|
luis@11
|
1
|
|
luis@11
|
2
|
|
luis@11
|
3 \subsection{Software Development}
|
|
luis@11
|
4
|
|
luis@11
|
5 Our survey shows that 66,6\% of researchers use more than one OS. Linux isn’t
|
|
luis@11
|
6 used as a single operating system by anybody, which seems to indicate
|
|
luis@11
|
7 that most Linux users use it for multi-platform developing or for
|
|
luis@11
|
8 specific software needs. SuperCollider, Android SDK/NDK, NET, PRAAT (Speech Researcher), CUDA-C
|
|
luis@11
|
9 (GPU Programming), Clojure, Presentation, R.
|
|
luis@11
|
10
|
|
luis@11
|
11 Most researchers (56\%) use version control. This kind of system is
|
|
luis@11
|
12 more widely used by PHD students and Postdocs/Research
|
|
luis@11
|
13 Assistants. When asked for what kind of technologies were used, SVN
|
|
luis@11
|
14 (14) and GIT (7) were the most popular systems. CVS (5) and Mercurial
|
|
luis@11
|
15 (4) were the other available options. Many users use more than one of
|
|
luis@11
|
16 these systems simultaneously.
|
|
luis@11
|
17
|
|
luis@11
|
18 When asked for the usage of code hosting services, 52\% of the researchers
|
|
luis@11
|
19 said their code stayed in their computers. The most used third-party
|
|
luis@11
|
20 source code hosting services is SourceForge (6 users). 10 users are
|
|
luis@11
|
21 using university source code version control tools. \textit{remove
|
|
luis@11
|
22 numbers, only leave percentages}
|
|
luis@11
|
23
|
|
luis@11
|
24 27 users do not produce or maintain software. 16 do, while 11 did not
|
|
luis@11
|
25 answer. \textit{percentages\ldots}
|
|
luis@11
|
26
|
|
luis@11
|
27 Most users (57\%) do not plan to make any software available.
|
|
luis@11
|
28
|
|
luis@11
|
29 When asked “Do you develop any software that you do not intend to
|
|
luis@11
|
30 publish?”, 52\% of the users answered no. Possible commercial use is
|
|
luis@11
|
31 the main justification given for not publishing the software at this
|
|
luis@11
|
32 point.
|
|
luis@11
|
33
|
|
luis@11
|
34 \subsection{Reproducible Research}
|
|
luis@11
|
35
|
|
luis@11
|
36 Most researchers (56\%) acknowledge they don't take the necessary
|
|
luis@11
|
37 steps to ensure sustainable and reproducible research. Many do not
|
|
luis@11
|
38 understand the concept of reproducible research. By analyzing this
|
|
luis@11
|
39 accordingly to the current position, we can see that PhD students are
|
|
luis@11
|
40 the ones that are less aware of the importance of reproducible
|
|
luis@11
|
41 research (even the ones that are almost finishing their PhD).
|
|
luis@11
|
42
|
|
luis@11
|
43 Many of the researchers that ensure they do the steps necessary to
|
|
luis@11
|
44 reproducibility say they only give the code and/or data to interested
|
|
luis@11
|
45 researchers. Some researchers also say that they publish their code in
|
|
luis@11
|
46 their own pages. At the same time, there are indications that this
|
|
luis@11
|
47 procedure can lead to unsustainability itself. Many researchers
|
|
luis@11
|
48 complain about the amount of time and/or complexity of making research
|
|
luis@11
|
49 reproducible. Also many of them make only parts of their work
|
|
luis@11
|
50 available. Some researchers also complain about copyright issues in
|
|
luis@11
|
51 releasing data.
|
|
luis@11
|
52
|
|
luis@11
|
53 Many researchers do not understand the full concept of
|
|
luis@11
|
54 reproducibility. Some assume that explaining the algorithm and the
|
|
luis@11
|
55 tools used is enough for other researchers to be able to reproduce
|
|
luis@11
|
56 their results. Finally, some typical (but not widely admitted) answers
|
|
luis@11
|
57 justify the decision not to embrace reproducibility due to messy code
|
|
luis@11
|
58 or code/data protection:
|
|
luis@11
|
59
|
|
luis@10
|
60
|
|
luis@10
|
61 %%% Local Variables:
|
|
luis@10
|
62 %%% mode: latex
|
|
luis@10
|
63 %%% TeX-master: "cannam"
|
|
luis@10
|
64 %%% End:
|
