Options

An offer from Dara Shayda

Do any of you Azimuth Code Project bosses have ideas for how to take advantage of this email? I'm not quite sure what to make of it. It's possible we could nudge him over to being interested in our El Niño project.


Hello John

How are you? this is Dara.

I have mastered (developed software for) the machine learning algorithms for forecasting time-series and moving now to volumetric data. Also using wavelets theory for denoising and noise measures. I have parallelized the algorithms effectively.

I have now 2 servers, each 16 cpu, one with additional large GPU cluster. Please see sample data:

This is one of the first exciting images from GPM satellite:

http://www.nasa.gov/mission_pages/GPM/main/#.U5GpZS-k8UU

I like to build a large open source development for GPM output for rain and other water related data. Hope to have your attention and cooperation.

I did some solve some YBE equations over finite fields ($\mathbb{Z}_9$) and Grassmannians, but that is for fun.

As always with much admiration,

Dara Shayda

Comments

  • 1.

    With 32 cores it becomes feasible to run some serious models. The methane emissions study of Boston springs to mind as the sort of imaging which Azimuth could take useful advantage of.

    Experiments with Gistemp data would certainly be another case.

    That would be wonderful especially as Dara has a setup and a system for managing it which I'm not sure would available to John or other Azimuth members with access to large university computer resources but I don't know.

    Dave Tanzer has a Linode machine which isn't online. (Dave if I helped with security would you consider running Nixos and some server like Snap (every instance is sandboxed so that it's nearly impenetrable) or nginx?).

    I'm writing this partly in my own interests as I'm about to ditch the cheapo Bluehost prototype server in a week I have the problem of losing use of my domain name and thus email address until I find the best alternative. I haven't found a cloud deployment server providing ISP services for less than about $800 per year. Azimuth models will have to be hosted somewhere and at the moment it seems there's going to have to be an Azimuth servers to deploy models developed on cloud servers like travis-ci or fpcomplete. Obviously I'd share the cost of an Azimuth server whatever it is.

    Comment Source:With 32 cores it becomes feasible to run some serious models. The methane emissions study of Boston springs to mind as the sort of imaging which Azimuth could take useful advantage of. Experiments with Gistemp data would certainly be another case. That would be wonderful especially as Dara has a setup and a system for managing it which I'm not sure would available to John or other Azimuth members with access to large university computer resources but I don't know. Dave Tanzer has a Linode machine which isn't online. (Dave if I helped with security would you consider running Nixos and some server like Snap (every instance is sandboxed so that it's nearly impenetrable) or nginx?). I'm writing this partly in my own interests as I'm about to ditch the cheapo Bluehost prototype server in a week I have the problem of losing use of my domain name and thus email address until I find the best alternative. I haven't found a cloud deployment server providing ISP services for less than about $800 per year. Azimuth models will have to be hosted somewhere and at the moment it seems there's going to have to be an Azimuth servers to deploy models developed on cloud servers like [travis-ci](http://www.travis-ci.com) or [fpcomplete](http://www.fpcomplete.com). Obviously I'd share the cost of an Azimuth server whatever it is.
  • 2.
    edited June 2014

    That would be wonderful especially as Dara has a setup and a system for managing it which I’m not sure would available to John or other Azimuth members with access to large university computer resources but I don’t know.

    I would probably be able to get access to large computer resources if I demonstrated that I could use them and were committed to doing so. However, to reach this point, it would help a lot to have some people in the Azimuth Project who have such resources and are successfully using them. Same principle as "it takes money to make money."

    The methane emissions study of Boston springs to mind as the sort of imaging which Azimuth could take useful advantage of.

    What's that? I don't remember that.

    Experiments with Gistemp data would certainly be another case.

    This is closer to the things we - especially Jan Galkowski - are already doing. El Niño calculations also leap to mind. I think that if we get Dara involved, it's important to try to get him to work on something we're already working on. We've got to focus our work to have a chance to do something interesting.

    What do you think, David Tanzer? When I mentioned "Azimuth Code Project bosses" I was jokingly alluding to you. I can talk to Dara, at least until it gets technical, but I'd like us to be on the same page.

    Comment Source:> That would be wonderful especially as Dara has a setup and a system for managing it which I’m not sure would available to John or other Azimuth members with access to large university computer resources but I don’t know. I would probably be able to get access to large computer resources if I demonstrated that I could use them and were committed to doing so. However, to reach this point, it would help a lot to have some people in the Azimuth Project who _have_ such resources and _are_ successfully using them. Same principle as "it takes money to make money." > The methane emissions study of Boston springs to mind as the sort of imaging which Azimuth could take useful advantage of. What's that? I don't remember that. > Experiments with Gistemp data would certainly be another case. This is closer to the things we - especially [[Jan Galkowski]] - are already doing. El Niño calculations also leap to mind. I think that if we get Dara involved, it's important to try to get him to work on something _we're already working on_. We've got to focus our work to have a chance to do something interesting. What do you think, David Tanzer? When I mentioned "Azimuth Code Project bosses" I was jokingly alluding to you. I can talk to Dara, at least until it gets technical, but I'd like us to be on the same page.
  • 3.
    edited June 2014

    I didn't want to suggest any response which might divert Dara from getting involved in the El Nino stuff after such kind offer.

    The estimate in the Boston study was about 7% methane leakage from the gas main. Whether the Koch brothers own that as well I don't know.

    The methane emissions graphic (pretty much like Dara's) is here and the paper is here.

    Comment Source:I didn't want to suggest any response which might divert Dara from getting involved in the El Nino stuff after such kind offer. The estimate in the Boston study was about 7% methane leakage from the gas main. Whether the Koch brothers own that as well I don't know. The methane emissions graphic (pretty much like Dara's) is [here](http://www.bu.edu/energy/research/technologies-engineered-systems/methane-emissions/) and the paper is [here](http://biology.duke.edu/jackson/ep2013.pdf).
  • 4.

    Well, it's time for me to reply to Dara so I'll go ahead based on the input I got so far.

    Comment Source:Well, it's time for me to reply to Dara so I'll go ahead based on the input I got so far.
  • 5.

    HI I've been distracted by some work related stuff, but I've been thinking about this question. I'll reply in the next hour our so.

    Comment Source:HI I've been distracted by some work related stuff, but I've been thinking about this question. I'll reply in the next hour our so.
  • 6.

    I would say something to the effect of:

    Hi Dara, thanks for your kind offer of computational resources to support our research work at Azimuth. Just by coincidence, a this time we're starting to look into writing programs to analyze temperature field data, with particular application to El Niño calculations. We're just now reading papers, analyzing them, and starting to work with some NOAA data. This may well lead to calculations which will require some substantial computing resources, such the one described in your generous offer. As of now we have no computational platform, other than our own individual computers. So this platform could turn out to be a good fit -- though it's too early to say today.

    We would welcome your participation in our preliminary discussions about this research, both in terms of you offering your perspective on the computational aspects -- and we'd like to hear more about your ideas for development for water and rain related data. In our group there are some mathematicians, programmers, computer scientists, and climate scientists, so there are a number of perspectives that come together.

    Here are some relevant discussion threads (all of the recent threads on El Niño calculations and papers).

    Let us know if you would like a Forum login.

    In any case, I will keep you posted of our progress, and hope that we will be able to collaborate in the future -- using the hardware platform, or otherwise.

    Comment Source:I would say something to the effect of: Hi Dara, thanks for your kind offer of computational resources to support our research work at Azimuth. Just by coincidence, a this time we're starting to look into writing programs to analyze temperature field data, with particular application to El Niño calculations. We're just now reading papers, analyzing them, and starting to work with some NOAA data. This may well lead to calculations which will require some substantial computing resources, such the one described in your generous offer. As of now we have no computational platform, other than our own individual computers. So this platform could turn out to be a good fit -- though it's too early to say today. We would welcome your participation in our preliminary discussions about this research, both in terms of you offering your perspective on the computational aspects -- and we'd like to hear more about your ideas for development for water and rain related data. In our group there are some mathematicians, programmers, computer scientists, and climate scientists, so there are a number of perspectives that come together. Here are some relevant discussion threads (all of the recent threads on El Niño calculations and papers). Let us know if you would like a Forum login. In any case, I will keep you posted of our progress, and hope that we will be able to collaborate in the future -- using the hardware platform, or otherwise.
  • 7.
    edited June 2014

    Jim wrote:

    Dave Tanzer has a Linode machine which isn’t online. (Dave if I helped with security would you consider running Nixos and some server like Snap (every instance is sandboxed so that it’s nearly impenetrable) or nginx?).

    Actually, I let my Linode account lapse, because I was only using it to fiddle around with Instiki -- which I prefer to do on the Rails Playground.

    It may be that a fast laptop with good memory will suffice for the analysis that we want to do on the NOAA data, in which case I would rather spend the time on defining the experiments, programming them and running them. And if we hit the limits of these machines with a real experiment that we want to perform, that itself will be a sign of progress. By that point, our specific machine requirements will have become more clear.

    Comment Source:Jim wrote: > Dave Tanzer has a Linode machine which isn’t online. (Dave if I helped with security would you consider running Nixos and some server like Snap (every instance is sandboxed so that it’s nearly impenetrable) or nginx?). Actually, I let my Linode account lapse, because I was only using it to fiddle around with Instiki -- which I prefer to do on the Rails Playground. It may be that a fast laptop with good memory will suffice for the analysis that we want to do on the NOAA data, in which case I would rather spend the time on defining the experiments, programming them and running them. And if we hit the limits of these machines with a real experiment that we want to perform, that itself will be a sign of progress. By that point, our specific machine requirements will have become more clear.
  • 8.
    nad
    edited June 2014

    I did some solve some YBE equations over finite fields (ℤ 9\mathbb{Z}_9) and Grassmannians, but that is for fun.

    I also did something with YBE equations in my dissertation (in 1996....though)!!

    So I was googling for Dara's works but isn't his name rather Dara O. Shayda?

    That is there is a Dara Shayda who writes religious texts: http://www.techofheart.co/2008/07/who-do-you-really-love-dara-shayda.html

    and one who is interested in DIY drones (or may be they are both the same): http://diydrones.com/profile/DaraShayda

    those links appear for me from the google buble - but this Dara Shayda does not seem to be interested in Grassmannians.

    there is though a Dara O. Shayda who writes on Kolmogorov Complexity: http://www.informatik.uni-trier.de/~ley/pers/hd/s/Shayda:Dara_O=

    Comment Source:>I did some solve some YBE equations over finite fields (ℤ 9\mathbb{Z}_9) and Grassmannians, but that is for fun. I also did something with YBE equations in my dissertation (in 1996....though)!! So I was googling for Dara's works but isn't his name rather Dara O. Shayda? That is there is a Dara Shayda who writes religious texts: http://www.techofheart.co/2008/07/who-do-you-really-love-dara-shayda.html and one who is interested in DIY drones (or may be they are both the same): http://diydrones.com/profile/DaraShayda those links appear for me from the google buble - but this Dara Shayda does not seem to be interested in Grassmannians. there is though a Dara O. Shayda who writes on Kolmogorov Complexity: http://www.informatik.uni-trier.de/~ley/pers/hd/s/Shayda:Dara_O=
  • 9.
    edited June 2014

    I emailed Dara. Here's what I wrote:


    Dear Dara -

    Hi! The Azimuth Project wants to analyze sea surface temperatures in the Pacific Ocean to see if we can predict El Nino events. We have some specific ideas for doing this, which you can read about here:

    http://forum.azimuthproject.org/discussion/1358/experiments-in-el-nino-detection-and-prediction/

    Briefly, we want to improve on a recent paper by Ludescher and coauthors:

    http://forum.azimuthproject.org/discussion/1360/paper-ludescher-et-al-improved-el-nino-forecasting-by-cooperativity-detection/#Item_42

    They actually have a couple of papers; I'll attach the one that is not freely available online. This new paper caused quite a stir, since it says an El Nino event is coming soon.

    If you would like to use your computers and skills to help us, that would be great. So far our project is just starting, so we don't need a lot of processing power yet, but if things work out well we could.

    Best,

    jb

    Comment Source:I emailed Dara. Here's what I wrote: <hr/> Dear Dara - Hi! The Azimuth Project wants to analyze sea surface temperatures in the Pacific Ocean to see if we can predict El Nino events. We have some specific ideas for doing this, which you can read about here: http://forum.azimuthproject.org/discussion/1358/experiments-in-el-nino-detection-and-prediction/ Briefly, we want to improve on a recent paper by Ludescher and coauthors: http://forum.azimuthproject.org/discussion/1360/paper-ludescher-et-al-improved-el-nino-forecasting-by-cooperativity-detection/#Item_42 They actually have a couple of papers; I'll attach the one that is not freely available online. This new paper caused quite a stir, since it says an El Nino event is coming soon. If you would like to use your computers and skills to help us, that would be great. So far our project is just starting, so we don't need a lot of processing power yet, but if things work out well we could. Best, jb
  • 10.

    Here is Dara's reply:


    Dear John

    Good to hear from you.

    1. I can allocate for you 2 16-cpu cores, one with an NVIDIA Tesla GPU, 64GB of physical memory, could easily go to Terra Byte of physical memory if needed, almost unlimited storage! So I could download all the volumetric and image data from NASA and other sources and keep our own copy :)
    2. Have Mathematica license for grid computing up to 16 kernels, already using it for parallel computing, so this license (quite pricy) is also available for your research and publication. This way we could do very complex analytical computations, and I show you the ones I already have very soon.
    3. We use C as well, I could do both CUDA and OPENMP for parallelization of forecasting algorithms. I use gcc for the c compeer which should be ok for your needs.
    4. iPython server is also installed on my servers, so you could make either Python notebooks or Mathematica notebooks to experiment and share with others.
    5. Easily I could add more servers as your needs grow
    6. I have IT experts who manage these servers, do not worry about that end of things

    This link is invalid, please resend: http://forum.azimuthproject.org/discussion/1358/experiments-in-el-nino-detection-and-prediction/

    so we don't need a lot of processing power yet, but if things work out well we could.

    I show you how the forecast algorithms chew up memory and cpu for even smallest forecasting, believe me you need large machines very quickly :)

    I keep in touch with some samples for you shortly, I read up your posts to see what you are up to and need to be coded.

    Dara


    The link I sent is apparently only readable when you're logged into the Forum, that's sort of a bug. I sent Dara a link to the wiki and explained how to join the Forum.

    Comment Source:Here is Dara's reply: <hr/> Dear John Good to hear from you. 1. I can allocate for you 2 16-cpu cores, one with an NVIDIA Tesla GPU, 64GB of physical memory, could easily go to Terra Byte of physical memory if needed, almost unlimited storage! So I could download all the volumetric and image data from NASA and other sources and keep our own copy :) 2. Have Mathematica license for grid computing up to 16 kernels, already using it for parallel computing, so this license (quite pricy) is also available for your research and publication. This way we could do very complex analytical computations, and I show you the ones I already have very soon. 3. We use C as well, I could do both CUDA and OPENMP for parallelization of forecasting algorithms. I use gcc for the c compeer which should be ok for your needs. 4. iPython server is also installed on my servers, so you could make either Python notebooks or Mathematica notebooks to experiment and share with others. 5. Easily I could add more servers as your needs grow 6. I have IT experts who manage these servers, do not worry about that end of things This link is invalid, please resend: http://forum.azimuthproject.org/discussion/1358/experiments-in-el-nino-detection-and-prediction/ > so we don't need a lot of processing power yet, but if things work out well we could. I show you how the forecast algorithms chew up memory and cpu for even smallest forecasting, believe me you need large machines very quickly :) I keep in touch with some samples for you shortly, I read up your posts to see what you are up to and need to be coded. Dara <hr/> The link I sent is apparently only readable when you're logged into the Forum, that's sort of a bug. I sent Dara a link to the wiki and explained how to join the Forum.
  • 11.
    edited June 2014

    Nad wrote:

    So I was googling for Dara’s works but isn’t his name rather Dara O. Shayda?

    The email I got is from Dara Shayda (dara@lossofgenerality.com), but I've been getting emails from Dara intermittently since at least 2010, and the early ones use the name Dara O. Shayda (dara@untiredwithloving.org). So, I think some of the different aspects you mention are all aspects of the same person.

    Comment Source:Nad wrote: > So I was googling for Dara’s works but isn’t his name rather Dara O. Shayda? The email I got is from Dara Shayda (dara@lossofgenerality.com), but I've been getting emails from Dara intermittently since at least 2010, and the early ones use the name Dara O. Shayda (dara@untiredwithloving.org). So, I think some of the different aspects you mention are all aspects of the same person.
  • 12.
    edited June 2014

    I started that thread under Azimuth Code Project, John can you check the visibility settings for that category and make it public if it's not -- thanks.

    Comment Source:I started that thread under Azimuth Code Project, John can you check the visibility settings for that category and make it public if it's not -- thanks.
  • 13.

    I just tried to make that category visible to Unauthenticated (but not Banned) viewers.

    Comment Source:I just tried to make that category visible to Unauthenticated (but not Banned) viewers.
  • 14.

    Okay, it worked - now Azimuth Code Project articles here on the Forum are visible to everyone in the world (since we haven't Banned anyone yet).

    Comment Source:Okay, it worked - now Azimuth Code Project articles here on the Forum are visible to everyone in the world (since we haven't Banned anyone yet).
  • 15.

    So, I think some of the different aspects you mention are all aspects of the same person.

    He seems to be a versatile man. I think his offer sounds very generous. Especially things like Mathematica licences are really expensive. (Finally since Jan Poeschko the creator of mathics is now at Wolfram, it is not clear how many open source computer algebra systems aside from Sage will be further available for use.) However given the current financial transparency of Azimuth it would be important to learn more about the financial backing of this offer.

    Comment Source:> So, I think some of the different aspects you mention are all aspects of the same person. He seems to be a versatile man. I think his offer sounds very generous. Especially things like Mathematica licences are really expensive. (Finally since Jan Poeschko the creator of <a href="http://www.mathics.org/">mathics</a> is now at Wolfram, it is not clear how many open source computer algebra systems aside from Sage will be further available for use.) However given the current <a href="http://www.azimuthproject.org/azimuth/show/Financial+Ressources+of+Azimuth">financial transparency of Azimuth</a> it would be important to learn more about the financial backing of this offer.
  • 16.
    edited June 2014

    His recently posted simulation results do suggest that he has a powerful computational and analytic platform at his disposal.

    I'm starting to experience the limits of laptop computing, not in terms of memory, but processor time. Graham's demo script, which plots the graph of median covariances -- which works very well, thanks Graham! -- takes about five minutes to run, and that is just using two values for the time offset, eight values for dLongitude, and one value for dLatitude.

    We may be able to get some mileage out of sampling / Monte Carlo techniques. For instance, to compute mean covariances, we should be able o pick appropriate random pairs of points, and take the means of the resulting covariances.

    It could be useful, in the near future, if we had a shell account and could run R programs on it. I'm also open to using other platforms, including Mathematica, but that would be more of a long term project to learn and explore. Also, I couldn't promise to use it in any particular manner or frequency, because of limited amounts of time for Azimuth.

    Comment Source:His recently posted simulation results do suggest that he has a powerful computational and analytic platform at his disposal. I'm starting to experience the limits of laptop computing, not in terms of memory, but processor time. Graham's demo script, which plots the graph of median covariances -- which works very well, thanks Graham! -- takes about five minutes to run, and that is just using two values for the time offset, eight values for dLongitude, and one value for dLatitude. We may be able to get some mileage out of sampling / Monte Carlo techniques. For instance, to compute mean covariances, we should be able o pick appropriate random pairs of points, and take the means of the resulting covariances. It could be useful, in the near future, if we had a shell account and could run R programs on it. I'm also open to using other platforms, including Mathematica, but that would be more of a long term project to learn and explore. Also, I couldn't promise to use it in any particular manner or frequency, because of limited amounts of time for Azimuth.
  • 17.

    On speed. In order to calculate a bunch of covariances between signal x and y, over a period N (eg 365) and a range of delays 0 to M (eg 0 to 200) it is sufficient to find 3(M+1) quantities:

    $$ \sum_{1}^{N} x_i, \sum_{2}^{N+1} x_i, \dots, \sum_{M}^{N+M} x_i $$ plus two more like that with y and xy in place of x. Since

    $$ \sum_{2}^{N+1} x_i - \sum_{1}^{N} x_i = x_{N+1} - x_1$$ and so on, this is not as much work as it first appears. Care would be needed to avoid numerical problems.

    Comment Source:On speed. In order to calculate a bunch of covariances between signal x and y, over a period N (eg 365) and a range of delays 0 to M (eg 0 to 200) it is sufficient to find 3(M+1) quantities: $$ \sum_{1}^{N} x_i, \sum_{2}^{N+1} x_i, \dots, \sum_{M}^{N+M} x_i $$ plus two more like that with y and xy in place of x. Since $$ \sum_{2}^{N+1} x_i - \sum_{1}^{N} x_i = x_{N+1} - x_1$$ and so on, this is not as much work as it first appears. Care would be needed to avoid numerical problems.
  • 18.
    edited June 2014

    Dave wrote:

    I’m starting to experience the limits of laptop computing, not in terms of memory, but processor time

    On my dual-core ~1.6MHz 4 MB RAM notebook (I can't stand the noise of my sparc server anymore) I think I have similar feelings and, although cloud compiling is reasonably cheap or free up to about 16 processors, anything more serious costs serious dosh, so I reckon Azimuth could soon use any extra power Dara can offer.

    Comment Source:Dave wrote: > I’m starting to experience the limits of laptop computing, not in terms of memory, but processor time On my dual-core ~1.6MHz 4 MB RAM notebook (I can't stand the noise of my sparc server anymore) I think I have similar feelings and, although cloud compiling is reasonably cheap or free up to about 16 processors, anything more serious costs serious dosh, so I reckon Azimuth could soon use any extra power Dara can offer.
  • 19.
    edited June 2014

    Nad wrote:

    However given the current financial transparency of Azimuth it would be important to learn more about the financial backing of this offer.

    I will ask Dara how he got ahold of his rather large computer resources. I don't know much about him, and I'm getting curious. I don't expect anything sinister to emerge, but transparency is indeed valuable.

    Since I'm a kind of internet mini-celebrity, lots of people read my stuff and get interested in doing things for me — or getting me to do things for them. For example, a couple of years ago in Singapore a Chinese billionaire invited me to lunch and offered to pay me to help him develop his theories on Taoism and physics. He said he could get me a position at a Chinese university near where he lives. (I politely said I was busy with other things.) On the more altruistic extreme, Greg Egan likes to solve math problems I pose, and often writes impressive computer programs to solve them or illustrate the issues.

    Comment Source:Nad wrote: > However given the current financial transparency of Azimuth it would be important to learn more about the financial backing of this offer. I will ask Dara how he got ahold of his rather large computer resources. I don't know much about him, and I'm getting curious. I don't expect anything sinister to emerge, but transparency is indeed valuable. Since I'm a kind of internet mini-celebrity, lots of people read my stuff and get interested in doing things for me &mdash; or getting me to do things for them. For example, a couple of years ago in Singapore a Chinese billionaire invited me to lunch and offered to pay me to help him develop his theories on Taoism and physics. He said he could get me a position at a Chinese university near where he lives. (I politely said I was busy with other things.) On the more altruistic extreme, Greg Egan likes to solve math problems I pose, and often writes impressive computer programs to solve them or illustrate the issues.
  • 20.

    Since I’m a kind of internet mini-celebrity, lots of people read my stuff and get interested in doing things for me — or getting me to do things for them.

    On the more altruistic extreme, Greg Egan likes to solve math problems I pose, and often writes impressive computer programs to solve them or illustrate the issues.

    Eventually Greg hopes that some of your fame helps to make his computer programs more widely known. I don't know may be he wants to set up a data visualization business.

    Comment Source:>Since I’m a kind of internet mini-celebrity, lots of people read my stuff and get interested in doing things for me — or getting me to do things for them. >On the more altruistic extreme, Greg Egan likes to solve math problems I pose, and often writes impressive computer programs to solve them or illustrate the issues. Eventually Greg hopes that some of your fame helps to make his computer programs more widely known. I don't know may be he wants to set up a data visualization business.
  • 21.
    edited June 2014

    It's rude to gossip about unpleasant theories about people's apparently nice behavior - especially if it's not relevant to our task here, and especially in a public discussion where the person being discussed can read the gossip. So, I won't continue discussing Greg Egan here.

    Comment Source:It's rude to gossip about unpleasant theories about people's apparently nice behavior - especially if it's not relevant to our task here, and especially in a public discussion where the person being discussed can read the gossip. So, I won't continue discussing Greg Egan here.
  • 22.

    I am sorry, but I don't see why my comment is rude gossip. I just commented on possibilities. I didn't say that Greg Egan would want to set up a data visualization business, I just more or less said that this could eventually be a possibility he might have thought of. He seems to be very good at this, and as it looks way faster than me. Finally he has to generate some income from somewhere and maybe he got tired of writing science fiction? But as said, that is what I could imagine about him and not necessarily what Greg Egan thinks about himself. Moreover it seems your comment that "Greg Egan likes to solve math problems" - "on the more altruistic extreme" might be equally unpleasant for Greg Egan, in particular I don't understand why setting up a data visualization business would be automatically "an unpleasant theory about people’s apparently nice behavior".

    Comment Source:I am sorry, but I don't see why my comment is rude gossip. I just commented on <em>possibilities</em>. I didn't say that Greg Egan would want to set up a data visualization business, I just more or less said that this <em>could eventually be a possibility he might have thought of.</em> He seems to be very good at this, and as it looks way faster than me. Finally he has to generate some income from somewhere and maybe he got tired of writing science fiction? But as said, that is what I could imagine about him and not necessarily what Greg Egan thinks about himself. Moreover it seems your comment that "Greg Egan likes to solve math problems" - "on the more altruistic extreme" might be equally unpleasant for Greg Egan, in particular I don't understand why setting up a data visualization business would be automatically "an unpleasant theory about people’s apparently nice behavior".
  • 23.
    edited June 2014

    As I said, I'm not interested in discussing your theories and speculations about Greg Egan here. He's a friend of mine, he enjoys his privacy, I know he would dislike your theories, and I regret ever having mentioned him here. It was an irrelevant digression. Let's stop here.

    Comment Source:As I said, I'm not interested in discussing your theories and speculations about Greg Egan here. He's a friend of mine, he enjoys his privacy, I know he would dislike your theories, and I regret ever having mentioned him here. It was an irrelevant digression. Let's stop here.
  • 24.

    It was an irrelevant digression. Let’s stop here.

    I was thinking wether I should let you have that final say, but I can't, because this is not an irrelevant digression! That is I think it is very connected to the question of how Azimuth will work on in the future. I had written before that I was thinking about wether I should set up myself a business by selling data visualizations. As you know I did some little things in that area and thats also why I might be more susceptible to what Greg Egan is doing and might do with his skills. In particular if I would start such a business he would be a business competitor - at least if Azimuth needs some visualizations. Since he is very fast and often also very good and since he hasn't yet even asked for money (as far as I understood?) and since there is no big need for much more visualizations I dare say that I see no chances to sell visualizations to Azimuth. As a matter of fact and last but not least Greg Egan is not the only example that is I meanwhile think by comparisions that I am way too slow and that I couldn't keep up a business by selling visualizations.

    Comment Source:> It was an irrelevant digression. Let’s stop here. I was thinking wether I should let you have that final say, but I can't, because this is not an irrelevant digression! That is I think it is very connected to the question of how Azimuth will work on in the future. I had written before that I was thinking about wether I should set up myself a business by selling data visualizations. As you know I did some little things in that area and thats also why I might be more susceptible to what Greg Egan is doing and <em>might</em> do with his skills. In particular if I would start such a business he would be a business competitor - at least if Azimuth needs some visualizations. Since he is very fast and often also very good and since he hasn't yet even asked for money (as far as I understood?) and since there is no big need for much more visualizations I dare say that I see no chances to sell visualizations to Azimuth. As a matter of fact and last but not least Greg Egan is not the only example that is I meanwhile think by comparisions that I am way too slow and that I couldn't keep up a business by selling visualizations.
  • 25.
    nad
    edited June 2014

    I meanwhile think by comparisions that I am way too slow and that I couldn’t keep up a business by selling visualizations.

    apart from this it seems the visualizations which I so far offered are not needed.

    Comment Source:> I meanwhile think by comparisions that I am way too slow and that I couldn’t keep up a business by selling visualizations. apart from this it seems the <a href="http://www.azimuthproject.org/azimuth/show/Does+global+warming+lag+or+lead+a+rise+in+greenhouse+gas+concentration%3F">visualizations which I so far offered</a> are not needed.
  • 26.
    edited June 2014

    I would like to do a blog article about this sometime, actually. I just worked a bit on the spelling, grammar and formatting of that page, a tiny step toward that goal.

    You emailed me a figure that's supposed to show a 2-year periodicity of something. It was too small to see. If I can get a big version, I can put it on the page:

    I've never heard of a 2-year periodicity in the Earth's climate. So, there could either be:

    1. a bug in your program

    2. a real effect I hadn't heard of, or

    3. errors in the standard databases

    4. a new discovery

    (in order of increasing interestingness).

    Comment Source:I would like to do a blog article about this sometime, actually. I just worked a bit on the spelling, grammar and formatting of that page, a tiny step toward that goal. You emailed me a figure that's supposed to show a 2-year periodicity of something. It was too small to see. If I can get a big version, I can put it on the page: * [Does global warming lag or lead a rise in greenhouse gas concentrations?](http://www.azimuthproject.org/azimuth/show/Does+global+warming+lag+or+lead+a+rise+in+greenhouse+gas+concentration%3F) I've never heard of a 2-year periodicity in the Earth's climate. So, there could either be: 1. a bug in your program 1. a real effect I hadn't heard of, or 1. errors in the standard databases 1. a new discovery (in order of increasing interestingness).
  • 27.
    nad
    edited June 2014

    I would like to do a blog article about this sometime, actually. I just worked a bit on the spelling, grammar and formatting of that page, a tiny step toward that goal.

    Thanks. Spelling and grammar is very welcome. Formatting is not yet so necessary since this (by far) is not a final version and I might change things, so your effort may not last.

    You emailed me a figure that’s supposed to show a 2-year periodicity of something. It was too small to see. If I can get a big version, I can put it on the page.

    I think there is a 2 year periodicity in the global surface temperatures anomalies (precisely: global Combined land CRUTEM4 and marine sea surface temperature (SST) anomalies from HadSST3, see Kennedy et al., 2011 temperature anomalies. ) I had sent you the image in the normal way, as I alsways do but I had lately quite some problems with my email I haven't yet figured out what to do about it, so frankly I don't know now how to send you the image. I currently think about wether I should upload it somewhere else and where, so that it can be linked to.

    I’ve never heard of a 2-year periodicity in the Earth’s climate. So, there could either be:

    1. a bug in your program
    2. a real effect I hadn’t heard of, or
    3. errors in the standard databases
    4. a new discovery

    Yep. Concerning 1): I hope it's not a bug in the program but I had already found one...however the temperature curve looks roughly the same in shape as the others in the internet, which I had looked at and it is not so easy to get a 2 year periodicity as a bug, so I think at least this 2 year periodicity is rather not a bug, but one can never exclude that there are bugs. 2): That's what I think is most likely. 3) I hope not since as I understood this is offical HadCrut data. However since this is some kind of averaged data it can't be fully excluded that this feature is due to some averaging routine. 4) As said I think 2) is most likely, since I find that 2 year periodicity rather obvious, actually I find it so obvious that I would find it sort of a bit ominous if this hasn't been seen before.

    Comment Source:>I would like to do a blog article about this sometime, actually. I just worked a bit on the spelling, grammar and formatting of that page, a tiny step toward that goal. Thanks. Spelling and grammar is very welcome. Formatting is not yet so necessary since this (by far) is not a final version and I might change things, so your effort may not last. >You emailed me a figure that’s supposed to show a 2-year periodicity of something. It was too small to see. If I can get a big version, I can put it on the page. I think there is a 2 year periodicity in the global surface temperatures anomalies (precisely: <a href="http://www.cru.uea.ac.uk/cru/data/temperature/">global Combined land CRUTEM4 and marine sea surface temperature (SST) anomalies from HadSST3, see Kennedy et al., 2011 temperature anomalies. </a>) I had sent you the image in the normal way, as I alsways do but I had lately quite some problems with my email I haven't yet figured out what to do about it, so frankly I don't know now how to send you the image. I currently think about wether I should upload it somewhere else and where, so that it can be linked to. > I’ve never heard of a 2-year periodicity in the Earth’s climate. So, there could either be: >1. a bug in your program 2. a real effect I hadn’t heard of, or 3. errors in the standard databases 4. a new discovery Yep. Concerning 1): I hope it's not a bug in the program but I had already found one...however the temperature curve looks roughly the same in shape as the others in the internet, which I had looked at and it is not so easy to get a 2 year periodicity as a bug, so I think at least this 2 year periodicity is rather not a bug, but one can never exclude that there are bugs. 2): That's what I think is most likely. 3) I hope not since as I understood this is offical HadCrut data. However since this is some kind of averaged data it can't be fully excluded that this feature is due to some averaging routine. 4) As said I think 2) is most likely, since I find that 2 year periodicity rather obvious, actually I find it so obvious that I would find it sort of a bit ominous if this hasn't been seen before.
  • 28.
    edited June 2014

    I'm beginning to feel the effect is real, so I think it would be great if you could write a short blog post about it for Azimuth. If you want to write something simple, you could just include a nice graph, explain how you got the data to create that graph, and ask what's going on. After all, this 2-year periodicity is interesting quite separately from the original question of whether greenhouse gas concentrations lead or lag temperature rises.

    This image is too small for me to see well, but otherwise it's nice:

    I think it would be even nicer if you left out the colored dots.

    I had sent you the image in the normal way, as I always do but I had lately quite some problems with my email I haven’t yet figured out what to do about it, so frankly I don’t know now how to send you the image. I currently think about wether I should upload it somewhere else and where, so that it can be linked to.

    You had included the image in the email itself, instead of creating an "attachment". If you can send it as an attachment, or put it anywhere and tell me the URL, that would be great.

    Comment Source:I'm beginning to feel the effect is real, so I think it would be great if you could write a short blog post about it for Azimuth. If you want to write something simple, you could just include a nice graph, explain how you got the data to create that graph, and ask what's going on. After all, this 2-year periodicity is interesting quite separately from the original question of whether greenhouse gas concentrations lead or lag temperature rises. This image is too small for me to see well, but otherwise it's nice: <a href = "http://www.randform.org/blog/?p=5572"> <img src = "http://www.randform.org/blog/wp-content/2014/06/TempAnom2Year450.jpg" alt = ""/> </a> I think it would be even nicer if you left out the colored dots. > I had sent you the image in the normal way, as I always do but I had lately quite some problems with my email I haven’t yet figured out what to do about it, so frankly I don’t know now how to send you the image. I currently think about wether I should upload it somewhere else and where, so that it can be linked to. You had included the image in the email itself, instead of creating an "attachment". If you can send it as an attachment, or put it _anywhere_ and tell me the URL, that would be great.
  • 29.
    edited June 2014

    I'm going to continue this conversation on another thread, since it's no longer about Dara's offer.

    Comment Source:I'm going to continue this conversation on [another thread](http://forum.azimuthproject.org/discussion/1375/quasibiennial-oscillation/?Focus=11132#Comment_11132), since it's no longer about Dara's offer.
Sign In or Register to comment.