#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Options

# Azimuth initiatives

This thread is a place to brainstorm about existing and potential Azimuth initiatives.

To my mind, the most effective paths to growth will be those which are based on specific initiatives. At present, we have:

• Network theory research -- green mathematics.

• Azimuth Code Project -- scientific programming for environmental applications, or green scientific programming.

Each of these gives a specific focus, towards which we can hope to organize our activities, and create spaces for new people to contribute in the future. Each initiative is a potential engine for generating articles on our multi-author blog.

But there could be more!

For example,

• Development of curriculum on science that really matters, (1) for general audiences, (2) for kids.

• Including: things about probability and stochastics that every citizen should know.

All of this would be more than enough to keep us busy for a long time. But still, it is worthwhile to consider other possible initiatives, which could work their way up in the priority queue.

What else?

• Options
1.

Here's a suggestion: maybe we don't need more projects. Maybe we need more people to do the projects!

If so, one main goal of the Azimuth Project should be to increase the number of people who regularly put time into the projects we're doing. There are people who used to do this but don't anymore: are they just busy, or could something be improved that would make them more eager to contribute? There are also lots of people who could conceivably join the project. Why aren't they?

I suspect a lot has to do with the inherent difficulty of organizing volunteer projects, combined with my poor management skills.

Comment Source:Here's a suggestion: maybe we don't need more projects. Maybe we need more people to do the projects! If so, one main goal of the Azimuth Project should be to increase the number of people who regularly put time into the projects we're doing. There are people who _used_ to do this but don't anymore: are they just busy, or could something be improved that would make them more eager to contribute? There are also lots of people who could conceivably join the project. Why aren't they? I suspect a lot has to do with the inherent difficulty of organizing volunteer projects, combined with my poor management skills.
• Options
2.

However, getting a clearly-defined list of projects and a strategy for carrying them out might be a good way to attract more people! My approach has been "if you want to do it, just do it!" — but that may leave people wondering what to do.

Comment Source:However, getting a clearly-defined list of projects and a strategy for carrying them out might be a good way to attract more people! My approach has been "if you want to do it, just do it!" &mdash; but that may leave people wondering what to do.
• Options
3.

I think we need to be more specific rather than more general (though there's no harm in brainstorming). "Scientific programming for environmental applications" contains a huge number of possibilities. I don't have difficulty coming up with things that I would like to do given time and energy, but committing to one thing instead of wondering about a dozen things is harder. Right now, I'm interested in the Zero carbon Britain 2030's energy model, see the diagram of the network in Azimuth code challenges. But is anyone else enthusiastic? Will I be in a month's time? Even here, I suspect, the project is too vague so far.

Comment Source:I think we need to be more specific rather than more general (though there's no harm in brainstorming). "Scientific programming for environmental applications" contains a huge number of possibilities. I don't have difficulty coming up with things that I would like to do given time and energy, but committing to one thing instead of wondering about a dozen things is harder. Right now, I'm interested in the [[Zero carbon Britain 2030]]'s energy model, see the diagram of the network in [[Azimuth code challenges]]. But is anyone else enthusiastic? Will I be in a month's time? Even here, I suspect, the project is too vague so far.
• Options
4.

I agree, specificity would make it much easier for us to focus on a project. I would love to spend more time on this, but it's really hard to just jump into one of the many floating ideas without knowing if I'm just repeating something that's been done by someone else here already..

Comment Source:I agree, specificity would make it much easier for us to focus on a project. I would love to spend more time on this, but it's really hard to just jump into one of the many floating ideas without knowing if I'm just repeating something that's been done by someone else here already..
• Options
5.
edited June 2014

I pretty much agree with everything that's being said here, including the initial comments that I made above. I should spell out more what I meant to say.

I listed four general "initiatives" above, which I proposed as a framework for engaging further participation in the project, and I asked if there were any others that people had in mind. I'm not sure initiatives was the best word, but I what I was driving at was general causes that people will really care about, on a gut level. Purposes may be a better word. Each one of these can set a context for defining specific projects -- so what I wrote above was just a starting point for brainstorming.

Think about what it would really take to get new people to volunteer some of their precious time to Azimuth. I absolutely agree that we need to get more specific about what are projects are, and how people can concretely chip in, say if they only have 5 hours per week, but are not so motivated to help us figure out how to define our projects. That is what I am, albeit slowly, trying to help the Azimuth Code Project along with, by organizing the TODO list, etc. Though I've gotten slowed down by the infrastructural work, and tsuris at work.

But on a deeper level than whether our projects are vaguely or clearly specified, there is the question of why the work could matter at all to the human condition; the need to answer the skeptical but valid questions: so what? Times are hard and looking like they'll get harder, so there needs to be an "emotional purpose" in order for people to become active despite the temptations to despair.

To me, at least the first three "initiatives" I listed above give a good enough answer to the So What question. The goal of scientific programming for environmental applications, or "green scientific programming," I find to be uplifting, and my gut reaction is, Yes, what can I do to learn more about this, and to help out? That's where we then need to provide some good specifics, in order to engage people to make the step from inspiration to active contribution. Think of all the structure that is used to productively engage programmers in large professional teams, with requirements analysts, software engineers, and what have you. Although we don't need a full-blown "scrum" system, we definitely need more specifics in order to engage people. So, you see, I am agreeing with you guys about this.

I see Green Mathematics as another "purpose." Perhaps John you could think, over time, about some agenda points to structure this, that might give people some guidance about how they can contribute in smaller ways. I know this may be too tall an order, given the complexity and depth of the research -- but maybe it's not. We could put together reading lists, encourage people to talk about category theory in simpler contexts, suggest ways that people can experiment with the parameters for the model of a growing plant leaf,...

John you may well be right that we don't need to add any more "purposes" or projects to our agenda. Perhaps Green Mathematics and Green Scientific Programming are more than enough for now. Think of this thread, then, as a placeholder, in case people come up with other meaningful purposes down the line.

And, to repeat myself for the sake of clarity, I'm not saying that these general goals are enough to engage people, but rather that they provide meaningful contexts for defining specific projects.

Comment Source:I pretty much agree with everything that's being said here, including the initial comments that I made above. I should spell out more what I meant to say. I listed four general "initiatives" above, which I proposed as a framework for engaging further participation in the project, and I asked if there were any others that people had in mind. I'm not sure initiatives was the best word, but I what I was driving at was general causes that people will really care about, on a gut level. Purposes may be a better word. Each one of these can set a context for defining specific projects -- so what I wrote above was just a starting point for brainstorming. Think about what it would really take to get new people to volunteer some of their precious time to Azimuth. I absolutely agree that we need to get more specific about what are projects are, and how people can concretely chip in, say if they only have 5 hours per week, but are not so motivated to help us figure out how to define our projects. That is what I am, albeit slowly, trying to help the Azimuth Code Project along with, by organizing the TODO list, etc. Though I've gotten slowed down by the infrastructural work, and tsuris at work. But on a deeper level than whether our projects are vaguely or clearly specified, there is the question of why the work could matter at all to the human condition; the need to answer the skeptical but valid questions: so what? Times are hard and looking like they'll get harder, so there needs to be an "emotional purpose" in order for people to become active despite the temptations to despair. To me, at least the first three "initiatives" I listed above give a good enough answer to the So What question. The goal of scientific programming for environmental applications, or "green scientific programming," I find to be uplifting, and my gut reaction is, Yes, what can I do to learn more about this, and to help out? That's where we then need to provide some good specifics, in order to engage people to make the step from inspiration to active contribution. Think of all the structure that is used to productively engage programmers in large professional teams, with requirements analysts, software engineers, and what have you. Although we don't need a full-blown "scrum" system, we definitely need more specifics in order to engage people. So, you see, I am agreeing with you guys about this. I see Green Mathematics as another "purpose." Perhaps John you could think, over time, about some agenda points to structure this, that might give people some guidance about how they can contribute in smaller ways. I know this may be too tall an order, given the complexity and depth of the research -- but maybe it's not. We could put together reading lists, encourage people to talk about category theory in simpler contexts, suggest ways that people can experiment with the parameters for the model of a growing plant leaf,... John you may well be right that we don't need to add any more "purposes" or projects to our agenda. Perhaps Green Mathematics and Green Scientific Programming are more than enough for now. Think of this thread, then, as a placeholder, in case people come up with other meaningful purposes down the line. And, to repeat myself for the sake of clarity, I'm not saying that these general goals are enough to engage people, but rather that they provide meaningful contexts for defining specific projects.
• Options
6.
edited June 2014

In other words, we can proceed from the general to the specific.

The Azimuth Code Project is one possible "implementation" of green scientific programming. The stochastic resonance program is one implementation within the Azimuth Code Project.

We need to work on our list of coding challenges, and evolve them into "requirements specifications" that can be acted upon by programmers without a lot of science background. Hence the role of supporting blog articles. To fully enact this vision also calls for some kind of organization, however informal it may be, in which scientists are involved in the discussion -- feeding good problems to the programmers, and answering the questions that come up in the course of development. We already have elements of this, with John and Nathan here, but as a group effort it clearly has not been actuated to anything near its real potential. There was a good burst a while back, when Glynn, Alan, Tim and Staffan were actively here, alas. This is the direction that I intend to keep chipping away at.

Comment Source:In other words, we can proceed from the general to the specific. The Azimuth Code Project is one possible "implementation" of green scientific programming. The stochastic resonance program is one implementation within the Azimuth Code Project. We need to work on our list of coding challenges, and evolve them into "requirements specifications" that can be acted upon by programmers without a lot of science background. Hence the role of supporting blog articles. To fully enact this vision also calls for some kind of organization, however informal it may be, in which scientists are involved in the discussion -- feeding good problems to the programmers, and answering the questions that come up in the course of development. We already have elements of this, with John and Nathan here, but as a group effort it clearly has not been actuated to anything near its real potential. There was a good burst a while back, when Glynn, Alan, Tim and Staffan were actively here, alas. This is the direction that I intend to keep chipping away at.
• Options
7.
edited June 2014

A very interesting conversation! There are lots of things to say. Here's something about "green mathematics".

For me, the manageable portion of "green mathematics" is "network theory": understanding the mathematics governing the many networks that make up the world, from food webs to smart grids, and also the networks we use to understand the world, like Bayesian networks.

I'm working hard with my grad students to develop the foundations of network theory. I'm advertising this project whenever I can, and some academics - mainly mathematicians, physicists and highly theoretical computer scientists - are getting interested. I have a huge amount to say about it that I haven't had time to say yet.

So, I'm happy with how the "green mathematics / network theory" project is proceeding, except for two things:

1. It's very theoretical, and I'm having a bit of trouble spinning off "bite-sized sub-projects" that would be good for people who like to write programs. I bet such projects could be invented, and could be useful... but someone may have to nudge me in that direction, because I naturally focus on figuring out stuff just by thinking. So, for example, Jim Stuttard and David Tanzer and Ken Webb have done work on software for stochastic Petri nets: either reviewing the existing software, or writing blog articles about Petri net programming, or actually doing such programming... but so far there's been a relative lack of "published products", perhaps because I never jumped in and said yes, I need you to simulate this particular stochastic Petri net! Why not? I can imagine some questions about Petri nets that require a lot of experiments to answer, where simulations would be crucial. But since I don't program, I tend to focus on other questions... and none of you have pushed me to change this habit.

2. It's not very practical yet. It would be great to choose a specific practical application project to work on, a kind of test case to focus the research a bit and let people see what the theoretical work can do. I have an idea: using climate networks to study El Niño. This is something in the news, something that lots of people are excited about and arguing about. Programming would be really essential here! Graham proposed another example: Zero carbon Britain. I don't understand this well enough yet, but it could be great. One advantage is that it's about saving the planet, not just understanding the planet.

What do the rest of you think about this? Can these reflections help us pick a specific project that programmers can get involved in? A project that's manageable, yet exciting enough to galvanize people into action?

Comment Source:A very interesting conversation! There are lots of things to say. Here's something about "green mathematics". For me, the manageable portion of "green mathematics" is "network theory": understanding the mathematics governing the many networks that make up the world, from food webs to smart grids, and also the networks we use to understand the world, like Bayesian networks. I'm working hard with my grad students to develop the foundations of network theory. I'm advertising this project whenever I can, and some academics - mainly mathematicians, physicists and highly theoretical computer scientists - are getting interested. I have a huge amount to say about it that I haven't had time to say yet. So, I'm happy with how the "green mathematics / network theory" project is proceeding, except for two things: 1. It's very theoretical, and I'm having a bit of trouble spinning off "bite-sized sub-projects" that would be good for people who like to write programs. I bet such projects could be invented, and could be useful... but someone may have to nudge me in that direction, because I naturally focus on figuring out stuff just by thinking. So, for example, Jim Stuttard and David Tanzer and [[Ken Webb]] have done work on software for stochastic Petri nets: either reviewing the existing software, or writing blog articles about Petri net programming, or actually doing such programming... but so far there's been a relative lack of "published products", perhaps because I never jumped in and said _yes, I need you to simulate this particular stochastic Petri net!_ Why not? I can imagine some questions about Petri nets that require a lot of experiments to answer, where simulations would be crucial. But since I don't program, I tend to focus on other questions... and none of you have _pushed_ me to change this habit. 1. It's not very practical _yet_. It would be great to choose a specific practical application project to work on, a kind of test case to focus the research a bit and let people see what the theoretical work can do. I have an idea: using climate networks to study El Ni&ntilde;o. This is something [in the news](http://www.wunderground.com/blog/JeffMasters/an-el-nio-coming-in-2014), something that lots of people are excited about and arguing about. Programming would be really essential here! Graham proposed another example: [[Zero carbon Britain]]. I don't understand this well enough yet, but it could be great. One advantage is that it's about _saving the planet_, not just _understanding the planet_. What do the rest of you think about this? Can these reflections help us pick a specific project that programmers can get involved in? A project that's manageable, yet exciting enough to galvanize people into action?
• Options
8.

I like the sound of the El Niño project, especially if programming could really help out!

Let's read up on the materials that John has pointed us to, and discuss. In software-ese we could call it preliminary requirements gathering.

Comment Source:I like the sound of the El Niño project, especially if programming could really help out! Let's read up on the materials that John has pointed us to, and discuss. In software-ese we could call it preliminary requirements gathering.
• Options
9.

Besides "green mathematics", the other project I feel we've been most successful at is "education on energy and environmental issues" - mainly through our blog posts, including guest posts, but also the article that David Tweed and I wrote.

I think we could a lot more here! We haven't done anything like

Development of curriculum on science that really matters, (1) for general audiences, (2) for kids.

We've been mainly focused on educating ourselves, and "people like us". But this may be okay. Even if we restricted our attention to educating scientifically literate people, we'd have a lot of work to do. For example: I've been pulling in guest bloggers, but the rest of you could do that too... no? If not everything were funneled through me, we could - in principle - increase our production of articles to the point where we could start thinking about assembling their information into a more systematic resource.

Comment Source:Besides "green mathematics", the other project I feel we've been most successful at is "education on energy and environmental issues" - mainly through our blog posts, including guest posts, but also the article that David Tweed and I wrote. I think we could a lot more here! We haven't done anything like > Development of curriculum on **science that really matters**, (1) for general audiences, (2) for kids. We've been mainly focused on educating _ourselves_, and "people like us". But this may be okay. Even if we restricted our attention to educating scientifically literate people, we'd have a lot of work to do. For example: I've been pulling in guest bloggers, but the rest of you could do that too... no? If not everything were funneled through me, we could - in principle - increase our production of articles to the point where we could start thinking about assembling their information into a more systematic resource. What do the rest of you think about this "educational" aspect of the Azimuth Project?
• Options
10.
edited June 2014

My impression is that we are better placed to teach scientifically literate people. Does anyone here have experience teaching kids?

On the other hand 'things about probability and stochastics that every citizen should know' will include some things that many non-mathematical scientists don't know, or don't understand well. I think a lot of non-mathematical scientists get a course in statistics which rushes through probability theory, and often leaves them able to calculate p-levels (or if they're lucky Bayes factors) but not really understand what they are doing or or why. I reckon the first basic probability concept that generally well-educated people usually trip over is conditional probability. But is that fixing that our problem?

Comment Source:> What do the rest of you think about this “educational” aspect of the Azimuth Project? My impression is that we are better placed to teach scientifically literate people. Does anyone here have experience teaching kids? On the other hand 'things about probability and stochastics that every citizen should know' will include some things that many non-mathematical scientists don't know, or don't understand well. I think a lot of non-mathematical scientists get a course in statistics which rushes through probability theory, and often leaves them able to calculate p-levels (or if they're lucky Bayes factors) but not really understand what they are doing or or why. I reckon the first basic probability concept that generally well-educated people usually trip over is conditional probability. But is that fixing that *our* problem?
• Options
11.
edited June 2014

Does anyone here have experience teaching kids?

I do have, but I wouldn't work for free in that area. That is if something is a peer-to-peer educational thing like Azimuth then this is at least somewhat of an educational exchange (depending on the quality of the blog comments etc.).

However -even if one can learn from kids- the educational flow is in the generic case mostly into one direction but this depends of course on age, intelligence etc.

I am actually currently thinking about whether I should teach again at high schools, unfortunately this would be something like a full-time commitment, since the Berlin school ministry offers no smaller part-time solutions. For Azimuth this would mean less contributions from my side if I should do that and if I should get the job (whether this is good or bad is another question...).

Comment Source:>Does anyone here have experience teaching kids? I do have, but I wouldn't work for free in that area. That is if something is a peer-to-peer educational thing like Azimuth then this is at least somewhat of an educational *exchange* (depending on the quality of the blog comments etc.). However -even if one can learn from kids- the educational flow is in the generic case mostly into one direction but this depends of course on age, intelligence etc. I am actually currently thinking about whether I should teach again at high schools, unfortunately this would be something like a full-time commitment, since the Berlin school ministry offers no smaller part-time solutions. For Azimuth this would mean less contributions from my side if I should do that and if I should get the job (whether this is good or bad is another question...).
• Options
12.

Graham wrote:

My impression is that we are better placed to teach scientifically literate people. Does anyone here have experience teaching kids?

I teach kids who are fresh out of high school.

But younger children? No.

I reckon the first basic probability concept that generally well-educated people usually trip over is conditional probability. But is that fixing that our problem?

My feeling is that "our problem" depends a lot on what we want to do... and that what we want to do can be discovered by looking at what we actually are doing.

In other words: instead of launching new projects that none of us are already doing, I tend to feel we should pay attention to what we actually are doing, and do more of that, and get more people involved doing that. However, I do feel that the programmers among us would like some guidance in finding cool but not-too-hard projects.

So I feel that right now the Azimuth Project is mainly focused on:

1. Developing network theory, with an eye toward environmental and energy applications.

2. Learning and explaining climate science, the challenges of getting carbon-free energy, etc. This is what Nad called a "peer-to-peer educational thing".

3. Writing programs that help people understand climate science.

Part 3 is where I sense the most frustration.

Comment Source:Graham wrote: > My impression is that we are better placed to teach scientifically literate people. Does anyone here have experience teaching kids? I teach kids who are fresh out of high school. <img src = "http://math.ucr.edu/home/baez/emoticons/tongue2.gif" alt = ""/> But younger children? No. > I reckon the first basic probability concept that generally well-educated people usually trip over is conditional probability. But is that fixing that _our_ problem? My feeling is that "our problem" depends a lot on what we want to do... and that what we want to do can be discovered by looking at what we actually _are_ doing. In other words: instead of launching new projects that none of us are already doing, I tend to feel we should pay attention to what we actually _are_ doing, and do more of that, and get more people involved doing that. However, I _do_ feel that the programmers among us would like some guidance in finding cool but not-too-hard projects. So I feel that right now the Azimuth Project is mainly focused on: 1. Developing network theory, with an eye toward environmental and energy applications. 1. Learning and explaining climate science, the challenges of getting carbon-free energy, etc. This is what Nad called a "peer-to-peer educational thing". 1. Writing programs that help people understand climate science. Part 3 is where I sense the most frustration.
• Options
13.

John wrote:

I tend to feel we should pay attention to what we actually are doing, and do more of that, and get more people involved doing that. However, I do feel that the programmers among us would like some guidance in finding cool but not-too-hard projects.

So I've been somewhat lacking in Azimuth direction.I had assumed that non-programming domain specialists would try producing some Petri net models using one of the most user-friendly cloud-based tools like NetLogo.

Nathan has listed candidate projects. I could only grok some aspects of a simple climate model from Gerald North's book. I coded up the equations until fig. 3.6 but neither I nor Glyn could work out a coherent overall model from the book. Getting the code from Gerald would help my climate education.

I'm still working on various automatically generated javascript models. I'd find any detailed requirements and specifications most welcome.

Comment Source:John wrote: > I tend to feel we should pay attention to what we actually are doing, and do more of that, and get more people involved doing that. However, I do feel that the programmers among us would like some guidance in finding cool but not-too-hard projects. So I've been somewhat lacking in Azimuth direction.I had assumed that non-programming domain specialists would try producing some Petri net models using one of the most user-friendly cloud-based tools like NetLogo. Nathan has listed candidate projects. I could only grok some aspects of a simple climate model from Gerald North's book. I coded up the equations until fig. 3.6 but neither I nor Glyn could work out a coherent overall model from the book. Getting the code from Gerald would help my climate education. I'm still working on various automatically generated javascript models. I'd find any detailed requirements and specifications most welcome.
• Options
14.
edited June 2014

Jim wrote:

I had assumed that non-programming domain specialists would try producing some Petri net models using one of the most user-friendly cloud-based tools like NetLogo.

Right now I'm not working on any problems that make me interested in actually running Petri net models. I do appreciate all your work reviewing Petri net software and will include a chapter about that in the book (which keeps getting delayed, but I hope to finish this summer). But since I don't tend to use software for computation (just communication), I tend to avoid math problems where software would be useful.

This might be something we could fix by working together, but it would require that both of us decide to do it, and communicate enough to make it happen.

More later...

Comment Source:Jim wrote: > I had assumed that non-programming domain specialists would try producing some Petri net models using one of the most user-friendly cloud-based tools like NetLogo. Right now I'm not working on any problems that make me interested in actually running Petri net models. I do appreciate all your work reviewing Petri net software and will include a chapter about that in the book (which keeps getting delayed, but I hope to finish this summer). But since I don't tend to use software for computation (just communication), I tend to avoid math problems where software would be useful. This _might_ be something we could fix by working together, but it would require that both of us decide to do it, and communicate enough to make it happen. More later...
• Options
15.
edited June 2014

... right now I'm actually more interested in getting someone to help me with some computations related to climate networks and El Niño! This seems like a quicker route to doing something related to global warming than Petri nets. Also, I've been invited to speak at NIPS, a conference on neural networks and machine learning, and if we could get something done by December I could present that work.

This is a short video that explains the issues in a simple way:

This is the new paper that created a stir:

It's not free, but I sent copies to some of you Azimuthers, and I can send it to others too. A lot of the methodology seems to be in in this free paper:

I suspect there are methodological problems with these papers, but that gives us things to do!

Anyone interested?

Comment Source:... right now I'm actually more interested in getting someone to help me with some computations related to climate networks and El Ni&ntilde;o! This seems like a quicker route to doing something related to global warming than Petri nets. Also, I've been invited to speak at NIPS, a conference on neural networks and machine learning, and if we could get something done by December I could present that work. This is a short video that explains the issues in a simple way: * [ScienceCasts: El Niño - Is 2014 the New 1997?](http://youtu.be/zaxPwASV2kY). This is the new paper that created a stir: * Josef Ludescher, Avi Gozolchiani, Mikhail I. Bogachev, Armin Bunde, Shlomo Havlin, and Hans Joachim Schellnhuber, Very early warning of next El Niño, _[Proceedings of the National Academy of Sciences](http://www.pnas.org/content/early/2014/02/07/1323058111)_, February 2014. It's not free, but I sent copies to some of you Azimuthers, and I can send it to others too. A lot of the methodology seems to be in in this free paper: * Josef Ludescher, Avi Gozolchiani, Mikhail I. Bogachev, Armin Bunde, Shlomo Havlin, and Hans Joachim Schellnhuber, [Improved El Niño forecasting by cooperativity detection](http://www.pnas.org/content/early/2013/06/26/1309353110.full.pdf+html), _Proceedings of the National Academy of Sciences_, 30 May 2013. I suspect there are methodological problems with these papers, but that gives us things to do! Anyone interested?
• Options
16.
edited June 2014

Okay, it sounds like David is interested:

I like the sound of the El Niño project, especially if programming could really help out!

Let’s read up on the materials that John has pointed us to, and discuss.

You should mainly ask questions, and make me answer them... though some reading would do us all good.

To copy and then improve on this paper on El Niño prediction:

we would first need to get ahold of daily temperature data for "14 grid points in the El Niño basin and 193 grid points outside this domain" from 1981 to 2014. That's 207 locations and 34 years. This data is supposedly available from the National Centers for Environmental Prediction and the National Center for Atmospheric Research Reanalysis I Project.

The paper starts by taking these temperatures, computing the average temperature at each day of the year at each location, and subtracting this from the actual temperatures to obtain "temperature anomalies". In other words, we want a big array of numbers like this: the temperature on March 21st 1990 at some location, minus the average temperature on all March 21sts from 1981 to 2014 at that location.

Then they process this array of numbers in various ways, which I can explain...

They consider all pairs of locations, so at some point they are working with 207 × 207 × 365 × 34 numbers. Is that a lot of numbers these days?

Comment Source:Okay, it sounds like David is interested: > I like the sound of the El Niño project, especially if programming could really help out! > Let’s read up on the materials that John has pointed us to, and discuss. You should mainly ask questions, and make me answer them... though some reading would do us all good. To copy and then _improve on_ this paper on El Niño prediction: * Josef Ludescher, Avi Gozolchiani, Mikhail I. Bogachev, Armin Bunde, Shlomo Havlin, and Hans Joachim Schellnhuber, [Improved El Niño forecasting by cooperativity detection](http://www.pnas.org/content/early/2013/06/26/1309353110.full.pdf+html), _Proceedings of the National Academy of Sciences_, 30 May 2013. we would first need to get ahold of daily temperature data for "14 grid points in the El Niño basin and 193 grid points outside this domain" from 1981 to 2014. That's 207 locations and 34 years. This data is supposedly available from the National Centers for Environmental Prediction and the National Center for Atmospheric Research Reanalysis I Project. The paper starts by taking these temperatures, computing the average temperature at each day of the year at each location, and subtracting this from the actual temperatures to obtain "temperature anomalies". In other words, we want a big array of numbers like this: the temperature on March 21st 1990 at some location, minus the average temperature on all March 21sts from 1981 to 2014 at that location. Then they process this array of numbers in various ways, which I can explain... They consider all _pairs_ of locations, so at some point they are working with 207 &times; 207 &times; 365 &times; 34 numbers. Is that a lot of numbers these days?
• Options
17.
edited June 2014

I am now creating wiki pages for (1) a basic explanation of climate networks, and (2) our new programming project to use climate networks for El Niño prediction.

I have copied all of the relevant references and information from this thread to these two new wiki pages. In a few minutes I will start the corresponding two threads. So let's carry on those conversations over there, and leave this thread as a "quieter" place for discussing the meta and strategy issues related to Azimuth initiatives -- and as a place for spawning off new specific projects. Already it worked to spawn off an El Niño programming project!

Comment Source:I am now creating wiki pages for (1) a basic explanation of climate networks, and (2) our new programming project to use climate networks for El Niño prediction. I have copied _all_ of the relevant references and information from this thread to these two new wiki pages. In a few minutes I will start the corresponding two threads. So let's carry on those conversations over there, and leave this thread as a "quieter" place for discussing the meta and strategy issues related to Azimuth initiatives -- and as a place for spawning off new specific projects. Already it worked to spawn off an El Niño programming project!
• Options
18.
edited June 2014

let’s ... leave this thread as a “quieter” place for discussing the meta and strategy issues related to Azimuth initiatives – and as a place for spawning off new specific projects.

Yes, good. By the way, I earlier moved this thread from the catch-all category "Azimuth Project" to the subcategory "Strategy". That's the right place for thinking at a broad strategic level about what the Azimuth Project should do.

Comment Source:> let’s ... leave this thread as a “quieter” place for discussing the meta and strategy issues related to Azimuth initiatives – and as a place for spawning off new specific projects. Yes, good. By the way, I earlier moved this thread from the catch-all category "Azimuth Project" to the subcategory "Strategy". That's the right place for thinking at a broad strategic level about what the Azimuth Project should do.