8 ms·

Relevant repository - https://github.com/achael/eht-imaging https://github.com/achael/eht-imaging

by ilarum 8y ago

darkpuma 8y ago
> https://github.com/achael/eht-imaging/graphs/contributors https://github.com/achael/eht-imaging/graphs/contributors Huh, I wonder how accurate this is. All the code is beyond me in any case, I'm in no position to judge the relative value of any of it.
- telesilla 8y ago
  This is deeply offensive. I'm the owner and founder of my company and I haven't written any meaningful code for our core products in 10 years. Our github repo has barely a scratch from me in it. Does this make my work, my hard long hours in managing my team and designing the product, worthless?
  - nightski 8y ago
    Honestly I'd take satisfaction in what I had built. As a leader I'd greatly enjoy giving credit to the team who committed to my vision and made it a reality.
  - vertis 8y ago
    It's offensive and frustrating. I don't understand why people have to try and pull holes in this kind of celebration like there is some conspiracy to promote Katie at the cost of others. If I read it right, she mentioned and praised her team as well.
- govg 8y ago
  I'm not sure what the idea behind posting this is? In a large scale project rarely does someone high up the command chain do any of the "grunt" work (i.e. programming). Also, by training she appears to be dealing more with theory than the actual implementation aspect of this all.
  - killjoywashere 8y ago
    ok, so just so everyone is clear, the point between parent and grandparent is Andrew Chael seems to have written a hell of a lot more code than Katie Bouman, for a lot longer * achael 566 commits 850,275 ++ 131,044 -- * klbouman 90 commits 2,410 ++ 1,265 -- However, at least at the level of reading the commit messages, Katie's are pretty math heavy: "fixed bug in the fake briggs weighting" "starting to fix chirp problems with polrep" "made it possible to do a min uv cut on closure phase when adding it a..." While Andrew's lean frequently toward code maintenance: "updated some docstrings in imager_utils" "moved imgsum to plotting.summary_plots" "modified README" That said, Andrew and others seem to have pretty good insight too.
    - ilarum 8y ago
      https://github.com/achael/eht-imaging/commit/886b07b8a00d142b23a70537511c79bef85e0042 https://github.com/achael/eht-imaging/commit/886b07b8a00d142... one commit with 524,306 additions. adding a model.
    - thecatspaw 8y ago
      she could have written 0 lines and be "the woman behind the picture". Writing code is easy, figuring out complex algorithms is something very different, and does not require coding knowledge
    - govg 8y ago
      True, but that need not be of high importance when assigning credit. Especially in projects that involve a lot of implementation, the high level algorithms and techniques are developed by someone who might be far removed from actually writing code, and that aspect is outsourced to someone much more competent at programming. A lot of PhD students / professors do not have the talent or have not developed the thought process to write complex code, simply because their focus has been on other things.
      - dagw 8y ago
        Exactly. One project I worked on a while back involved me taking a big pile of very clever and complicated Matlab code and rewriting it in python in a way that made it easy to use from other projects and able to read a couple of additional input file formats. If you just look at the commits on that project it would look like I was responsible for 90+% of the entire thing, while in truth I was basically just doing transcribing and cleaning and had basically nothing to do with any of the difficult parts.
- ilarum 8y ago
  Number of lines is not an indicator of contribution. The reason I posted this repo was so that folks curious about the programming behind this science could have their fill. I am just happy to see an academic project using good version control practices released under a proper license.
black_puppydog 8y ago
It's also in the links in that repo, but may be more interesting to some folks here: Katie's paper on VLBI reconstruction: https://arxiv.org/abs/1512.01413 https://arxiv.org/abs/1512.01413 This is how I learned about the topic and I think it's well suited for computery folks, since it was published in CVPR.
- pbhjpbhj 8y ago
  There are 6 authors for that paper, the way you linked it suggested it's solely by her; are the other authors listed honorarily? Similarly the title suggests she worked alone on the project. Which seems exceedingly unlikely given the need for telescope time and computing time and the wide range of disciplines I imagine the project covers ... did she work alone. That must be almost unique in experimentalism nowadays?
  - black_puppydog 8y ago
    It depends a bit on the scientific field, but in mine (computer vision, which this was also published in) the first author is the "main" author, usually doing most if not all the "ground work", then you have collaborators and co-authors, and finally the supervisors (in the case of the PhD student that she was at the time). It is becoming more common to see papers with "Author1, Author2, Author3..." lists, where the "*" authors count with equal contribution. This is important for attribution and the infamous metrics that funding often depends on, but it's not the case here. So I'm certain those authors did their part, so maybe yes, I should have linked this as "Bouman et al." but I wouldn't expect this to be six equal contributions either. That all being said, she's certainly standing on the shoulders of a pyramid of giants there. Edit: to the people downvoting the parent, maybe explain? I didn't take this as a bad faith comment. It can be genuinely confusing to someone who doesn't know the ins and outs of academic attribution...
    - SiempreViernes 8y ago
      I'd argue that in practice science is more of a pile of giants.
toomanybeersies 8y ago
That code definitely wouldn't pass code review by me. Terrible commit messages too.
- noonespecial 8y ago
  Once you post your impossibly distant black hole images, I'm sure they'll consider your pull request cleaning it up...
  - toomanybeersies 8y ago
    For those who didn't realise above, I was being flippant. On a slightly serious note though, I wonder how much productivity is lost in the scientific community due to poorly written and documented code? I've heard stories of 40 year old Fortran code written by long deceased professors that was written to crunch physics numbers or whatever, and when it's come time to modify or add to it, nobody can make head nor tail of it and they have to write it from scratch. There's a reason why in the non-academic world we have coding standards and code review. Code isn't written in a bubble, other people will look at it and work on it. That's not to belittle or criticise the work done in the slightest. Cleanliness of code is orthogonal to functionality. You can have beautifully written, clean and documented code that doesn't do what it's meant to, and likewise you can have a complete mess of code that performs some genius function perfectly.
    - gentaro 8y ago
      I agree with you that the scientific community is way behind industry standards, but the reason for that is much less of their code is actually designed for reuse. The overwhelming majority of their work is just "let me try writing this code and see what results I get." Industry professionals are forced to take the approach of "I need to write this code to be as maintainable and flexible as possible" because they have no idea what the business is going to want next and generally have no set timeframe for how long they may have to maintain any particular project.
      - 8y ago

Relevant repository - https://github.com/achael/eht-imaging https://github.com/achael/eht-imaging

by ilarum 8y ago

Relevant repository - https://github.com/achael/eht-imaging https://github.com/achael/eht-imaging

> https://github.com/achael/eht-imaging/graphs/contributors https://github.com/achael/eht-imaging/graphs/contributors Huh, I wonder how accurate this is. All the code is beyond me in any case, I'm in no position to judge the relative value of any of it.

telesilla 8y ago
This is deeply offensive. I'm the owner and founder of my company and I haven't written any meaningful code for our core products in 10 years. Our github repo has barely a scratch from me in it. Does this make my work, my hard long hours in managing my team and designing the product, worthless?
- nightski 8y ago
  Honestly I'd take satisfaction in what I had built. As a leader I'd greatly enjoy giving credit to the team who committed to my vision and made it a reality.
- vertis 8y ago
  It's offensive and frustrating. I don't understand why people have to try and pull holes in this kind of celebration like there is some conspiracy to promote Katie at the cost of others. If I read it right, she mentioned and praised her team as well.
govg 8y ago
I'm not sure what the idea behind posting this is? In a large scale project rarely does someone high up the command chain do any of the "grunt" work (i.e. programming). Also, by training she appears to be dealing more with theory than the actual implementation aspect of this all.
- killjoywashere 8y ago
  ok, so just so everyone is clear, the point between parent and grandparent is Andrew Chael seems to have written a hell of a lot more code than Katie Bouman, for a lot longer * achael 566 commits 850,275 ++ 131,044 -- * klbouman 90 commits 2,410 ++ 1,265 -- However, at least at the level of reading the commit messages, Katie's are pretty math heavy: "fixed bug in the fake briggs weighting" "starting to fix chirp problems with polrep" "made it possible to do a min uv cut on closure phase when adding it a..." While Andrew's lean frequently toward code maintenance: "updated some docstrings in imager_utils" "moved imgsum to plotting.summary_plots" "modified README" That said, Andrew and others seem to have pretty good insight too.
  - ilarum 8y ago
    https://github.com/achael/eht-imaging/commit/886b07b8a00d142b23a70537511c79bef85e0042 https://github.com/achael/eht-imaging/commit/886b07b8a00d142... one commit with 524,306 additions. adding a model.
  - thecatspaw 8y ago
    she could have written 0 lines and be "the woman behind the picture". Writing code is easy, figuring out complex algorithms is something very different, and does not require coding knowledge
  - govg 8y ago
    True, but that need not be of high importance when assigning credit. Especially in projects that involve a lot of implementation, the high level algorithms and techniques are developed by someone who might be far removed from actually writing code, and that aspect is outsourced to someone much more competent at programming. A lot of PhD students / professors do not have the talent or have not developed the thought process to write complex code, simply because their focus has been on other things.
    - dagw 8y ago
      Exactly. One project I worked on a while back involved me taking a big pile of very clever and complicated Matlab code and rewriting it in python in a way that made it easy to use from other projects and able to read a couple of additional input file formats. If you just look at the commits on that project it would look like I was responsible for 90+% of the entire thing, while in truth I was basically just doing transcribing and cleaning and had basically nothing to do with any of the difficult parts.
ilarum 8y ago
Number of lines is not an indicator of contribution. The reason I posted this repo was so that folks curious about the programming behind this science could have their fill. I am just happy to see an academic project using good version control practices released under a proper license.

black_puppydog 8y ago

It's also in the links in that repo, but may be more interesting to some folks here: Katie's paper on VLBI reconstruction: https://arxiv.org/abs/1512.01413 https://arxiv.org/abs/1512.01413 This is how I learned about the topic and I think it's well suited for computery folks, since it was published in CVPR.

pbhjpbhj 8y ago
There are 6 authors for that paper, the way you linked it suggested it's solely by her; are the other authors listed honorarily? Similarly the title suggests she worked alone on the project. Which seems exceedingly unlikely given the need for telescope time and computing time and the wide range of disciplines I imagine the project covers ... did she work alone. That must be almost unique in experimentalism nowadays?
- black_puppydog 8y ago
  It depends a bit on the scientific field, but in mine (computer vision, which this was also published in) the first author is the "main" author, usually doing most if not all the "ground work", then you have collaborators and co-authors, and finally the supervisors (in the case of the PhD student that she was at the time). It is becoming more common to see papers with "Author1, Author2, Author3..." lists, where the "*" authors count with equal contribution. This is important for attribution and the infamous metrics that funding often depends on, but it's not the case here. So I'm certain those authors did their part, so maybe yes, I should have linked this as "Bouman et al." but I wouldn't expect this to be six equal contributions either. That all being said, she's certainly standing on the shoulders of a pyramid of giants there. Edit: to the people downvoting the parent, maybe explain? I didn't take this as a bad faith comment. It can be genuinely confusing to someone who doesn't know the ins and outs of academic attribution...
  - SiempreViernes 8y ago
    I'd argue that in practice science is more of a pile of giants.

toomanybeersies 8y ago

That code definitely wouldn't pass code review by me. Terrible commit messages too.

noonespecial 8y ago
Once you post your impossibly distant black hole images, I'm sure they'll consider your pull request cleaning it up...
- toomanybeersies 8y ago
  For those who didn't realise above, I was being flippant. On a slightly serious note though, I wonder how much productivity is lost in the scientific community due to poorly written and documented code? I've heard stories of 40 year old Fortran code written by long deceased professors that was written to crunch physics numbers or whatever, and when it's come time to modify or add to it, nobody can make head nor tail of it and they have to write it from scratch. There's a reason why in the non-academic world we have coding standards and code review. Code isn't written in a bubble, other people will look at it and work on it. That's not to belittle or criticise the work done in the slightest. Cleanliness of code is orthogonal to functionality. You can have beautifully written, clean and documented code that doesn't do what it's meant to, and likewise you can have a complete mess of code that performs some genius function perfectly.
  - gentaro 8y ago
    I agree with you that the scientific community is way behind industry standards, but the reason for that is much less of their code is actually designed for reuse. The overwhelming majority of their work is just "let me try writing this code and see what results I get." Industry professionals are forced to take the approach of "I need to write this code to be as maintainable and flexible as possible" because they have no idea what the business is going to want next and generally have no set timeframe for how long they may have to maintain any particular project.
    - 8y ago