Okay let's go through this point by point.

"That's a very dangerous generalization. It is not true. Some scientists may be biased towards one direction, some others towards another, and many are actually only relying on facts. You have several groups working on similar topics and there is thus a competition. Competition is healthy. That's how new approaches are built."

Yes scientists may or may not have one bias or another.     But I think you're mistaken when assuming there will be many scientists equally working in competition.    Scientists require funding to do their research.   Facts don't matter when your job, your paycheque and your research grants depend on you producing a particular set of facts for your boss, be that a government or corporate organization.    If you provide the facts they want you get paid more.    If you don't then you lose your research grant and get fired, or the results of your research get surpressed and aren't allowed to be published.    

"First, I would like to push forward the fact that journals are not belonging to any governments. Those are just private companies trying to make as much as money as possible... Universities all around the world are paying yearly fee so that their employees can access the information. For the rest of the world, it is a field dependent statement. Particle physics papers are mostly open access, for instance, as we pay for that."

Cool but what about the people who want to read those journals and scientific papers who aren't in those particular universities?   What if you just want to do some research or are just curious?    Science should be something everyone can practice not just academic elites in their ivory towers.    You nailed it on the head there, yes scientific journals are run by publishing houses that just want to make money and that's the problem which inhibits real decentralized science.

"Some are already thinking about make scientific result available to everybody. For instance, on the steemit blockhain, have you checked the pevo project? Now, outside steemit, have you heard about the arxiv platform?"

No actually I haven't heard of those projects and will have to look into them thank you.

"Taking your proposal, who do you think could vote? Anyone? I don't think this is good. I would feel very uncomfortable to vote for something outside my field. I don't have any expertise to do so. Science must be reviewed by peers. The general audience can comment of course, but does Mr. or Ms. everybody have any knowledge to assess the quality of a scientific paper?"

Why would you feel uncomfortable voting on someone outside our field?   We make recommendations and give reputation to those outside our field of expertise all the time.    You don't need to be an expert mechanic to see if a professional fixed your car correctly.   If you're wanting to have corrective criticism on your craft that's one thing but I would put forward that those with "degrees" aren't the only ones who can think a) logically b) critically think and use the scientific method and c) have insight onto your paper.   I've read through scientific papers.    While the vocabulary is a bit advanced in places and they can get a bit tedious they're not impossible to understand even for Joe Everyman, he'd just need wikipedia to help with the vocabulary.      And if people are reading through scientific papers and studying science isn't that how one develops an education and expertise in the first place?   By studying?     

"Why being anonymous? This makes no sense to me. If you are in academia, you want your peers to know what you are working on. Also, scientists are not looking for money for themselves. That is a wrong assumption from the start. They are however looking for funding to hire other people to help them on their research."

Why be anonymous?   Well to prevent situations like what's going on in U.S. right now or what happened in Canada just a few years ago under Harper.    Governments making it illegal for scientists to publish their data that didn't meet with government approval.    And/or governments making it illegal for scientists to criticize government policies, especially environmental policies.   Why are the scientists in the U.S. that work for government taking their names off the papers they want published?   Because if they don't the papers can't be published.   That's why.    And it's more important to the scientists to get the data out there than it is to have their name on the research paper and receive reputation.     That's also why there are alt scientific twitter accounts being made and other such things so even if the scientists can't officially release the data and get paid and acknowledged at least they can release the data.

It doesn't matter if scientists want the money for themselves or to further their research.   That's irrelevant.    The fact is they need money to continue their research.   And that dependency on money means that if they produce the wrong facts from their research they can get their funding cut.    

"This is again a very dangerous generalization. Open access data exists. Moreover, while I agree that data should be made available to the public after sometime, this should not be done blindly. Scientists may want to use their data and analyse it themselves first. They after all built the experiments and get this by their work. Then, after some time, they should however release the data (for reproducibility, etc...)."

And why can't those scientists use their data while other people also use their data?   It's not like they can't keep a copy of it.    As for analysis more eyes make for better analysis than just two.    The more people analyzing the data the faster it happens.    The only reason you'd want to keep your data safe is to produce a product and then the question needs to be asked: What's in the product and how is it made?

All in all I think you're far too trusting of institutions.