But because publication is the goal, improving software isn't prioritized. #FOSDEM2020
-
-
Prikaži ovu nit
-
Similarly, education of software dev. skills aren't something in which there's a lot of investment.
#FOSDEM2020Prikaži ovu nit -
There's a trust issue too! Scientists would rather have all the code under their control for small projects. For larger projects they often go to commercial software.
#FOSDEM2020Prikaži ovu nit -
There's a tendency to not publish software, because what if people find bugs in it?!?!
#FOSDEM2020 (gasp)Prikaži ovu nit -
For academics, software development is a side gig. But it's not rewarded, no time is allocated for it, and it's low priority.
#FOSDEM2020Prikaži ovu nit -
Ex: ODMLTABLES 1 developer 0 contributers 1-10 users If the developer moves on, this project is dead.
#FOSDEM2020Prikaži ovu nit -
Ex: nest (simulator for spiking ANN models) 5 active developers 1 funded position for docs >10 active contributors This project is much more robust.
#FOSDEM2020Prikaži ovu nit -
What can us software developers do? * chat with scientists!!! * get involved in an existing project * make docs better for non-experts!!! (docs docs docs cc
@emckean) * advertise within potential user community#FOSDEM2020Prikaži ovu nit -
Make scientific projects robust and sustainable! * get more users and contributors * simplify usage and contribution * advertise
#sustainability#FOSDEM2020Prikaži ovu nit -
As scientists: * use existing open source projects * "make sure your software outlives your career" * link to other packages
#FOSDEM2020Prikaži ovu nit -
Great talk by Julia Sprenger!
#FOSDEM2020Prikaži ovu nit
Kraj razgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.