Researcher Behavior Access Controls at a Library Proxy Server are Not Okay

After the reaction to a presentation of an October Scholarly Networks Security Initiative (SNSI) webinar , I learned that Peter Murray tells us that User Behavior Access Controls at a Library Proxy Server are Okay.

They are not OK.

I expressed my opinion about this last invention in my reaction Academics band together with publishers because access to research is a cybercrime. There I argued that academic managers are to be blamed even more than publishers, because they support the publishers in the detriment of the researchers. But this is not all.

Here I want to add that there is a trend, started at least with the inventions of green and gold OA, which were so beneficial to the publishers. Decades of advances were lost because the fight ignored researchers needs.

What do they need? Something arXiv like with a Sci-hub like interface.

What do researchers got? They have to pay $2000 to publish an article they wrote [added: or even $10000 for an article in Nature]. And moreover, they are suspect of cybercrime. But be calm, because user behaviour access controls at a library proxy server are okay.

The trend is to make deals over the heads of researchers.

Publishers with managers, publishers with librarians, IT department with librarians, and so on. If you look at how BOAI started, the initiator of the gold (in the pockets of publishers) open access style, it was librarians with publishers.

We, researchers, understood that librarians were scared by publishers that their important role will decay. We understood that managers want to apply to us the criteria which were designed for journals.

But it is time to understand that researchers have to be at the core of any deal, because without researchers there is no need for librarians, university IT administrators, managers or scientific publishers.

To make deals over the researchers heads is not Okay.

To be clear, please at least return the respect you received from the researchers. Please stop treating researchers as users which have to be herded to the publishers needs. This is not your job.

There is no pure commutative structure except the trivial one

I have a sketch of a proof that there is no other emergent algebra than the trivial commutative one, among those who satisfy the COLIN relation (a sort of right-self-distributivity), see these notes for a formulation of the problem and this post here at chorasimilarity.

Recall that LIN relation implies that we have a structure of conical group, which is one of a non-commutative vector space. The more powerful relation SHUFFLE gives a structure of a commutative vector space. The same result can be achieved by using instead the pair of LIN and COLIN relations, which are therefore equivalent with SHUF.

Geometrically, deviation from LIN is a curvature and deviation from COLIN is the bracket in the non-commutative tangent space. So they look like two independent phenomena.

It is then natural to ask if there are pure commutative structures, i.e. if there are COLIN but not LIN structures.

The answer seems to be “no”. The proof seems to indicate that the reason for this is that COLIN induces more constraints that LIN. Indeed, even if they are symmetric, in the context where we also have the R2 for emergent algebras, COLIN implies a symmetric of R2 and this is what is needed to conclude the proof. Looking in the mirror, by the same proof LIN brings R2, which is already available, and that’s why COLIN is stronger than LIN.

I still might be wrong somewhere in the proof, maybe there still are some exotic pure commutative examples.

Freedom of Speech and Section 230

Section 230 of the Communication Decency Act, 47 U.S. Code § 230, is “one of the most valuable tools for protecting freedom of expression and innovation on the Internet”, as the Electronic Frontier Foundation writes.

The text of Section 230 which is of interest is the following:

“(c) Protection for “Good Samaritan” blocking and screening of offensive material

(1) Treatment of publisher or speaker

No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.

(2) Civil liability

No provider or user of an interactive computer service shall be held liable on account of—

(A) any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable, whether or not such material is constitutionally protected;

or

(B) any action taken to enable or make available to information content providers or others the technical means to restrict access to material described in paragraph (1)”

The Electronic Frontier Foundation text explains that:

“The Internet community as a whole objected strongly to the Communications Decency Act, and with EFF’s help, the anti-free speech provisions were struck down by the Supreme Court. But thankfully, CDA 230 remains and in the years since has far outshone the rest of the law.”

[…]

“This legal and policy framework has allowed for YouTube and Vimeo users to upload their own videos, Amazon and Yelp to offer countless user reviews, craigslist to host classified ads, and Facebook and Twitter to offer social networking to hundreds of millions of Internet users. Given the sheer size of user-generated websites […] Rather than face potential liability for their users’ actions, most would likely not host any user content at all or would need to protect themselves by being actively engaged in censoring what we say, what we see, and what we do online.”

The Universal Declaration of Human Rights, adopted in 1948, formulates in the Article 19 the freedom of opinion and expression:

“Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.”

These rights (condesated name: “free speech”) are in many countries constitutions.

In US there is the First Amendment to the United States Constitution:

“Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.”

The right of free speech is, in US, protected against the government. The Article 19 of the Universal Declaration of Human Rights has a more powerful form, not being restricted to protect against goverments.

But governments are no longer the most powerful force, the one we need protection for our free speech.

Corporations are! more specifically, any private entity who was formed and grown by the promise of a better public forum.

Section 230, as is, does not protect free speech as defined in the article 19. It does not protect free speech more than the First Amendment, who protects only against goverment. It only protects these rights in an indirect way, namely that without Section 230 corporations “would likely not host any user content at all or would need to protect themselves by being actively engaged in censoring what we say, what we see, and what we do online”, as EFF explains. So even more, when corporations are actually actively engaged in censoring, with the best intentions even, Section 230 does not protect free speech.

What can be done? The simple thing to add to the wonderful and very useful Section 230 would be a form of article 19 or an extension of the First Amendment text, with the effect that:

Any provider of user created content is under the same constraints concerning free speech as the state is by the First Amendment.

In this way, Section 230 will turn from an unfair advantage of monopolies who control our world view, into an explicit protector of freedom of expression and innovation on the Internet.

Plan U, almost 30 years too late, what we need instead

Plan U [site] [article] says that

“Arguably the most effective mechanism for providing free, immediate access to research has been the non-profit preprint server arXiv.”

“If all research funders required their grantees to post their manuscripts first on preprint servers — an approach we refer to as “Plan U” — the widespread desire to provide immediate free access to the world’s scientific output would be achieved with minimal effort and expense.”

Looks like a great idea, only that it is too late. By almost 30 years. A generation of researchers careers were lost by this delay. How was this possible? After all this fight for open access?

The fight was fake. BOAI made arXiv “green OA”, i.e. an archiving place. What a damage.

OK, it seems that now, with Plan U, we finally arrive where we should be. The most powerful point is, in my opinion, that

“because it sidesteps the complexities and uncertainties of attempting to manipulate the economics of a $10B/year industry, Plan U could literally be mandated by funders tomorrow with minimal expense, achieving immediate free access to research and the significant benefits to the academic community and public this entails.”

What could be achieved with an extra $10B/year for research? Right now this is stolen from reseach and the academic managers help the publishers to achieve that goal. Why?

The problem with Plan U is that in practice it is already achieved. Most of new research is available online, free and fast. Most of published research is available via Sci-Hub. I don’t know how Sci-Hub does this, the publishers claim that this is cybercrime, but the outcome, i.e. the easy availability of all published research, can’t be in itself a crime.

So what would be a Plan which is realistical and not another generation loss?

If you don’t like Sci-Hub, make one. It is technically possible. If academics don’t know how to do it then they should ask the professionals.

It should be redundant, scientific results accessible as simple as possible: by DOI or other identifier. It should be resilient in time. It should be decentralized (big problem), because one cannot trust big companies, like Google, who could do it fast and well, to treat respectfully scientific data.

Couple it with Open Science and make sure that we don’t end with another BOAI like fake solution: be sure that a scientific bit is just as expensive to host as any other bit.

All this is possible and fast to achieve, only that the ones who have decision power do not want that. But suppose they will find time for this.

There is a problem with decentralization. Besides redundancy, there is another reason for this. Just imagine that tomorrow academic managers discover the internet and they declare that arXiv.org is publication (or some variant of Plan U). Then every researcher in the world who hunts for promotion points will send a deluge of low quality minimal unit of research publications there. In a year arXiv.org will decay.

Instead, there should instead be many instances of arXiv.org (coupled with Open Science), which interact by scientific arguments one with the other.

Well, dreams. Decentralization is frowned upon. We have algo-human single point of truth now.

The single point of truth is a necessary thing for a centralized mindframe. Probably this is what expect us, researchers, in the next 30 years.

Unless good hackers give us fast, usable, no nonsense technical solutions.

Wanted: Left quasigroups which are right distributive but not left distributive

Excuse me, I am left-handed, which makes me prone to confusions about what is left- and what is right-. Here is the link to the question on mathoverflow.

UPDATE: I reformulated the question and put a description of the problem in these notes. See there how left-distributivity is related to curvature and right-distributivity is related to commutators.

UPDATE 2: It seems that these objects are exceedingly rare in the literature. I can’t locate any such example, there are hints that they exist, but I am not sure if I look at one, perhaps because I am not familiar with the language in the field of quasigroups.

UPDATE 3: Solved, there are none! arXiv:2110.08178 and some ore comments here.

______

______

I look for as many examples as possible, my preference is for infinite examples, of idempotent quasigroups which have this peculiar algebraic structure:

a set X with an operation denoted by a dot “.”, such that

(idempotent) x . x = x

(left quasigroup) the equation a . x = y has a unique solution denoted by x = a * y

(right distributive) (x ? y) ! z = (x ! z) ? (y ! z) for any choice of the operations ? and ! among the operations . and *

but not

(left distributive) x ? (y ! z) = (x ? y) ! (x ? z) for any choice of the operations ? and ! among the operations . and *

Remark that idempotent left quasigroups which are left distributive are quandles. What are idempotent left quasigroups which are right distributive but not quandles?

Academics band together with publishers because access to research is a cybercrime

This is the world we live in. That is what I understand from reading about the Scholarly Networks Security Initiative. and it’s now famous webinar, via Bjorn Brembs october post.

I just found this, after the post I wrote yesterday. I had no idea about this collaboration between publishers and academics to put spyware on academic networks for the benefit of publishers.

UPDATE: see also Researcher Behavior Access Controls at a Library Proxy Server are not Okay.

What I find worrying is not that publishers, like Elsevier, Springer Nature or Cambridge University Press, want to protect their business against the Sci-hub threat. This is natural behaviour from a commercial point of view. These businesses (not sure about CUP) see their activity atacked, so they fight back to keep their profit up.

The problem is with the academics. Why do they help the publishers? For whose benefit?

I wrote again and again in the past that it is not enough to criticize the publishers for the bad bahaviour. Academic managers are to be blamed because they band with the publishers. Why does nobody asks them why?

Take Elsevier as example. I signed the Cost of Knowledge and I try as much as I can to not fold to the pressure of legacy publishers. But Elsevier does not have any direct way to force anybody to use their products.

At the end of the day, is the academic managers who pressure the researchers in the favor of publishers, is them who manage the libraries who pay big bucks from public funds to the publishers. Now we see that in universities is proposed to introduce spyware in the favor of publishers. Whose fault is this?

Now we get to the gist: it is cybercrime. It is indeed, because research is valuable for states, who fund it and benefit from it. When a piece of US research, say, is made available by a Russian site, say Sci-hub, then the advantage of one state in that research direction is lost in the favor of all the other states. So everybody can copy and produce derivative works.

Science should be free. Do you remember that disclosure of mathematical knowledge was considered cybercrime?

The gist is not really that the threat to the publisher’s business is now branded as cybercrime. The fact is that publishers and universities are naturally together as a part of the state. Powerful states need powerful propaganda and publishers together with universities are an important part of this. When they are in trouble it is only natural that they resort to another state pillar and find together a powerful name: cybercrime.

Every state does that. Does it mean that states are against science? It is also propaganda to say that a state is against science and another state support generous efforts to make science free. It’s so complicated, but it’s all propaganda.

It is propaganda which harms every state! It is the most stupid way to proceed today. Because today is very different from yesterday. If we make the research free to access then we create an evironment where better research appears, and faster. Scarcity is a very bad idea. We all know it.

So I don’t buy that publishers and academic managers banded together to fight cybercrime. I believe academics will produce better results without the artificial scarcity created by legacy publishers.

At best, some state bureaucrats proved they are ready to harm their states, by ignorance. At worst, academic managers have non-declared interest to keep the legacy publisher alive.

Sci-hub provides an easy to use and necessary product to researchers from the world, who want to do their research. This is the strength of it. Now that people saw that it is technically possible, the rest is spin.

A site of the Sci-hub creator Alexandra Elbakyan

I found this site of Elbakyan, (not only) in my opinion the creator of one of the greatest libraries of science ever, Sci-hub.

Here is the site:

https://sci-hub.do/alexandra

and if it will stop then here is the archived version.

UPDATE: now is https://sci-hub.ru/alexandra

I wrote several times about Sci-hub or about Elbakyan here, last time in this post. I found her new site by looking for a replacement of the links which are no longer valid in that post.

Sci-hub is a technical solution (and a solution driven by the contributions from many people) for a problem which should not exist. You may hate Sci-hub, you may think that in some places is not legal, but fact is that it is a very easy to use site, which is indeed used by a huge number of people. They are not using it because they are poor, rich or ideologically motivated. They are using the site because it works, responding to a problem they have. I suspect almost all of them are researchers or students, because of the content of the library available.

Ideologically, the existence of this solution shows that the Open Access is at best a misleading fight. Indeed, who cares if it can be read for free from the publishers (some of them so cynical that they take their money from the article creators), or the article is hidden behind a paywall. The solution makes the article available anyway.

This solution makes obvious why a hacker is better than an army of nerds. With a negligible percentage of funding, compared with Google Books, the hacker made a huge gift to researchers. In the same time, Google sits with its fat nerd ass over millions of books, not making them available to anybody, after they spent a fortune to scan them.

Yes, there is the problem with the copyright. What a shame is it when it comes to scientific works! Because scientific knowledge occupies such a small number of bits, compared with the metadata exhaust which is collected and processed, but the commercial interest leads to keeping it walled.

Open Science, in the pure sense of content which is scientific, i.e. it can be validated independently, is not touched by this solution. It does not have to be, Elbakyan did enough. I noticed several articles in her site, about OS, written in Russian, I shall read (a brute English translation of) them.

It is though remarkable that despite all the very well paid Open Access fight, which birthed the greed monster called Gold OA, the true OA was solved by one, or a few, individuals. Likewise Open Science can be practiced by anybody, individually, without waiting for the whole society of researchers to reform. It is enough that you, researcher, put online as much as possible of your work, so that another researcher, if curious, could reproduce your findings (and contradict your interpretations/opinions).

Pure See expressivity and linear variables

Pure See [draft] in it’s present form tries to be both a graph rewrite system and a term rewrite system. I arrived to split the semantic part from the syntactic part by the introduction of linear variables. (The draft will be soon updated).

As a programming language, I aim Pure See to be expressive for a human. At least that any construct to admit a reading which makes sense for a human.

Here are two examples related to lambda calculus. Recall that we have, in lambda calculus, two operations:

• application, denoted by AB, defined for any two terms A and B
• abstration, denoted by λ x. A, defined for any variable x and term A

In Pure See we can write the definition of application as:

as A from B see AB;

or

apply A over B as AB;

Similarly, In Pure See the definition of abstraction is:

see A from x as λ x. A;

Therefore the term (λ x. A)B appears to be

as (see A from x) from B see (λ x. A)B;

which in lambda calculus reduces to A[x=B] and in Pure See reduces to

A as (λ x. A)B;

B as x;

Chemlambda animated book

I decided it is time to make it. As the name tells, it is animated, meaning that it will be full of animations and of simulations. This is why it cannot be made into an e-book, not with the present technology.

But we may try, right?

For those who followed chemlambda through the years, it will be a self-sufficient place to enjoy the animations, play with the simulations and dream on. Thus a part of it will be based on the 264 animations with comments, most of them with simulations, of the chemlambda collection.

I intend to use though the original animations, currently available only from my professional page. This is because the animations alone are 1GB, so I can’t put them on Github.

For those who desire to know what is the relation between chemlambda and a lambda calculus graphic reducer, there will be a reducer of lambda terms, which is also available here, Notice that you may reduce lambda terms not only with chemlambda, but also with dirIC, a version of directed interaction combinators.

The story of the ouroboros (which is proved that is NOT immortal) makes the transition from a limited but interesting lambda calculus perspective to chemlambda quines.

You will be able to play with quines and discover more of them with a combination of the quine lab and the page on how to test a quine.

There will be an introduction concerning the history of chemlambda, what it is and how is related to other graph rewriting formalism. I shall base on the history reference page, but I shall rather concentrate on chemlambda and dirIC, perhaps also on Interaction Combinators.

For the algorithms used for graph rewriting and why are they interesting for models of real chemistry, there will be a more extended version section 6.5 “Local machines”, arXiv:2003.14332 [cs.AI] . Perhaps combined with the Molecular computers with pdf version on (figshare) or arXiv:1811.04960 [cs.ET] article.

The introduction will adapt the Chemlambda for the people presentation and I will consider if I add the two chemical sneakernet stories (internet of smells) (archived) and   (Home remodeling) (archived).

All in all, it will make into a ~ 350 pages of text with animations (264) and simulations (~ 200).

Depending where I share it, if it will be put on Zenodo then I may add many hours of raw movies from my personal collection. The simulations used in the animations (i.e. the old ones, not the new, js only ones) are already archived in the Chemlambda collection of simulations (1GB) here:

Buliga, Marius (2017): The Chemlambda collection of simulations. figshare. Dataset. https://doi.org/10.6084/m9.figshare.4747390.v1

__________

As you see, in this projected book there will be the part of chemlambda which appeared in relation with artificial life and molecular computers.

There will be nothing in the book about the unlikely source of this, namely the computing with space project. There will not be anything about pure see, or about medial emergent algebras, or about emergent graph rewrites. Not even about Zip Slip Smash. This part is ongoing research, very exciting to pursue, independently of the possible applications of the computing with space in chemistry.

Maybe I will mention chemski though, because it is in the same vein as chemlambda.

Where we are now with the computing with space project (nov 2020)

I take this information apart from a previous post, to show where we are now with the computing with space project. Please read that post because it contains more than what I extract and slightly edit here.

While I try to increase the readability, it becomes clear to me that even if most of the composing pieces are available online it is rather hard for the non-specialist to compile a up to date unitary version. I shall write such a version soon, moreover with parts which are still on paper.

So here is the global image, as it is available today.

All in all we have the following formalisms, in a decreasing order of generality:

• the formalism of emergent algebras, or dilation structures, which can be turned into a graph rewriting formalism over decorated graphs (with nodes and edges which are decorated) and with graph rewrites which take into account the decorations. Various attempts for this graph rewriting formalism are: Braided spaces with dilations [1] section 4, the Computing with space [2] sections 3-6, lambda-scale calculus [3], em-convex rewrite system [4], with a satisfactory final version to appear.
• less general is the formalism of linear emergent algebras, which can be used to (or are compatible with) a pure graph rewriting formalism over (non-planar) diagrams of tangles, because (LIN) in this case is the rewrite (R3). It is not known if this graph rewriting formalism is Turing complete though, or more precisely if there is some natural correspondence with Interaction Combinators, say. The purpose of the Zip Slip Smash formalism is to provide this correspondence, where we add to the Reidemeister rewrites some rewiring rewrites in order to achieve this.
• even less general is the formalism of medial emergent algebras, which turns out to be capable to serve as a semantics for dirIC, thus for the Interaction Combinators.

Therefore, the image is, in increasing order of generality:

multiplicative linear logic (SHUF) < graph rewriting on decorated knot diagrams (LIN) < differential calculus in emergent algebras

The interest would be to understand the implications of these inclusions, but first to internalize that linear logic is, as I said repeatedly, a commutative version of a more powerful formalism. This seems hard to believe, but it is indeed so. A price to pay for passing to a more powerful level is to renounce at the false generality of cartesian etc categories, because this is a generalization from a too particular example. From here we may look to provide versions of linear logic which are not commutative, but they are still LINear, therefore at the level of (LIN), not only (SHUF). Another effort would be to decrease the differential calculus from the most general level provided by dilation structures to the level of (LIN), not only (SHUF) (where it is the usual one known to anybody), so that we can compare it on common ground with the linear logic, in a natural, unforced, less naive way.

There are many other directions to explore, some of them in a more evolved state than others, but with patience we can do great things.

Election result

I was not right with my Election prediction (with proof). I hope that I was right with this other more important prediction 🙂

This post is just a mark in time, that at a certain moment I believed a certain thing.

EDIT 5 (Dec 12th): Let me be clear about why those elections were mentioned here: because they mark an important time in history where a corporation was stronger than the strongest state. It is not unseen, the previous example I am aware about is the East India Company. You may tell me that it is an obvious result and I’ll agree that the underdog does not deserve much attention in this real movie. This is not my country, instead this is my internet, affected by these elections, again.

EDIT 4 (Nov 20th): After the Most Favored Nation rule, maybe something about Section 230?

EDIT 3 (Nov 12th): After legacy media and the media protected by the section 230 weighted on it, the scale didn’t move in the expected direction. The problem of ruling by making the map is that when the map is very far from reality, for a very long time, it becomes useless.

EDIT 2 (Nov 5th): It looks that I guessed wrong. My question is then: who will repeal a section 230 😉 ?

EDIT: Today Nov 4th, searching for informations…

ha, ha, ha, who’s the boss? who makes the map.