Ciro Santilli @cirosantilli 35

They have a lot of stuff, well done.

CC BY-NC-SA by default unfortunately, but what can you do...

With CC BY-SA of course. Not the evil CC BY-NC-SA.

Public university and also universities with charity status, e.g. since Section "Most British universities are registered as charities".

Any lectures must be recorded. But of course, text is cheaper than video.

Just fucking pay teachers a decent salary comparable to industry instead of basically requiring them to try and sell their knowledge to complement their income. They must get paid, and their knowledge must

We don't need that many teachers teaching the same subjects over and over. We just need a few good men and women who can truly teach the masses. And we must pay those men and women a decent wage.

The perfect platform for this will of course be OurBigBook.com ;-)

Bibliography:

Follows the "certified teacher only" approach which is in Ciro Santilli's opinion a fatal flaw of most elearning systems out there, OurBigBook.com won't suffer from that!

But that is a very, very good project.

All notes appear to have been extracted from existing notes, as noted on the bottom of each page.

Appears to have mixed licenses. E.g.:

TODO how does it work exactly? Do they ask for permission from authors in every case, including when the content has open license? Or when it has open license, do they just do it? In some cases, the notes have no license, so they must have asked.

TODO what is the source code that authors write? LaTeX or something else? LaTeX feels extremely likely given that it is what most original materials were already written in.

They are attempting a "model up this entire university" thing: phys.libretexts.org/Courses which is good. E.g. they have a bunch of "quantum mechanics ones under: phys.libretexts.org/Bookshelves/Quantum_Mechanics

Appears to be UC Davies-based mostly.

They claim to use this closed source backend: www.nice.com/resources/cxone-expert-knowledge-management? Seriously? For a publicly funded project with low-tech requirements?? It is mind blowing.

Some issues:

OK let's database it:

CC BY-NC-SA unfortunately.

Upload is actually optional, and it appears that teachers do retain their copyright: ocw.mit.edu/ans7870/global/MIT_OpenCourseWare_FAQs.pdf Hmmm, so how have they convinced so many teachers to do it?

Some possible/not possible sources that could be used to manually bootstrap content:

Existing lecture notes by students:

Lecture note upload website:

Exams uploads:

This is an interesting initiative which has some similarities to Ciro Santilli's OurBigBook project.

The fatal flaw of the initiative in Ciro Santilli's opinion is the lack of user-generated content. We will never get there without UGC and algorithms, never.

Also as of 2021, it mostly useless business courses: learn.saylor.org unfortunately.

But it has several redeeming factors which Ciro Santilli aproves of:

Licensing appears to be a mixed mess between the dreaded CC BY-NC-SA and the good CC BY, e.g.:?

The founder Michael J. Saylor looks a bit crooked, Rich people who create charitable prizes are often crooked comes to mind. But maybe he's just weird.

As en.wikipedia.org/w/index.php?title=ZX-calculus&oldid=1071329204#Diagram_rewriting tries to explain but fails to deliver as usual consider the GHZ state represented as a quantum circuit.

How can we easily prove that that quantum circuit equals the state:?

The naive way would be to just do the matrix multiplication as explained at Section "Quantum computing is just matrix multiplication".

However, ZX-calculus provides a simpler way.

And even more importantly, sometimes it is the only way, because in a real circuit, we would not be able to do the matrix multiplication

What we do in ZX-calculus is we first transform the original quantum circuit into a ZX graph.

This is always possible, because we can describe how to do the conversion simply for any of the Clifford plus T gates, which is a set of universal quantum gates.

Then, after we do this transformation, we can start applying further transformations that simplify the circuit.

It has already been proven that there is no efficient algorithm for this (TODO source, someone said P-sharp complete best case)

But it has been proven in 2017 that any possible equivalence between quantum circuits can be reached by modifying ZX-calculus circuits.

There are only 7 transformation rules that we need, and all others can be derived from those, universality.

So, we can apply those rules to do the transformation shown in Wikipedia:

and one of those rules finally tells us that that last graph means our desired state:because it is a Z spider with $m=3$ and $n=1$.

Bibliography: