What Open Source Means to LibreOffice Users
Being an open source project is one of the pillars of LibreOffice. Understanding the people who not only build and maintain but also use LibreOffice is important to anyone concerned about the sustainability of open source. So we asked the community what aspects are important and present the results now.
The survey followed GitHub which ran a similar open source study about ‘the attitudes, experiences, and backgrounds of those who use, build, and maintain open source software’ earlier this year. Among the very interesting results, one question serves as the basis for us: “What do open source users value in software?”.
We asked the same question, but split the matrix into separate questions and added short explanations since our audience is likely not so technically-minded like GitHub developers.
- How important is User Experience for your?
Good user experience means functional requirements are covered completely, the program is easy to use, and it is a pleasure to work with it.
- How important is Replicability for you?
Anyone can read the source code, understand how a function computes, and use it for their own projects.
- How important is Transparency for you?
The project permits use, study, and modification of the source code without restriction, and as a normal user you are regularly informed about all aspects
- How important is it to have a Trusted Producer?
Trust in the producer (or rather publisher) means you know the organization and people for years and can rely on the product in the future
- How important is the availability of Technical Support for you?
Technical support in terms of the availability of wikis and forums, as well as the possibility to contact experts and even developers.
- How important is Stability for you?
Stability means the program functions properly and does not hang or crash on unexpected conditions.
- How important is Security for you?
Security is the protection from information theft or damage as well as from disruption or misdirection of services.
- How important is the Cost factor for you?
Open source software is usually free of charge but might have follow-up costs for training and administration, for instance.
- How important is Compatibility for you?
This terms means to load data processed with old software versions (backward compatibility) as well as the compatibility across different programs and platforms.
- How important is Customizability/Modifiability for you?
The term means you can adapt the program to your personal needs and preferences regarding its visual appearance as well as the workflow.[collapse]
Together with a couple of independent factors, the results  show where LibreOffice users put highest value in the project.
In the first questions we asked how important aspects are on a five-level Likert scale from Not at all important to Extremely important. Since we are mostly interested in a comparison with the other study, both are shown in the graphic, in gray for GitHub and black for LibreOffice results.
The results of both studies are comparable with higher appraisal of stability, UX (user experience), and compatibility where transparency, customizability, costs, and replicability are rated lower. The most striking result is that having a Trusted Producer (or publisher) is much more relevant for users of LibreOffice.
Interesting for marketing is what office tools are used regularly. This question was asked as multiple choice and figure 2 shows – on the left – how often one of the predefined answers was checked and on the right how many different tools are used in total. Not surprisingly Microsoft Word and Google Docs are used most often.
Participants had the chance to freely enter text for Other tools. AbiWord, Hancom Office, Lotus Smartsuite, Mobisystems, Polaris Office, Quattro Pro, Typora, Zoho and some text editors were mentioned but also Collabora Online is seen as a separate product.
Interesting in this context is how often an online suite is used. The answer options were presented again with a Likert scale.
And of course we wanted to know how loyal our users are and asked for how many years they have been using LibreOffice. The average is 9.3 with two peaks at 5 and 15 years (figure 3) possibly because people do not remember exactly when they started.
Next question was about expertise. As self-estimation of being an expert is sometimes not easy we rather asked whether a colleague or friend would ask to solve a problem with LibreOffice. The majority answered with Yes (n=997 vs. 262), which is an indicator that the sample is biased.
Nevertheless it is interesting if those participants who are more expert than others have a different view on the open source aspects. And the results make sense: the ‘advanced’ categories Transparency, Replicability, and Trusted Producer are valued significantly higher. The results in figure 4 (and following) shows data from LibreOffice only, the asterisks indicate significant differences based on MAONVA statistics.
LibreOffice ships two default versions, fresh and stable, where the first contains all newly introduced features while stable is the long-term and feature-frozen release. Furthermore, nightly builds allow a very early look into the development.
|Bleeding edge with nightly builds||25|
|I don’t care at all||31|
|No idea, whatever the updater recommends||120|
|Nothing else than ‘fresh’||495|
|I’m always on ‘stable’||618|
Table 2 shows that fresh and stable are nearly at par, and knowing if users of fresh are more brave is thrilling. And while both groups do not differ on most categories, the users of Stable favor stability, which is not too much surprising, and Fresh look for improvements in the area of UX (figure 5).
The affiliation is quite important for an open source project as it lives by to the community. Participation is not restricted to development or special knowledge. Localization, contribution to the help, doing quality checking and not least reporting bugs or enhancement requests is essential for LibreOffice.
It was asked again as multiple choice question, and figure 6 shows the breakdown of single answers where on the right-hand side the number of answers is listed.
This result comes along with the number of members at The Document Foundation. We had n=106 members vs. 1179 who are not. Obviously there is still some potential to contribute!
Finally, and always a good one, is the battle between operating systems.
|Linux||macOS||Microsoft XP,2000,Vista,NT||Microsoft Windows 7||Microsoft Windows 8,10|
And of course we find the preconception represented in the open source aspects (figure 7). Linux users value transparency and replicability much higher whereas Windows users are more interested in the costs.
The evaluation takes only Windows and Linux into account (merging all Windows variants), there are not enough participants for macOS.
For users of LibreOffice, stability, usability, and compatibility are paramount, like other GitHub projects but to an even higher extent. Less relevant aspects like transparency and replicability are more appreciated by experts and developers. Quite interesting is the importance to know the publisher (trusted producer), meaning to know the organization and people for years and relying on the product in the future. This difference may result from the missing description at the GitHub survey but it’s also caused by the appreciation of our work. Many thanks for this compliment.
The participation in the project is not bad compared to other projects – about 25% of the participants state some kind of affiliation. Nevertheless anyone can contribute; it’s very easy to get involved. And when you become a member of The Document Foundation you could have even more impact on the future of the project. So why not join today?
 Raw data and R script of evaluation are available freely.
Nice survey, thank you. Is there anything that these results will change in LibO development or in decision making?
We will have a session during the LibreOffice conference that is based on these results.
Congratulations on the successful survey and on the openness of data and methods! In my opinion the fact that more enquired users were not involved in the project shows that the survey universe was unbiased.
For a future edition I think it would make sense to ask what is the user’s primary Office suite.
Since the most valued quality is Stability maybe TDF could consider a tender or even a hired developer for fixing regressions (nearly 15% of reported bugs)?
Despite the fact that TDF pays a small team (including me) from donations https://plus.google.com/+libreoffice/posts/ech7j2TvT6L, wouldn’t hiring a developer contradict with the other aspects? How much community and free-beer is open source when money comes into play?
Ok. Then just fund tenders (it is one of the items on the leaflet) t
Money has no conflict with Open Source and freedom. As long as the funding is transparent and serves the goals of the community it is as appropriate as hiring staff.