XuMiao

The claim that “tourism has little financial benefits to the local economy in Venice” is debatable and context-dependent. Here's a detailed breakdown addressing both why the claim may be true in some aspects, and why it may be misleading or false in others.

---

Arguments Supporting the Claim:

1. High Leakage of Tourist Revenue

Much of the tourist spending in Venice ends up outside the local economy:

Many hotels, cruise lines, and travel agencies are owned by foreign or non-local entities.

Revenue often flows to large tour operators, not to Venetians themselves.

Day-trippers (especially cruise passengers) spend very little per capita.

2. Overtourism and Cost Externalization

The externalities of mass tourism (e.g. garbage collection, water bus crowding, maintenance of ancient infrastructure) are borne by the municipality and residents, not by tourists.

The economic cost of wear and tear on fragile historical structures is immense and undercompensated.

3. Loss of Local Businesses and Services

Traditional shops and services (bakeries, fishmongers, schools) are being replaced by souvenir shops and Airbnbs, which often serve short-term tourists.

This creates a "hollow economy" where real life becomes unviable for locals.

4. Depopulation and Real Estate Inflation

Real estate is increasingly purchased by investors for short-term rentals, pushing locals out and reducing residential density.

Venice’s population has dropped from ~175,000 in 1950s to under 50,000 today in the historic center.

5. Low Multiplier Effect

Much of the employment created is low-paid, seasonal, precarious, and lacks career development.

Limited reinvestment into the community fabric (education, public health, sustainable infrastructure).

---

Counterarguments (Why Tourism Still Brings Economic Benefit):

1. Tourism Is a Major Employer

A significant portion of Venetian jobs is in hospitality, transport, and retail, all tied to tourism.

Completely removing tourism would collapse the current local job market.

2. Tax Revenues

The city imposes tourist taxes (tassa di soggiorno) on accommodations and more recently, even entrance fees for day-trippers.

These can help fund infrastructure and conservation—if well-managed.

3. Export Substitute

Venice doesn’t have a diversified industrial base. Tourism is one of the few export-equivalent services Venice can offer due to its geographic isolation and fragile ecosystem.

---

Conclusion

While tourism contributes significantly in gross economic terms, the net local financial benefit is undermined by:

revenue leakage,

rising costs of living,

poor job quality,

and infrastructure stress.

Thus, the statement is partially true: mass tourism as currently structured in Venice is unsustainable and offers diminishing marginal returns to locals, especially compared to the burdens it imposes

Algebra geometry view makes sense to me. Considering ML as a learning to approximate scheme algorithm. Tensor representation is similar to the SDP trick achieving the optimal max-sat approximation. The difference is that DL approximates from inside of the high dimensional space (concave) while SDP approximates from outside (convex). The later one turns into polynomial algorithm, but the former one remains NP-hard. The success of DL just proved that there is a long way to go for P equals NP. Whenever we figure that out, symbolic approach and Tensor approach will merge.

US was wiser than others because we were able to look at our mistakes and fix them admirably.

It's only a weak character who tries to demonize others in exchange for self-comforting. Sadly, I'm seeing more and more people doing that.

To be honest, many potential murders sacrifice themselves so much that I can live safely. I really should have thanked them for not killing me.

From the physics point of view, an intelligence is to keep the universe running in the perfect order, no waste of energy to increase the temperature. All accurate predictions align our current states to the future. If we look deep enough, we always find information coming from the past and the future. The only trouble is that our computation power is bounded, and computing itself generates heat waste.

Human always feel that we have free will to design. It's likely a mistake in our intelligence unless the universe is infinitely dimensional where we can process the energy flow in any way we want.

If quantum reality is objective, then probabilistic reality is objective.

Deterministic probability like Chaos or second law of thermodynamics imply the existence of incomplete information. This type of probabilistic reality might be subjective.

> This is a review paper. It's long. Past the first sentence I find it readable and well organized. It mentions some of the work on interpretability I'd expect to see, Finale-Velez, Rudin, Wallach, and LIME, but does not appear to mention Shapley. The bottom line conclusion is "In the end, the important thing is to explain the right thing to the right person in the right way at the right time." That's both an obvious truth and a differentiating mindset in research-first space. It's worth a skim.

People want to know whether some mathematical formulas can work. Then how do they work? Then what can make them work in a different way.

Explanability or interpretability leads to controllability at the end.

I rather see NNs with semantic meanings instead of semantic meanings from NNs. If human would like to control NNs, why not make them meaningful modules that can be composed like a regular program.

For example, instead of using CNNs or RNNs, we simply make a model by stating the definition:

Jaywalk :: (p: Person, scene: Image) := p in scene & exist s: Street in scene, walk_cross(p, s) in scene & not exist z: ZebraCross in scene, inside(p, z) in scene

Here predicates, walk_cross and inside, are neural network modules that might be used in many different problems. We can identify cases where the model make wrong predictions and modify the definition accordingly.

This is much human friendly development than tweaking parameters. After all, not everyone is fond of programming in NNs directly.

Exactly.

a+b should be the same as a.add.b

A high level programming language should be designed for communication instead of dictating the computations.

In terms of the statistical learning framework, the Bayesian optimal classifiers are NP hard to find under distribution free conditions. The models have rather large confidence intervals to compare with each other. Scientific methods can only lead you to a certain level of certainty, the rest are purely subjective.

If they don't buy it from the store, they have to make it themselves. Everybody are satisfied in the transaction because they played well on this trivial zero sum game instead of falling into a negative sum swirl.

By the way, economic concepts are statistics. Arguments are only meaningful on large number of samples or over long period of time.

Supervised learning algorithms assume that the input data are iid of the future. This is not valid in most of the real applications. The observation that we see men more than women in programming does not necessarily generalize to the future. That's why online learning provides an exploitation vs exploration mechanism to minimize the bias in the hindsight. In many applications, people just forgot about this simple strategy and blame the bias caused by supervised learning to the black box model.

Of course, black box AI itself is not the right solution. As more and more cross domain multitask settings emerge, open box AI will gradually take off. It is about compositional capability like functor and monad in functional language. Explanable or not is just a communication problem which is parallel to the ultimate intelligence problem. It is very possible that human intelligence is bounded.

The fear is that jobs left to humans are those low level ones. Technology divide up the society into elites who control machines and labors who is controlled by machines. And gradually, the elites become less and less until the day we humans are all batteries.

Python does have a lot of problems, but it's so far the most flexible language like English.

High level programming language is meant to communicate, not to dictate the computation. Python happens to be easy to understand by humans.

Most of human knowledge are represented in logic. Semantic web is designed from description logic. A less powerful logic than first order logic.