ASCL.net

Astrophysics Source Code Library

Making codes discoverable since 1999

Tweets from and about the code sharing session at AAS223

The code sharing crowd took over the AAS Twitter feed, it seems, during the Special Session on code sharing at AAS 223. Bottom up is the best way to read these, as the most recent tweet is on the top, and please note they aren't strictly in order of occurrence and I likely missed some (there were so many!). I'm happy to add those I missed if someone tells me about them. Thanks to all those who tweeted throughout the session!

  1. ASCL ‏@asclnet 10 Jan
    @    shaka_lulu     I keep a list of articles of possible interest to #astroCodeShare folks here: http://asterisk.apod.com/viewtopic.php?f=35&t=21544 …. #aas223
  4. Nuria Lorente ‏@NoTruerAlien 7 Jan
    @    augustmuench     @bruceberriman Absolutely, but NOT releasing code also comes at a price, which is often forgotten. #aas223 #astroCodeShare
  5. Zach Pace ‏@zpacefromspace 7 Jan
    Just got finished with an awesome breakout session at #aas223 on code sharing. The moral: your code may be crap, but release it anyway!
  6. Nuria Lorente ‏@NoTruerAlien 7 Jan
    Morin et al: Informative paper on Sw licensing for Scientist-Programmer. MT @augustmuench: http://bit.ly/QlZKDP #astroCodeShare #aas223
  7. Chrissy Madison ‏@cmmadiso 7 Jan
    See. It happens! RT @bathompso: Pulling a @cmmadiso: my phone has 1% battery after the #astroCodeShare session. #AAS223
  8. Ben Thompson ‏@bathompso 7 Jan
    Pulling a @cmmadiso: my phone has 1% battery after the #astroCodeShare session. #AAS223
  10. Adrian Price-Whelan ‏@a_p_w 7 Jan
    RE: writing quick/dirty code to get papers out. "Weeks of programming can save you hours of planning." #aas223
  12. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Hanisch: there should also be a prize for software, esp since Webber prize is for hardware only #aas223 #astroCodeShare
  13. August Muench ‏@augustmuench 7 Jan
    .@AlexaVillaume note: that paper is for software. licensing of *data/papers* is distinct but VERY important thing. #aas223 #astroCodeShare
  15. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Licensing: BSD or MIT and forget about it- but we should discuss it more as community - @davidwhogg #astroCodeShare #aas223 cc @jonmccann
  16. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    @    jonmccann     crap I had to answer an email and missed license discussion. Maybe check #astroCodeShare tag if someone else got it
  17. Christopher Hanley ‏@chanley 7 Jan
    @    eblur27     Projects should include a citations file in repo right next to LICENCE.txt and README. Make it easier to be cidted #astroCodeShare
  18. August Muench ‏@augustmuench 7 Jan
    so @aaccomazzi brings us back to licensing: unlicenced code is the WORST. @davidwhogg echos this point #aas223 #astroCodeShare
  19. Lia Corrales ‏@eblur27 7 Jan
    Hey #astroCodeShare, I still want to know how I should cite software I get through github. Could help with fluid contributor lists #aas223
  20. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Something to be gleaned from size of room v attendance v perceived necessary size of room & how code is valued. #aas223 #astroCodeShare
  21. Ben Cook ‏@bacook17 7 Jan
    Reference in #aas223 code sharing session. http://wssspe.researchcomputing.org.uk
  22. Kelle Cruz ‏@kellecruz 7 Jan
    #    AAS223     lots of different ways to share code...but I really want to spend time and energy making it expected & common practice.
  23. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Comment in back: u make good science by making good investments- invest in quality code by encouraging code sharing #aas223 #astroCodeShare
  24. Kelle Cruz ‏@kellecruz 7 Jan
    #    aas223     I care less about how we data & code share. tech will work itself out. I want to make it a *requirement* for funds and publications.
  25. Alexa Villaume ‏@AlexaVillaume 7 Jan
    A mortifying story of a misplaced 2 in a program causing 8 years of research going down the drain. Share your code. It'll be ok. #aas223
  26. Meredith Rawls ‏@merrdiff 7 Jan
    .@kellecruz If it's easy to share code & get credit, we'll do it. Reminds me of this: http://theoatmeal.com/comics/game_of_thrones … #aas223 #astroCodeShare
  27. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    There are more ppl in this room than were in the Kepler session I attended yesterday. #aas223 #astroCodeShare
  28. August Muench ‏@augustmuench 7 Jan
    @    jradavenport     225: Panel: MMMMMM Q: FMMFFFMM A: all M except 1 comment by F audience #astroCodeShare
  29. August Muench ‏@augustmuench 7 Jan
    So @eteq is gonna drop the mic: papers have fixed author lists. software authorship if fluid and grows. Et tu, ADS? #astroCodeShare #aas223
  30. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Reply: if it goes on arXiv you can never update contributor list, so subsequent contributors don't get credit #aas223 #astroCodeShare
  31. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Prsa: would be great if announcement of code went up on arXiv (I think they often do as release papers, e.g. emcee) #aas223 #astroCodeShare
  32. August Muench ‏@augustmuench 7 Jan
    interesting point: Montage built under contract to NASA; astropy built by cats, hosted on a cat based website #astroCodeShare #aas223
  33. Meredith Rawls ‏@merrdiff 7 Jan
    Andrej Prsa implores everyone in #astroCodeShare session to post code on astro-ph every time you submit a paper. #aas223
  34. Meredith Rawls ‏@merrdiff 7 Jan
    My advisor has said not to "waste time" writing generalized code; contradicting this is troubling. Mixed messages. #aas223
     #astroCodeShare
  35. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Cost to sharing: making code useable by anyone req more time than just making it work for you then publishing w it #aas223 #astroCodeShare
  36. Meredith Rawls ‏@merrdiff 7 Jan
    Recurring theme of how do I maximize research productivity and make my code useable for others? Not an easy Q. #AAS223 #astroCodeShare
  37. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    That is, proprietary data sometimes equates to leverage- there is prob some analogy in code community- @davidwhogg #aas223 #astroCodeShare
  38. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    panelists doubt ppl are being hired for a specific code as opposed to skill, but must be analogy w proprietary data #aas223 #astroCodeShare
  39. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Q: how do we reward ppl in ways that don't req keeping code proprietary? As in, ppl get hired bc they have the code #aas223 #astroCodeShare
  40. Lia Corrales ‏@eblur27 7 Jan
    Sad I'm missing #astroCodeShare, but reports of a packed room and massive twitter coverage are letting me stay comfy in COS session #aas223
  41. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Besides, every little thing you think no one else needs- *someone* will prob find it useful #aas223 #astroCodeShare
  42. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    A: do what you need, if no one else needs it then that's fine, you haven't made anyone's life worse - @davidwhogg #aas223 #astroCodeShare
  43. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Q: what's the balance bt needing to make code work for yourself vs making it useful for everyone always? #aas223 #astroCodeShare
  44. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Comment in back: u make good science by making good investments- invest in quality code by encouraging code sharing #aas223 #astroCodeShare
  45. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Katz: not really, unlikely beyond a few years' horizon at a time #aas223 #astroCodeShare
  46. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    No long term stewardship of code like there is for results (i.e. pubs)- does NSF have plans for that? - @davidwhogg #aas223 #astroCodeShare
  47. Ben Thompson ‏@bathompso 7 Jan
    . @kellecruz starting off the #astroCodeShare question session strong. Why no AAS reps here? #AAS223
  48. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Do you use other people's codes? Do you modify them or use them as is? #astroCodeShare #aas223
  49. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Do you share code? If not, why not? #astroCodeShare #aas223
  51. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Benefits: perceived priority on work, visibility & good will, citations, bug-catching, and moral high ground - DWH #astroCodeShare #aas223
  52. Alexa Villaume ‏@AlexaVillaume 7 Jan
    Releasing code establishes priority and good will. Benefit from bug catching. Also, you get to be smug. #aas223
  53. Laura Watkins ‏@laurawatkins_ 7 Jan
    @    davidwhogg    : if you're not embarrassed by the code you released, you released it too late. #aas223
  54. Ian Paul Freeley ‏@ianpaulfreeley 7 Jan
    If your not embarrassed by your code/website, you launched too late--Hogg #aas223
  55. Ben Thompson ‏@bathompso 7 Jan
    If you’re not embarrassed by your code, you’re releasing it too late #AAS223 #astroCodeShare
  56. Meredith Rawls ‏@merrdiff 7 Jan
    .@davidwhogg debunks cons to code sharing. Only real cost is email & support requests. He knows of NO example of being scooped. #aas223
  57. August Muench ‏@augustmuench 7 Jan
    hogg: "if you're not embarrassed by the code/website you put out there then you put it out there too late." so good. #astroCodeShare #aas223
  58. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Cost: embarrassment! You know your code is crap, but if yr not embarrassed you released too late. -@davidwhogg #aas223 #astroCodeShare
  59. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Costs: getting scooped? @davidwhogg knows of no cases of scooping caused by *release of code* #astroCodeShare #aas223
  60. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    All papers, grant writing, etc - not just code - are developed out in the open since 2005. - @davidwhogg #astroCodeShare #aas223
  61. Matthew Turk ‏@powersoffour 7 Jan
    @    augustmuench     Not all good or new software is developed using github. Platforms should be transcended by applications. #astroCodeShare
  62. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    And boom, @davidwhogg right on time. Also, who mic'd him? #astroCodeShare #aas223
  63. Laura Watkins ‏@laurawatkins_ 7 Jan
    +1 MT @augustmuench "and this fact terrifies me because we have no idea collectively what sharing should look like. #astroCodeShare #aas223”
  64. August Muench ‏@augustmuench 7 Jan
    and this fact terrifies me because education -- we have no idea collectively what sharing should look like. #astroCodeShare #aas223
  65. August Muench ‏@augustmuench 7 Jan
    Who enforces software/data sharing in astronomy? YOU DO! WE DO! PEER REVIEW DOES! not nsf/nasa. #aas223 #astroCodeShare
  66. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Its UP TO YOU to include good data management plan as part of panel reviews. The community must enforce importance. #aas223 #astroCodeShare
  67. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Data management plans in NSF proposals are required to detail how results/data/software will be shared. - Katz #astroCodeShare #aas223
  68. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    NSF policy for sharing research results: supposed to share not only the data and the results but the software #astroCodeShare #aas223
  69. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    NSF does include "products" in addtn to pubs in bio sketches, but could be better abt following up on code release #AAS223 #astroCodeShare
  70. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Do we have policies that mandate code release in conjunction w publication or receipt of fed funds? #aas223 #astroCodeShare
  71. August Muench ‏@augustmuench 7 Jan
    Software that enables all this new software: Github, Travis, Sphinx, Jenkins. #aas223 #astroCodeShare
  72. August Muench ‏@augustmuench 7 Jan
    Agreed. RT @kellecruz: .@augustmuench we need to make data/code sharing requirements part of AAS journal policy. those two things. #aas223
  73. Ben Thompson ‏@bathompso 7 Jan
    Testing code is an important part of code sharing. #aas224 session? #AAS223
  74. Kelle Cruz ‏@kellecruz 7 Jan
    .@augustmuench we need to make data/code sharing requirements part of AAS journal policy. those two things. #aas223
  75. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    If you build it, they will code - Tollerud #aas223 #astroCodeShare
  76. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Need infrastructure, a few software ppl to do housekeeping, let scientists do whatev & set expectations - Tollerud #astroCodeShare #aas223
  77. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Most ppl who have contributed code to AstroPY have never met each other - all via @github - Tollerud #astroCodeShare #aas223
  78. Ben Thompson ‏@bathompso 7 Jan
    Almost 60 people (who have not met) have all worked together to build @astropy #astroCodeShare #AAS223
  79. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    AstroPY: a python library for and by astronomers, developed by self-herding astronomers since 2011 - Tollerud #astroCodeShare #aas223
  80. August Muench ‏@augustmuench 7 Jan
    I hoping that we see some cool diffs between the @astropy and montage *support* networks in the open discussion in #astroCodeShare #aas223
  81. August Muench ‏@augustmuench 7 Jan
    . @merrdiff "research objects" is I think the new age terminology. #aas223 #astroCodeShare
  82. David Morrison ‏@drmorr0 7 Jan
    @    merrdiff     Best advice I have: learn to use Git (or SVN, if you must), and use it for every single piece of code you write. #astroCodeShare
  84. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    I have used this exact cat herding graphic in Erik Tollerud's talk in an LSST talk hehe #aas223 #astroCodeShare
  85. August Muench ‏@augustmuench 7 Jan
    the @astropy project -- cat herding software development from @eteq at #astroCodeShare #aas223
  86. Ian Paul Freeley ‏@ianpaulfreeley 7 Jan
    Damn it--tweets from code sharing session sounded cool, but I got here late and now crowd out the door. #aas223
  87. Alex Parker ‏@Alex_Parker 7 Jan
    I'm nodding so vigorously at the #astroCodeShare tweets that I might need to ice my neck later.
  89. Dr Chris Tibbs ‏@chris_tibbs 7 Jan
    Love the fact that my timeline is currently full of great tweets about code sharing and EPO #aas223
  90. Kelle Cruz ‏@kellecruz 7 Jan
    #    aas223     ok, maybe 20% women in code sharing session but still disproportionately tweeting. #interesting
  91. Alexa Villaume ‏@AlexaVillaume 7 Jan
    "I wrote my first fortran code when Apollo 12 was on the moon." #aas223
  92. August Muench ‏@augustmuench 7 Jan
    Decision to code Cloudy in C++ was partly motivated to use industry grade lang & give students real world job ops! #astroCodeShare #aas223
  93. Meredith Rawls ‏@merrdiff 7 Jan
    Learning about the CLOUDY code, but speaker has no visuals :( Jokes that the code can be opaque; "C++: write once, read never." #aas223
  94. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Complaining astros aren't comp scientists is like saying they shldn't learn math bc they aren't mathematicians #astroCodeShare #aas223
  95. Timothy Pickering ‏@te_pickering 7 Jan
    #    preach    ! RT @shaka_lulu: I'll paraphrase @mjuric here: code is to modern astronomy what calculus once was. #aas223 #astroCodeShare
  96. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    I'll paraphrase @mjuric here: code is to modern astronomy what calculus once was. #aas223 #astroCodeShare
  97. Kelle Cruz ‏@kellecruz 7 Jan
    #    aas223     could someone in the back of the code sharing session do a quick attendance & gender count? I'm in the front row...
  98. Jessica Lu ‏@jlu_astro 7 Jan
    @    kellecruz     Tell me if you figure it out!
  99. Kelle Cruz ‏@kellecruz 7 Jan
    #    aas223     code sharing room is packed! I'm curious what brought them all here...
  101. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Q: how much do you think we fail to educate our young researchers to write good code? #astroCodeShare #aas223
  102. Ben Thompson ‏@bathompso 7 Jan
    Q on why students are not educated on how to write good code (or code at all!) #AAS223
    We have all failed here.
  103. August Muench ‏@augustmuench 7 Jan
    The code base under question is Montage http://bit.ly/1aELvEz , dev'd & now volunteerly supported by IPAC scientists #aas223 #astroCodeShare
  104. August Muench ‏@augustmuench 7 Jan
    "Releasing your code comes with a price" -- @bruceberriman Hmm, let's see if this pivots to the positive+solutions! #aas223 #astroCodeShare
  105. Lucianne Walkowicz ‏@shaka_lulu 7 Jan
    Lastly, resist the pundit-technician divide. - Weiner #aas223 #astroCodeShare
  106. August Muench ‏@augustmuench 7 Jan
    I completely agree with @cloud149: a lot of our concerns about sharing code are "pseudo" or hypothetical problems. #aas223 #astroCodeShare
  107. Michelle Collins ‏@michelle_lmc 7 Jan
    We are failing to teach students how to write GOOD code in astronomy. Need to do better. Some programs in place, but not standard #aas223
  108. Laura Watkins ‏@laurawatkins_ 7 Jan
    "Do we do enough to teach our researchers how to write good code?" No. Fundamental skills but so many are left to learn alone. #aas223
  109. Kelle Cruz ‏@kellecruz 7 Jan
    #    aas223     really interesting that nearly 100% of the women in the code sharing session are tweeting...all 4 of us. #exaggerating
  110. Laura Watkins ‏@laurawatkins_ 7 Jan
    Standing room only at the code sharing session. Apparently this is more popular than anticipated (this can only be a good thing)! #aas223
  111. Meredith Rawls ‏@merrdiff 7 Jan
    Astrophysics code sharing session 225 at #aas223. Let's stop re-inventing the wheel. Our hardware is built to last; why not software?
  112. Michelle Collins ‏@michelle_lmc 7 Jan
    Oh, there are no women on the code sharing panel. Are we not sharing code? I'm currently not, but i'm here to learn how to #aas223
  113. Ben Thompson ‏@bathompso 7 Jan
    Excited for the Astronomy Code Sharing session. Wondering what to do with all my research programs. #AAS223
  114. Erik Tollerud ‏@eteq 7 Jan
    #    aas223    , Tues@2pm: talking about lesson's from @astropy on how code can be shared, along side @owlice @davidwhogg @bruceberriman @cloud149
  115. Benjamin Weiner ‏@cloud149 7 Jan
    My talk "Occupy Hard Drives" for code session Tues 2 pm #aas223 is here: http://bit.ly/1acClmg @davidwhogg @bruceberriman @owlice @eteq
  116. ADASS ‏@astroADASS 7 Jan
    Follow discussion on astronomy code sharing at the #aas223 meeting using #astroCodeShare hashtag.
  117. Benjamin Weiner ‏@cloud149 7 Jan
    Tues 2pm #aas223 I aim to provoke on astro code sharing and why we don't respect software with @davidwhogg @bruceberriman @owlice @eteq

  118. Astropy ‏@astropy 6 Jan
    At the #aas223? Don't miss Tuesday's 14:00-15:30 session on code sharing - including a talk by @eteq about @astropy!

  119. David W. Hogg ‏@davidwhogg 6 Jan
    Tues at 2 see @owlice Hanisch Teuben @cloud149 @bruceberriman Ferland Katz @eteq and me get all crazy about #code sharing at #aas223 in NH5
This entry was posted in AAS, conferences, discussion, ideas, special sessions on January 13, 2014 by Alice Allen.

Leave a Reply