The good, the bad and the iawful: OA, measures of scientific output and bad legislation

What happens in the real world, the one of the  powers that be, as concerns open access, peer-review and communication of research results? Let’s see.

The good:  San Francisco Declaration on Research Assessment (DORA).

Funding agencies, institutions that employ scientists, and scientists themselves, all have a desire, and need, to assess the quality and impact of scientific outputs. It is thus imperative that scientific output is measured accurately and evaluated wisely. […]

A number of themes run through these recommendations:

  • the need to eliminate the use of journal-based metrics, such as Journal Impact Factors, in funding, appointment, and promotion considerations;
  • the need to assess research on its own merits rather than on the basis of the journal in which the research is published; and
  • the need to capitalize on the opportunities provided by online publication (such as relaxing unnecessary limits on the number of words, figures, and references in articles, and exploring new indicators of significance and impact).

Read it. Disseminate it. Sign it.

________________

The bad:  The apparatus of research assessment is driven by the academic publishing industry and has become entirely self-serving. In this article by  Peter Coles you find:

The involvement of  a company like Elsevier in this system just demonstrates the extent to which the machinery of research assessment is driven by the academic publishing industry. The REF is now pretty much the only reason why we have to use traditional journals. It would be better for research, better for public accountability and better economically if we all published our research free of charge in open archives. It wouldn’t be good for academic publishing houses, however, so they’re naturally very keen to keep things just the way they are. The saddest thing is that we’re all so cowed by the system that we see no alternative but to participate in this scam.

________________

The iawful:  From this g+ post by Peter Suber we find out that:

Elsevier, NewsCorp, Facebook, and Yahoo are some of the major players in NetChoice, an industry group “promoting convenience, choice, and commerce on the net.”
http://www.netchoice.org/about/

NetChoice has a watch list for bad legislation that it calls iAWFUL (Internet Advocates’ Watchlist for Ugly Laws). The latest version of iAWFUL includes the White House OA directive plus the state-level OA bills in California, Illinois, and North Dakota. (Yes, there was a bill in ND, and no, NetChoice doesn’t seem to know about the OA bill in NY.)
http://www.netchoice.org/2013-may-iawful/

Insofar as NetChoice has an argument for opposing these OA initiatives, it’s a crude bolus of false assertions and assumptions. I haven’t seen this kind of motivated distortion since the days of PRISM and the Research Works Act.
http://www.netchoice.org/2013-may-iawful/4-forcing-journals-to-make-their-works-publicly-available/

UPDATE (23.05.2013):Elsevier distances itself from open-access article

The publisher Elsevier has disassociated itself from an article by a trade association it belongs to that condemns proposed open-access mandates in several US states.

________________

So, things happen … eventually. But slowly. I bet many of us, not entangled with the high politics or management in academia, wish for a faster pace.  For my part, I would rather play the Game. It has a very low Coase cost, you know?

________________

UPDATE:  if you still wonder  about Gold OA, is it good? is it bad?, here is a tweet about Elsevier and iawful:

Sets, lists and order of moves in graphic lambda calculus

Suppose that we want to group together three arrows in graphic lambda calculus. We have this:

sets_1

We want to group them together such that later, by performing graphic beta moves, the first arrow available to be 11′, then 22′, then 33′. Moreover, we want to group the arrows such that we don’t have to make choices concerning the order of the graphic beta moves, i.e. such that there is only one way to unpack the arrows. The solution is to “pack” the arrows into a variant of a list. Lists have been defined here, in relation to currying.

sets_2

Basically we take a zipper and we close it.  Further we see how to unpack this list.

sets_3

The dashed red curve encircles the only place where we can use a graphic beta move. The first move frees the 11′ arrow and then there is only one place where we can do a graphic beta move, which frees the 22′ arrow and finally a last move frees the 33′ arrow and produces a loop which can be eliminated.

The uniqueness of the order of moves is true, in fact, if we accept as valid beta moves only those from left to right (i.e. those which eliminate gates). Otherwise we can go back and forth with a beta move as long as we want.

There is another way to pack the three arrows, under the form of another graph, which could aptly be called a “set”. This time we need a graph with the property that we can extract any arrow we want from it, by one graphic beta move. Here is the solution:

sets_5

Indeed, in the next figure we see that we have three places, one for each arrow, which can be independently used for extraction of the arrow of choice.

set_4

In between these extremes, there are other possibilities.  In the next figure is a graph which packs the three arrows such that: there are three places where a graphic beta move can be performed, as in the case of the set graph, but once a beta move is performed, the symmetry is broken. The performed beta move does not free any arrow, but now we have the choice between the other two possible beta moves. Any such choice frees only one arrow, and the last possible beta move frees the remaining two arrows simultaneously.

Here is the figure:

sets_6

The graph from the left hand side is not a list, nor a set, although it is as symmetric as a set graph.  There are 3 \cdot 2 = 6 possible ways to unpack the graph. So this graph encodes all lists of two arrows out of the three arrows.

We, researchers, just need a medium for social interaction, and some apps

… so that we can freely play the game of research. Because is a game, i.e. it is driven by curiosity, desire to learn, does not depend on goals and tasks, it is an extension of a child attitude, lost by the majority of adults. Let the vanity aside and just play and interact with other researchers, on equal foot. Let the creativity manifest freely.

Two    Three  Four examples:

Rap Genius is a very well-loved and well-used online tool for annotating rap songs.  Only, not so surprisingly, people are starting to use it to annotate other things.  Like scientific papers.

  • Olivier Charbonneau writes

    Actually, that’s an interesting take on mass data visualization – imagine creating an algorithm that could parse a dataset of bibliographic information into minecraft (for example) – what would that research “world” look like?

  • Hermann Hesse’s   Das Glasperlenspiel (aka Magister Ludi)
  • Timothy Gowers, some time ago, in this post, writes:

What I think could work is something like a cross between the arXiv, a social networking site, Amazon book reviews, and Mathoverflow.

 

 

 

_________

Context:

Graphic lambda calculus used for quantum programming (Towards qubits III)

I want to make a bit more clear one of the goals of the research on graphic lambda calculus, which are reported on this blog.  I stress that this is one of the goals and that this is live research,  in the making, explained here in order to attract, or invite others to join, or use this exploration for their purposes.

More precisely, further I present several justifications for two series of posts

which have as common goal the application of graphic lambda calculus to some form of quantum programming (probably some version of a quantum lambda calculus). I use the informative linked wiki page on quantum programming  for citing. Please click on the links to go where the real information is.

Efforts are underway to develop functional programming languages for quantum computing. Examples include Selinger’s QPL,[1] and the Haskell-like language QML by Altenkirch and Grattage.[2][3] Higher-order quantum programming languages, based on lambda calculus, have been proposed by van Tonder,[4] Selinger and Valiron [5] and by Arrighi and Dowek.[6]

Simon Gay’s Quantum Programming Languages Survey has more information on the current state of research and a comprehensive bibliography of resources.

I hope that in some finite time I can prove that there is a “quantum lambda calculus” sector in graphic lambda calculus. Let me explain why.

Basically, leaving much detail aside, quantum computation needs a  mix of at least two ingredients:

  • some algebraic structure, which contains objects like complex vector spaces, real projective spaces, unitary transformations, projections, etc,
  • some logical structure overarching the algebraic one (purists may say that in principle a lambda calculus would do).

The algebraic structure is not needed entirely, i.e. the needed part is the web of relations between the various algebraic operations. For example, the vector space operations are needed and not the points of the vector space. Likewise, we need “linearity”, “unitarity” and not particular linear or unitary transformations. Enough is to know how linearity and unitarity interact with the algebraic operations.

In the same way, as concerns the logic part, we need (say, if we are interested in a quantum lambda calculus) an abstraction an an application operations (like in lambda calculus) which interact well with the algebraic structure. Right?

There is one more ingredient needed: some form of evaluation procedure. There we can see a difference between a quantum and a classical lambda calculus. A quantum lambda calculus is more geometrical, less commutative than a classical one. One has to take care of phases, of the order of evaluations more than in the classical one.

Graphic lambda calculus seems to be a welcoming host for all these demands. Indeed, let’s see.

Graphic lambda calculus encodes algebraic structures in the barest way, by using only one gate: the emergent algebra gate \bar{\varepsilon}, with the parameter \varepsilon in a commutative group. This \varepsilon models “scale”, it is usually taken in (0, \infty) or in \mathbb{Z}. However, phase is a kind of scale, i.e. the formalism works well with the choice of the commutative group of scales to be \mathbb{C}^{*}.   Any algebraic operation and any algebraic computation in complex vector spaces, or in real projective spaces, may be expressed into graphic lambda calculus by the intermediary of the emergent algebra gate. Moreover, even some of the differential calculus (needed but not mentioned previously) can be embedded into graphic lambda calculus, in a kind of constructive way. This is the “emergent algebra” point of view, introduced in arXiv:0907.1520 .

So, shortly said, in graphic lambda calculus we have the algebraic structure needed. It “emerges” from the \bar{\varepsilon} gate, when we take the scale parameter to be in \mathbb{C}^{*}. With the barycentric move BAR from Towards qubits part I   we get the algebraic structures of vector spaces (see  how to get projective spaces in   part II, work in progress). More interesting, without the barycentric move we get Carnot groups, i.e. non-commutative vector spaces.

Question 1. What we obtain if in the formalism of quantum mechanics we renounce at complex vector spaces and we replace with their non-commutative version, the Carnot groups?

(This is the motivation for the series of posts Gromov-Hausdorff distances and the Heisenberg group part 0, part I, part II, part III  in this blog.)

For the logic part, we know that graphic lambda calculus has a sector which corresponds to untyped lambda calculus. In quantum programming it would be interesting to find a quantum version of the lambda calculus which interacts well with the algebraic structure. But in graphic lambda calculus are allowed interactions between the lambda calculus gates,  (or logical gates) of abstraction and application, and the algebraic gates. We don’t need more, that is what I shall try to convince you eventually. Indeed, probably obscured behind the lambda scale calculus  (which is a first, non-graphical version of the graphic lambda calculus), this was already explored in section 4 “Relative scaled calculus” of arXiv:1205.0139, where we see that to any scale parameter \varepsilon is associated a relative lambda calculus. This was done in whole generality, but for the needs of a quantum lambda calculus  “linearity moves” like in the  “Towards geometric Plunnecke graphs” series could be applied selectively, i.e. only with respect to  the (0, \infty) part of \mathbb{C}^{*}, thus obtaining a relative lambda calculus which is phase-dependent.

Question 2.  What would a relative scaled lambda calculus look like in graphic lambda calculus?

Finally, for the evaluation procedures which are adapted to quantum world, in this respect, for the moment, I have only results which indicate how to get usual evaluation procedures in graphic lambda calculus by destroying it’s geometrical nature (that’s what I call the “cartesian disease“, if you care), which are explained in some detail in   Packing and unpacking arrows in graphic lambda calculus    and Packing arrows (II) and the strand networks sector.

Question 3.  Design evaluation procedures in graphic lambda calculus which are geometrical, in the sense that, at least when applied to the yet vague quantum lambda sector of the graphic lambda calculus, they give evaluation procedures which are useful for quantum programming.

So, that’s it, I hope it helps a bit the understanding. You are welcome to join, to contradict me or to contribute constructively!

I have a question about this idea of mixing games with peer-reviews

I don’t get it, therefore I ask, with the hope of your input. It looks that the Gamifying peer-review post has found some attentive ears, but the Game on the knowledge frontier not. It is very puzzling for me, because:

  • the game on the frontier seems feasible in the immediate future,
  • it has two ingredients – visual input instead of bonus points and peer-review as a “conquest” strategy – which have not been tried before and I consider them potentially very powerful,
  • the game on the frontier idea is more than a proposal for peer-review.

My question is: why is the game on the frontier idea less attractive?

Looking forward for your open comments. Suggestions for improvement of such ideas are also especially welcomed.

_______________

UPDATE:  Olivier Charbonneau writes:

Actually, that’s an interesting take on mass data visualization – imagine creating an algorithm that could parse a dataset of bibliographic information into minecraft (for example) – what would that research “world” look like?

 

What group is this? (Parallel transport in spaces with dilations, II)

I continue from Parallel transport in spaces with dilations, I.   Recall that we have a set X , which could be see as the complete directed graph X^{2}. By a construction using binary decorated trees, with leaves in X, we obtain first a set of finite trees FinT(X), then we put an equivalence relation \sim on this set, namely two finite trees A and B are close A \sim B if A \bullet B is a finite tree. The class of finite points PoinT(X) is formed by the equivalence classes [A] of finite trees A  with respect to the closeness relation \sim.

Notice that the equality relation is \leftrightarrow , in this world.  This equality relation is generated by the “oriented Reidemeister moves”  R1a and R2a, which appear also as moves in graphic lambda calculus. (By the way, this construction can be made in graphic lambda calculus, which has the moves R1a and R2a. In this way we obtain a higher level of abstraction, because in the process we eliminate the set X. Graphic lambda calculus does not need variables. More about this at a future time.) If you are not comfortable with this equality relation than you can just factorize with it and replace it by equality.

It is clear that to any “point” x \in X is associated a finite point [x] \in PoinT(X). Immediate questions jump into the mind:

  • (Q1)  Is the function x \in X \mapsto [x] \in PoinT(X) injective? Otherwise said, can you prove that if x \not = y then x \bullet y is not a finite tree?
  • (Q2)  What is the cardinality of PoinT(X)? Say, if X is finite is then  PoinT(X) infinite ?

Along with these questions, a third one is almost immediate. To any two finite trees A and B is associated the function [AB] : [B] \rightarrow [A]  defined by

[AB](C) = A \circ (B \bullet C) .

The function is well defined: for any C \in [B] we have B \bullet C \in FinT(X), by definition. Therefore [AB](C) \in [A], because A \bullet \left( [AB](C) \right) \leftrightarrow B \bullet C .

Consider now the groupoid ParaT(X) with the set of objects PoinT(X) and the set of arrows generated by the arrows [AB] from [B] to [A].  The third question is:

  • (Q3)  What is the isotropy group of a finite point [A]   (in particular [x] ) in this groupoid? Call this isotropy group IsoT(X) and remark that because the groupoid ParaT(X) is connected, it follows that the isotropy groupoid does not depend on the object (finite point), in particular is the same at any point x \in X (seen of course as [x] \in PoinT(X) ).

In a future post I shall explain the answers to these questions, which I think they are the following:

  • Q1:  yes.
  • Q2: infinite.
  • Q3: a kind of free nilpotent group.

But feel free to contradict me, or to propose solutions. Of course, I shall cite any valuable contribution, even if it appears in a blog  (via +Graham Steel).

MMORPGames at the knowledge frontier

I think we can use the social nature of the web in order to physically construct the knowledge boundary. (In 21st century “physical”  means into the web.)

Most interesting things happen at the boundary. Life on earth is concentrated at it’s surface, a thin boundary between the planet and the void. Most people live near a body of water. Researchers are citizens of the boundary between what is known and the unknown.  Contrary to the image of knowledge as the interior of a sphere, with an ever increasing interface (boundary) where active research is located, no, knowledge, old or new, is always on the boundary, evolving like life is, into deeply interconnected, fractal like niches.

All this for saying that we need an interesting boundary where we, researchers, can live, not impeded by physical or commercial constraints.  We need to build the knowledge boundary into the web, at least as much the real Earth was rebuilt into the google earth.

Game seems to be a way. Because game is both social and an instrument of exploration. We all love games, especially researchers. Despite the folklore describing nerds as socially inept, we were the first adopters of  Role Playing Games, later evolved into virtual worlds of the Massively Multiplayer Online Role Playing Games.  Why not make the knowledge frontier into  one of these virtual worlds?

It looks doable, we almost have all we need. Keywords of research areas could be the countries, places. The physics of this world is ruled by forces with articles citation lists as force-carrying bosons.  Once the physics is done, we could populate this world and play a game of conquest and exploration. A massively multiplayer online game.  Peer-reviews of articles decide which units of places are wild and which ones are tamed. Claim your land (by peer-reviewing articles), it’s up for grabs.  Organize yourselves by interacting with others, delegating peer-reviews for better management of your kingdoms, collaborating for the exploration of new lands.

Instead of getting bonus points, as mathoverflow works, grab some piece of virtual land that you can see! Cultivate it, by linking your articles to it or by peer-reviewing other articles. See the boundaries of your small or big kingdom. Maybe you would like to trespass them, to go into a near place? You are welcome as a trader. You can establish trade with other near kingdoms by throwing bridges between the land, i.e. by writing interdisciplinary articles, with keywords of both lands. Others will follow (or not) and they will populate the new boundary land you created.

After some time, you may be living in complex, multiply-connected kingdom cities, because you are doing peer-reviewed research in an established, rich in knowledge field. On the fringes of such rich kingdoms a strange variety of creatures live. Some are crackpots, living in the wild territory, which grows wilder with the passage of time.  Others are explorers, living between your respectable realm and wild, but evolving into tamer territory.   From time to time some explorer (or some crackpot, sometimes is not easy to tell one from another) makes a break and suddenly a bright avenue connects two far kingdoms. By the tectonic plate movement of this world, ruled by citations, these kingdoms are now one near the other.  Claim new land! Trade new bridges! During this process some previously rich, lively, kingdoms might become derelict. Few people pass by, but there’s nothing lost: like happened in Rome, the marble of ancient temples was used later for building cathedrals.

If you are not a professional researcher, nevermind,  you may  visit this world and contribute. Or understand more, by seeing how complex, how alive research is, how everything is interwoven. Because an image speaks a thousand words, you can really walk around and make an idea of your own about the subject you are curious about.

Thinking more about peer-reviews, which are like property documents, as in real life some are good and some are disputable.  Some are like spells: “I feel that the article is not compelling enough …”. Some are frivolous nonsense: “I find it off-putting when an author  does not use quotation marks as I am used to”. Some are rock-solid: “there’s a gap in the proof” or “I have not been able to find the error in the proof, but here is a counter-example to the author’s theorem 1.2”.

So, how can it be done? We (for example by a common effort at github) could start from what is available, like keywords and citations freely available or easy to harvest, from tools like google scholar profiles, mathscinet, you name it.  The physics has to be written, the project could be initially hosted for almost nothing, we could ask for sponsors. We could join efforts with established international organisms which intend to pursue somehow similar projects. The more difficult part will be the tuning of interactions, so that the game starts to have more and more adopters.

After that, as I said, the knowledge frontier will be up for grabs. Many will love it and some will hate it.

______________

Context: The richness of knowledge comes from this web of interactions between human minds, across time and space. This knowledge is not reserved to the statistically few people doing research. We grow with it, during school, we live within, no matter what we do as adults, we talk about and we are curious about it. Even more, immensely more after knowledge has been liberated by the web.

In a short lapse of time (at the scale of history) it has become obvious that research itself needs to be liberated from outdated habits. Imagine a researcher, before the web.  She was a dual creature: physically placed somewhere on the physical earth, living in some moment in time, but  mentally interconnected with other researchers all over the world, anytime in the history. However, the physical place of living impeded or helped the researcher to reach further in the knowledge world, depending on the availability of virtual connections: books, other physically near researchers, local traditions. We can’t even speculate about how many curious minds did not accessed the knowledge web, due to the physical place and moment in time where they lived, or due to society customs. How many women, for example?

But now we have the web, and we use it, as researchers. It is, in some sense, a physical structure which could support the virtual knowledge web. The www appeared in the research world, we are the first citizens of it.  The most surprising effect of the web was not to allow everybody to access the knowledge boundary. Instead, the most powerful effect was to enhance the access of everybody to everybody else. The web has become social. Much less the research world.

Due to old habits,  we loose the pace. We are still chained by physical demands. Being dual creatures, we have to support our physical living. For example, we are forced by outdated customs to accept the hiding behind paywalls of the results of our research.  The more younger we are, the more is the pressure to “sell” what we do, or to pervert the direction of our work in order to increase our chances of success in the physical world. In order to get access to physical means, like career advancements and grant money.

Old customs die hard. Some time ago a peasant’s child with a brilliant mind had to renounce learning because he needed to  help his family, his sister was seen as a burden, not even in principle considered for eventual higher education. Now young brilliant minds, bored or constrained by the local research overlord or local fashion, rather go doing something rewarding for their minds  outside academia, than slicing a tasteless salami into the atoms of publishable units, as their  masters (used to) advice them.