Wednesday, 22 May 2013 by Heiko Tietze and BjÃ¶rn Balazs
We conducted a large study about strength and weakness of file managers in may 2013. In this article we present first results, discuss issues and questions that occur during the study, and present the schedule for the statistical analysis.
In may 2013 we asked users about their preferences for file managers (FM). The online survey did not contain questions regarding Dolphin specifically, but addresses file managers in general to be able compare usability and user experience of different tools. Out of 1310 users that started the questionnaire, 749 finished the first part with questions about the When and What (42% drop-out) and 669 finished the second part with ratings on the chosen file manager (10% drop out). The last part with demographics was finished by 651 probands (2% drop-out).
Most participants (56%) use Dolphin (which is KDE’s default), followed by Nautilus (10%), CLI (8%), Thunar (5%), Konqueror (4%), and Krusader (3%). The predominance of KDE / Qt within the study is also shownÂ in one of the last questions about the OS. It’s a pity that we weren’t able to attract more users of other OS with the survey.
Two third of all users chose their FM deliberately. Most prioritize functionality (31%) and efficiency (18%). This result is according to developers expectations (cf. Quo vadis, Dolphin? Results from the developers study).
And at a first glance the users are satisfied with their FM: 31 report low values (1 for displeased and 2) while 627 report high ratings (5 and 6 as fully satisfied).
The rather high quote of drop-outs probably comes from too ‘strange’ questions and a more complex study design in contrast to simple ‘Like’ polls. The basic idea is that software in general is applied by a person (individual aspects were asked in the demographics section), in a certain situation, and with a specific purpose (both asked in the first part of the study). We admit that the particular situation in which a tool is being used, needs some reflection. Of course, FM’s are used to copy files, to add or remove folders, and to manage properties. But if you think about the situation in which you apply a tool you might come to the conclusion that sometimes a command line interface (CLI) fits better and sometimes it is nice to have a GUI with browsing functions. And perhaps you generally prefer an orthodox FM like Krusader because of its familiarity but use Dolphin when browsing through files.
Furthermore we received some criticism concerning the depending questions about the rating. The most simple approach for an analysis is to count ‘likes’. We believe in more elaborated methods and statistical evaluation that exceeds those introspection. In terms of usability the ISO 9241-110 ‘Ergonomic requirements for office work with visual display terminals’ defines seven different dialog principles. The question is whether or not users who expect efficiency from their chosen FM also state high values on the efficiency scale, and at the same time report low values on other scales. For instance, the ISO demands ‘learnability’ for a good usability, i.e. the usage of a dialog is guided or at least supported, or alternate interaction is proposed by tool-tips or the like. But nobody does expect those features from a CLI, which we assume to be efficient primarily. However, it is worth to discuss the norm in general as users are barely able to answer the plain (yet not simple) questions based on the wording of the norm.
Most interesting are analyses in relation to FMs, OS, or user’s personal preferences. Some ideas:
- Are there differences in the satisfaction for different FMs? And if so: On what basis?
- Do Gnome users expect more simplicity from their FM compared to KDE users?
- Which part ofÂ Dolphin needs most attention for the future development?
And some questions with a rather academic background:
- Can we predict future changes due to age effect, i.e. do elder people request different features (except from accessibility)?
- Are motifs useful to predict expectations? E.g. high values in prestige goes along with more individualization. (cf. LibreOffice user research â€“ Summary).
We will conduct those statistics within the next days and will report the results on this channel. If you have more questions in advance feel free to comment.