also means you should be clearer about what it is allowed to touch.

Microphone icon

Voice input.

Send button

Actually sends the request and starts execution.

main

This is usually the current Git branch. You can think of main as the primary, stable line of the project.

Refresh icon

This usually refreshes or resyncs the current workspace or branch state. It is useful when something changed and you want the UI to catch up.

9. The best way to use Codex if you are not a developer

A simple and effective workflow is:

1. one important project = one dedicated folder 2. one goal = one thread 3. begin by asking for a plain-language explanation before any change is made 4. before large changes, ask Codex to restate what it will change and what it will not touch 5. make a commit after each stable stage 6. automate only the tasks that already make sense manually 7. store long-term preferences in personalization

10. The three most important ideas

If you only remember three things, remember these:

• thread = task memory
• commit = stable checkpoint
• project folder = boundary

11. Why structure matters more than code at the beginning

For many non-programmers, the main problem is not lack of coding knowledge. It is lack of structure:

• mixed projects
• mixed goals
• no stable checkpoints
• unclear rules

Learning to structure work is often far more valuable at the beginning than learning syntax.

12. Summary

> Do not think of Codex first as a coding tool. Think of it first as a workspace that helps organize, execute, remember, and move a project forward.

Once threads, project boundaries, commits, and automation make sense, the interface becomes much easier to use and much more powerful.

A Simple Question

After reaching a certain stage in my career, I started asking a more fundamental question: what is a successful life?

If health is not there, all achievements lose their meaning...

My Story

1987, I was born in a small town in Sichuan, China...

(Full article coming soon)

(Full article in English coming soon)

(Full article in English coming soon)

Background

In 2015, I had just finished my Master's...

(Full article coming soon)