HN user

magic-chicken

47 karma
Posts1
Comments26
View on HN

The process to remove religions from public institutions started 50-70 years ago. It's a process, and its going to take a long time, as there is 400 years of history to deal with. Saying that the whole process is BS because there was some hesitation, or because there are still streets named after catholic Saints is just a logical fallacy.

I understand how it can be perceived this way. However, did you look at Quebec history? You will find that Quebec was very religious until 50-70 years ago. Education was provided by members of catholic institutions. At that point in time, teachers were wearing their religious uniform. Quebec made the decision to re-take ownership of the education system and to get the religions out of there. In the beginning, religious people were still providing the educuation, but they were asked to stop wearing religious symbols.

The current movement is a continuity of the work that was started 50-70 years ago to get religious symbols out of public institutions, and in that sense, it's coherent.

The point is that the crucifix was removed from the most symbolic place it could be removed from. If thats not a strong, consistent signal that Quebec wants religions out of the public institutions, then what is?

About religon: you are correct, it's not racism, it's a different approach to multi-culturalism. The problem is, the ROC is intolerant and only consider that one valid approach to multi-culturalism exists: their own. Anything else is "racist".

What? Are you suggesting that left-leaning voters have a lower level of testosterone than right-leaning voters?

After all, lead poisoning was suggested as one of the possible reasons for the decline of the Roman empire.

What are you trying to imply here? Microplastic -> Low Testorone -> Socialist government -> Decline of civilisation?

He asked : why is it important ? You answered : because it is important. His comment has value because he's questionning an ill defined 'authority'. You accused him of a commiting a logical fallacy while committing one yourself. I learned more by reading the answers to his question than by reading your comment.

I think that for most people in the article, their house was not under a flight path when they bought it :

>> The multibillion-dollar program is changing the way air traffic is managed, moving it from radar to satellite navigation. Proponents say it makes the air traffic system more efficient because it allows planes to fly more direct routes to their destinations.

But the shift has angered residents, who live in neighborhoods that are below the new flight paths. Residents in Northwest Washington sued the FAA over the changes but lost in court. A suit filed by the state of Maryland is pending.

I did not detect any bashing in the video, except toward the whole Euro. On the contrary, I found he was trying very hard not to point fingers at anybody. Also, he asked many times the audience to challenge future panelists (ex. Steve Bannon) on their ideas, and he did so in a very respectful manner. English is not my first language, so I may have missed some things. Could you expand on what part you think he was bashing on Germany or being condescending ?

TL/DR

Tesla ADS is the best on the market but it is marketed as an autopilot, which is wrong and caused some overconfident owners to die.

Assuming you are talking about software design:

- Make it work, then make it simple, then make it fast. In that order.

- Don't do at runtime what you can do at compile time.

- You are not going to need it.

- Encapsulate your dependencies.

- Push code complexity / logic to the edges (aka Tell, don't ask)

- Build only your core business software components. Use 3rd party software or open source for the rest.

- If at least one part of your system is not redundant, your system is not redundant.

- The persistance mechanism of your application is an implementation detail.

- Stateful systems are harder to debug than stateless systems.

- Almost always develop a business logic core for your application that is independant from it's distribution mechanism.

- When you make an API public, make everything you can to make the changes to it backward compatible. Also, use semantic versioning.

- If you are developing a distributed system, accept from the start that the other services on which you rely won't be availaible 100% of the time.

- Fail fast, fail often.

- Caching is hard to get right. Use it as a last resort.

If you were talking about visual design :

- Design for mobile first. Porting it to desktop will be easy afterwards.

- Prefer text over images. The more explicit the better.

- Read about typography to choose a good combination of font, line height, line width and letter spacing.

- Don't use colors to give meaning to something. Use shapes and colors instead to help people with color blindess.

- On a page, show all the information the user needs to make a decision or an action. No more, no less.

- Optimize actions that are performed often or that needs to happen fast. The less input the user needs to provide the better.

- Make navigating between popular sections of the site easy. It should not take more than one or 2 actions to go anywhere on your site / application.

- Learn about reading "hot spots" to decide how to arrange your content.

- Read about colors and how people of different cultures perceive them.

- Be consistent.

From my experience, regulated industries use a combination of the following techniques :

- Dont use passwords for databases. Make services use integrated authentication to interact with the database.

- Developers / operations dont have read / write access to production tables. It can be obtained momentarily for ad-hocs operations.

- When the database needs to be modified, the script goes through a change management system and it is executed at a planned time, ideally by another person.

- If you use private keys, ex for symmetric encryption, don't store them with the data it's used to encrypt.

- Use database level encryption on sensitive data to prevent anyone with read access to snoop on PII.

- Enable database level audit and change tracking.

- Developers don't push their own changes to production.

A given prediction can be wrong, an experiment may be biaised, my point is that you choose to ignore that the vast majority of the experiments and measurements point in the same direction.

>Researchers tend to stop looking for bugs in the model when it returns the results that they expect.

Again... they are _ALL_ biaised ?

So you test the system with human drivers for a few miles, and if none of them dies then it's secure ? You actually rely on the subjectivity of uninformed users to assess the reliabilty of a safety-critical system ?

> Not really. It's mostly subjective.

No. The quality of a safety-critical system is not subjective.

> I think most would generally agree with the quality assessment.

You are not qualified to make that claim, and neither are your users.

As you can see in the article, it already exists in Canada. Some engineering schools like École de technologie supérieure (ÉTS) offers a 4 year Software Engineering program that enable you to become a member of the Ordre des ingénieurs du Québec. It's a self-regulatory body that governs Quebec's professional engineers. There is a ceremony where they give you an iron ring, mostly to remind you to be humble and to always consider the public interest first when making decisions. Can software engineers become P.E. in the U.S ?