jojo2000

Let me explain my line of reasoning here (been there at least three times, situation gave 3 different outcomes) :

- case 1 : need to make calculations with a specialized library in C (precision arithmetics). Build a bare FFI, later replaced with CFFI.

- case 2 : loop-intensive on very simple calculations, time-constrained development. Identified as horrible performance in python : tried numba didn't work, ended up using cython, worked really well.

- case 3 : optimize numpy-intensive routine for performance. Tried numba, didn't work. Too expensive to recode in cython. Look at numpy C/C++ interfaces, and numpy-friendly C++ alternatives. Also try to trick numpy to function better (do not ever try this, it'll give worse results). Ended up doing nothing as time to develop was the main constraint here.

If you have a lot of time and work on a small program, maybe you can spend the time to optimize. Team producing lots of complicated algorithms and no way to re-develop everything, stick to python, identify perf losses and choose wisely what you'll optimize.

Numba has progressed but is still not a "drop-in decorator" as advertized. Can even give worse performance in some cases. Nevertheless the idea is good and I praise the effort, when it's done it'll be massive !

this one ? https://www.statnews.com/2020/09/09/astrazeneca-covid19-vacc...

I persist about what I said, waiting for someone to demonstrate this is not the case ? Furthermore, there are a lot of controversies regarding HepB and HPV vaccines.

It's one thing to induce our inner defense mechanisms, it's something different from using our cells as a factory like we use bacterias by modifying their dna. Everything is crystal clear, no need to complicate things !

numba is a true horror to use, because supported subset of python language is an ever-moving target and overall small. cython is good but needs time to meddle with. For loop-intensive, simple function I'd advise FFI.

Having worked for a thriving competitor of those brands, I can tell you that the app was not the major factor of success for the company. Our infra, at the time, was retarded, the code quality was mediocre, and we witnessed major downtimes, fun times. You have to come at the right time (for vc money), organize operations efficiently (recruit drivers + passengers) and make some lobbying to not be banned outright, while managing angriness from drivers, and unhappy clients. This is the hard part. The tech is basic !

I would work as a scientist for a better planet, working on advanced topics that may not achieve anything or may change things forever. I'd work on OSS in my spare time to give back to the community.

I'm a physicist and have worked on rocket science before, but the pay was a pity and the boss was shitty.

SaaS is the only way to make an ample living in my country, the only jobs where competence makes a change instead of which university you graduated...

Congrats guys :) Your product will get the traction it deserves !

> "The Damascus-like sample was significantly stronger, holding up to about 20 percent more stretching force. It didn’t reach the strength of a typical, traditionally made maraging steel, but the researchers note that this requires “a time-consuming and costly post-process ageing heat treatment.”"

Of course. Damascus steel gets its properties because of folding metal times and times and hammering. Piling layers with different properties has nothing to do with making "something that looks alike but is not at all" damascus steel...

If everyone agreed on everything, the world would be less fun ! My main problem with nuclear fission is that we generate a lot of waste that is almost impossible to dispose of and very dangerous. We don't even know, in case of a civilization breakdown, how people would manage the situation if knowledge was lost. [0]

edit : ray of hope here [1b]

Currently there has been many accidents with nuclear power, and we don't know how to manage those. Fukushima is still releasing lots of radioactivity in the sea. Yet radio activity about this is close to zero. [1]

And new ways to tap into fission power (pressurized reactors) are dangerous and hard to build.

Yet, I shall agree with you, that there is no definite solution to the energy problem, consuming less would be a good start.

[0] https://en.wikipedia.org/wiki/Into_Eternity_(film)

[1] http://www.asahi.com/ajw/articles/13228942

[1b] https://futurism.com/the-byte/nobel-prize-lasers-destroy-nuc...

Model around consumption should get deeply revised if we want to leave the world in a good state for our children.

Refurbish rather than build new buildings

Build sturdy hardware with replaceable parts

Build from natural materials as much as possible, and invent a new concrete

Stop polluting by sprinkling deadly stuff to grow crops and make things

Develop recycling aggressively (98%)

Stop using oil derivates everywhere

Stop using nuclear fission as a power source

...

Problem is that it is antipodal with current economical growth models

There are a lot of interesting developments in the haptics world, yet we have to see something really interesting and different from vibrations to get to the next level..

I tried to build a startup around haptic technology which could have changed things in that space. It didn't work out as we couldn't get funded. Still, a prototype is laying on a shelf somewhere...

> Despite the coronavirus outbreak causing a downturn in home sales and listing activity through March and April, Redfin managed to beat expectations with its first quarter earnings as the company posted a net loss of $60 million. A year ago, Redfin had reported a larger net loss of $67 million, for comparison.

That's a weird way of "beating expectations"

I couldn't find a single radioactivity sensor in europe urad's network. regarding the first website, I couldn't find a way to have a range bigger than 24h.

my bad, didn't look at the options enabling wider time frame, thanks !

yes, great resource. Still,

I know there exists and official network of radiation detection in Europe yet data is not accessible (except maybe here [0])

and we would need historical data to see fluctuations over natural threshold (more than 24H)

[0] https://era.org.mt/en/Pages/Ionising-Radiation-Monitoring-St...

In order to assess the danger you need to monitor realtime exposure to radioactive particles. There is no measurements here. It seems the data exists in Europe but not publicly accessible which is a shame.

The Tchernobyl accident released a lot of radioactivity in the air.

The soil contamination is twofold :

- spotty : some rocks that blew thru the explosion highly contaminated some spots.

- distributed : the plume contaminated the soil with material of different half-lifes.

The problem here is that trees are recycling organic matter that was tainted by radioactivity and so, if they burn they release it in floating materials.

We cannot infer something meaningful from the plume's direction as the atmosphere has high-flowing air currents in upper atmosphere that redistribute the pollution.

Only a raise of radioactivity in realtime measurements will tell. I'm sure the effect can be seen in Europe currently, yet no one wants to scare people so information is hidden.

Goverments have a problem. People can exchange information without oversight.

This is frightening, to say the least.

It's a narrow viewpoint to only see it through the lens of death percentage.

Of course it is a game-changer.

You eradicate the viral load after 6 days instead of 20+. Meaning :

- reduced contagion (avoiding big peak of infections), no confinement needed, so the better for the economy - you avoid complications, meaning less load on hospital, so you can save a lot of people who would have died otherwise. - people get better in less time so less damage to their finances.

Yes, once they're on a vent, your main problem is not even the viral load, but rather that the body reacts to the infection in desperate ways like cytokine storm [0] which destroys organs.

[0] https://en.wikipedia.org/wiki/Cytokine_release_syndrome

It's the most heavily tested area of the world [0], and also the only place where everyone who is positive gets a treatment (hydroxychloroquine + azithromycine). Other protocols are listed here [1]

Regarding status, it seems the peak is being reached there [2]

[0] https://twitter.com/raoult_didier/status/1245978149206228992...

[1] https://twitter.com/raoult_didier/status/1242808646997880832...

[1] https://twitter.com/raoult_didier/status/1246003916325748736...

> In a sense, this is an example of the US responding in kind to something that Europe has already done.

Is this meant to be funny ? /s

Seems like a kiddy moral statement. Europe has done so many bad things in the past, but US currently has the lead !

This affair learns us some interesting things :

1) nation states are the current structure that defends people, a statement which was challenged by the creation of EU, and will try to protect their people before others.

2) Global trade cannot abolish the aforementioned statements.

Pick your conclusion. Mine currently is :

States have lost self-reliance based on global trade doctrine and this trend should be reversed. Chicago school has made enough ravages.

Seems like few people really read the stuff.

One sentence made me choke :

> "The network needs to provide specific QoS and security policies based on user identity, rather than mapping to something instead"

Well, no, sorry. It's an orwellian design proposal.

Cloudflare seems like a trustworthy and technically competent company. Kudos.

This. The leader is the target and his/her actions will be scrutinized and critisized.

Yet if you build upon your team and create mutual trust, people will follow you and work really hard. That's the rewarding part of the job.

Regarding higher-ups, you have to choose wisely, most of the time it's very difficult because you give them what they want, but it costs a lot of time, energy and money.

Be pedagogic, explain, and explain. The most difficult part is when you want to take action that doesn't give reward instantly (CI/refactoring/...). Until they've been bitten by this, they won't understand. But you won't know beforehand, so ask the question in interviews !

Promote your best teammates eagerly and give them responsibilities.

Share as much information as possible with the team.

Enforce collective decision processes based on technical merit and cost (RFC-style), to rally everyone to the chosen solution.

Hello, and thank you ! Exploring the dependency graph of this paper is a good starter [0]

[0] https://www.nature.com/articles/s41421-020-0156-0

Yes, you start with molecular studies, then in-vitro, then in-vivo, then in humans, and in the end, sometimes it gives good results, sometimes not.

You point the exact problem here.

As a scientist the first test is intuition. The second test is having people getting better with treatment. The third test is conducting a full, deep study with the utmost scientific rigor.

So now there is a dilemna. People die in front of you. You try something. It seems to work. It uses a widely-used set of molecules (Hydroxychloroquine + azythromycine (+Zinc in some protocols). Will you try it, or will you make a double-blind study with half the people at risk of dying ?

History of aviation. People just couldn't believe that it was possible.

http://www.wright-brothers.org/History_Wing/Aviations_Attic/...

"Not wishing to be miserly with his information, Root sent an eye-witness account of what the Wrights were doing to the Scientific American, with a letter telling the editor he was free to use it. But the editor was not to be taken in and made no effort to investigate what Root had dropped into his lap. Though the Scientific American printed in 1905 many articles about flying, nearly all were about devices that maybe ought to be tried."

I don't fail in anything, I just don't need to engage in endless debates about methodology. Half of the studies in psychology fail to replicate. There are problems everywhere in science, because geniuses are still rare and now ostracized because they produce new researched and are judged by lords of the old order. If you scratch hard, you can always find something.

Some people criticize, others just act. The people who create are always going to be the targets of the people who spend their time criticizing.

Now it's time to act. Not pull over waiting for some authority to decide for yourself. The list of publications is here to show that this available treatment doesn't come from a magical hat.

In those troubled times, a lot of people are just critisizing instead of doing something useful. There is a bibliography studying the effects of chloroquine and derivates to fight viruses. As scientists, the base of our work is to read papers and search for bibliography. Targeting people with ad hominem attacks is plain lame. Do some science FFS, and act swiftly in times of war.

https://www.sciencedirect.com/science/article/pii/S092485792...

https://aac.asm.org/content/53/8/3416

https://www.sciencedirect.com/science/article/pii/S0006291X0...

https://link.springer.com/article/10.1186/1743-422X-2-69

https://www.sciencedirect.com/science/article/pii/S016635420...

https://www.sciencedirect.com/science/article/pii/S147330990...

https://www.sciencedirect.com/science/article/pii/S016635421...

https://www.thelancet.com/journals/laninf/article/PIIS1473-3...

https://link.springer.com/article/10.1007/s00535-009-0132-9

https://link.springer.com/article/10.1186/1743-422X-3-39

Bad blocks bad blocks what ya gonna do, what ya gonna do when they come for you.

Hard drives fail hard or smooth, SSDs fail hard ! On a hard drive, we can picture how the blocks are stored (sequentially). On an SSD, there is a remapping of blocks in silicon+firmware, and this added layer adds its own bugs [0] [1].

[0] https://blog.elcomsoft.com/2019/01/why-ssds-die-a-sudden-dea...

[1] https://www.anandtech.com/show/6503/second-update-on-samsung...

(edit : added one reference more)

There are papers out there that qualify chloroquine and derivates as effective in case of viral infections with viruses related to covid-19 [0] and try to explain what the mechanism could be :

« The exact mechanism of antiviral intervention by chloroquine is not yet elucidated but is possibly a multitarget mechanism, depending on the time point at which the drug is added »

[0] https://aac.asm.org/content/53/8/3416

> The possible side effects include permanent blindness.

Yes, but only with massive doses or when treatment lasts for more than 5 years.