End-of-year-surveys/2020/Discussion

Jump to: navigation, search

Meetup on EOY 2020 survey results (session 1/2)

Held on the 23rd of January 2021.

Setting

Context

A community wide survey was conducted at the end of 2020. The goal was to better understand the community's needs and to provide a baseline for improvements.

Goal

The goal was to come together on an reasonable understanding of the results, as well as to discuss priorities and ways forward. On the agenda:

  • addressing pain points visible from the results
  • improving knowledge transfer
  • promotion

Summary

Ordered by "most to least discussed".

Lack of integration

The participants agreed to foster changes towards a simplification and better integration of the platforms, catering in particular for new users, in response to the following concerns:

  • the path from opensuse.org to downloading, finding docs and installing is unnecessarily complicated
  • information is scattered (docs in particular: wiki and official docs)
  • there is insufficient structure, making it difficult to tell what's really important from what is less
  • basic information about the distributions ("more about...") is missing from opensuse.org
  • the message conveyed by Lives images not being recommended for installation was found confusing
  • "rebuild" images for Leap were found missing

Promotion

The participants agreed to foster changes towards better promotion, welcomed the idea of press-kits, and several participants expressed their sympathy for a more "aggressive" promotion, in response to the following concerns:

  • there is generally insufficient visibility of the openSUSE brand, of its value-proposition
  • openSUSE could better promote its name by having users intervene on different social platforms, or by having influential people featured in live events (podcasts, talks, conferences)
  • promotional material could be enriched, improved, and made more accessible via "press kits"
  • both promotion and documentation in general could find inspiration in the August 2019 issue of "Getting Started with Linux"

Knowledge transfer

Even though there was no unanimous endorsement of the ideas, the participants viewed positively the ideas of organizing workshops on specific, technical subjects (i.e. openQA) and of making it easier for certain groups to work with a mentor-apprentice workflow, in response to the following concerns:

  • there are many skilled people in the community, but onboarding is difficult, and many activities and projects lack visibility
  • some teams work beyond capacity, and the perspective of learning can motivate new contributors to step in.

Details

Lack of integration

  • Adrien: actionable outcomes of this meeting should go to etherpad (to document ideas) and Progress / Pagure (to act)
  • Adrien: the survey suggests that making the web platforms navigable is key
  • Gerald: regular page and wiki are confusing; it is desirable to improve the docs by creating an area where it's official and up-do-date
  • Lars: we could organize a new survey with free text to better understand harware pain points, also present good points
  • Chris: opensuse.org mentions tools, but there is room for adding FAQs and the other informations about the distributions
  • Sasi: easier access to the docs was not a guiding constraint of the new design, but it would help Sasi to have a clear path from entry point to specific things, such as documentation
  • Axel: more rich content, include Codex and make it more discoverable
  • Sasi: opensuse.org cannot contain everything about distros, moving distros to another pages is feasible, but spreading each distro under a separate subdomain is not easy for the user to comprehend; also there will be soon a single entry point for each distribution (get.opensuse.org)
  • AJV: the entry point should disclose the different distributions, one shouldn't have to ask a web engine to find about them
  • Adrien: feasible to organize "sprint sessions" to improve the web platform?
  • Sasi: cannot write content alone, but Sasi OK with lightly mentoring a couple of volunteers willing to help with contents and design
  • Adrien: Why installation + Live? And then why not recommended to install?
  • Axel: not sure about the argument for not recommending installation from Live image
  • AJV: the Live images can be much easier to use because they are packaged with more forgiving drivers (usually more up-to-date and compatible)
  • Adrien to Axel: How to go about this integration change? Where to push change?
  • Chris: the live installers should run through the same amount of tests
  • Axel: should include rebuild images (software packages from x weeks ago) as well
  • Andreas: maintenance and curation of the images (every x weeks, openQA on Leap rebuilds)

Knowledge transfer

  • Adrien: as far as knowledge transfer, we could generalize mentoring; time investment at start but repaid for the benefits of all
  • Doug: could set up workshops on openQA
  • AJV: an underlying thread in all this is fragmentation: we need to bring resources together, we need less channels so that people can be focussed on where to get stuff, and more integration
  • Milan: it's not always clear where to find the correct info, about packages, etc.; we should take a page off the arch wiki because it's much more orderly
  • Axel: we should make it sure that we make the most of channels for what they're good for, and for example support and technical information is difficult to pick up on instant messaging applications
  • Sasi: using the mailing list out of the browser is user friendly and provides a nice middle-ground between chat and "classic" ML

Promotion

  • Adrien: perhaps we could make sure people understand the value proposition of TW
  • AJV: it's not really a matter of TW / Leap, but we should make sure people perceive the oS "brand"
  • Chris: depends who you're considering for audience
  • Maurizio: Fedora converted their wiki into real documentation; what live presence can we offer?
  • AJV: we could use the material from the magazine as quick start + tutorial and expose this as a entry point for even the newcomer
  • Axel: the CT magazine could feature us
  • Adrien: feature ourselves on social platforms or get featured in live events
  • Chris: Leap releases should be featured on social media
  • Sasi: press kit need to be made, easier access to branding material
  • Andreas: people to demonstrate their use of an oS distro, people that show that things work / can be done on oS distros
  • AJV: it's easier to get good feedbacks when things fall together easily
  • Ivo: ready to publish video interviews of "special users" on platforms

Meetup on EOY 2020 survey results (session 2/2)

Held on the 30th of January 2021.

Setting

  • Doug organized the meeting and prepared slides
  • Ivo recorded the meeting (TBA)
  • 15-20 people attended
  • Adrien took these minutes

Context

A community wide survey was conducted at the end of 2020. The goal was to better understand the community's needs and to provide a baseline for improvements.

Goal

The goal was to come together on an reasonable understanding of the results, as well as to discuss priorities and ways forward. On the agenda:

  • tools driving switchers to openSUSE (Where are users coming from)
  • discuss flagship project/s
  • expanding global users
  • increasing diversity
  • increase usage with people under 34

Summary

User base & flashipping

The participants didn't settle on a full-fledged approach towards expanding the user base or flagshipping a single distribution; however clear shortco'mings and pathways for incremental improvements emerged:

User base. Our users are either experienced users or users coming from the Windows/Ubuntu ecosystems and likely to be less experienced.

  • this suggests that we should spotlight advantages our distributions offer to little experienced users without "dumbing down" our message about more advanced features.

Press kits. As with the last session, press kits were deemed a crucial asset for promotion and for creating a single frame of reference for people interested in talking positively of our distributions.

Leap. It excels in stability, which makes it well suited to the enterprise world, especially if familiar with SLE, but could benefit from simpler update procedures.

  • maintenance teams are going to be got in touch with to seek simplifications

Tumbleweed. It has a bit of an "geek" flair and might seem slightly more difficult to maintain for the end users, in particular regarding the requirement to reboot for installing low-level packages

  • mitigating solutions: offline updates, transactional updates, (post-scriptum from Adrien: even though it does not address the rebooting issue, `tumbleweed-cli` should be given more spotlight in docs & good practices to make updates easier to control), also `btrfs` should be presented as strong recommendation

Moreover, its strong points should be better underlined: fresh kernel; fresh packages; thoroughly tested; forgiving through snapshots, It was agreed that this combination of features made Tumbleweed the most distinctive advantage over similar and competing distributions.

Feedback mechanisms, communication, diversity

Survey. The recent "End of the Year" survey was welcomed and its usefulness acknowledged; however certain blind-spots and shortcomings were identified:

  • we should make sure to run surveys periodically to monitor change (probably once a year)
  • surveys should allow better integration between questions (allowing for better clustering of profiles) and allow for more direct feedback, with some feedback questions designed to be accessible even with basic knowledge of the distributions and the community.
  • surveys may also have certain parts designed with surveys from other distributions in mind, to invite comparisons.

Communication

  • the landing page of the "Welcome" desktop application could implement desktop notifications listening to important news
  • news-o-o could use per-category filtered web feeds (so as to mitigate potential opt-outs in reaction to an heterogeneous stream of topics) and web push notifications

Community life & diversity. The goal of more bonding opportunities to better integrate and coordinate among sub-communities and local communities was seen as very important. Were seen favorably:

  • parties for released versions of Leap
  • parties centered around discussions of survey results (if surveys become part of the oS tradition)

It was also acknowledged that the Western demographic center might act as "too heavy" a center of gravity for information, documentation and frame of reference.

  • to mitigate this it was proposed to practice the policy of always translating important news to other languages (Spanish, Arabic, Russian, Chinese are being considered for starters, depending on our translators' availability).

Details

  • Gerald: thanks everyone and rejoices that we can work as a "team of teams" and choose where you'll be putting focus together
  • Neal: happy that we can do this and is sorry he missed last session
  • Ariez: we are here as part of the "Project", how do we manage to get more end users to join us in our endeavor; we should make sure we are not just among ourselves; we should make sure everybody has a say

Switchers

  • Doug presents slides about switchers and underlines the switching dynamics
  • Neap: we would be more successfuly by helping more new users, because we are competing with the same "market share" as with Ubuntu
  • Gerald: perhaps it's more like people have spent quite some time on Ubuntu, and we need to cater for more advanced users
  • Neal: difficult for advocating is "What can I talk about"? He would need to have something to talk about
  • Ariez:
   1. when you try to start a VPS or project, you go to a provider, and everyone has only Ubuntu / CentOS, oS distros are quite rare there; it would be nice to reach to servers vendors to show that oS distros is good we'd gain visibility there;
   2. for new users, the mindset of people is dictated by search engines, and Ubuntu will always rank higher than us in the results; Ubuntu has built a presence we cannot really change easily
  • Neal:
   1. agrees, yet there is lack of confidence in oS as desktop
   2. we lack a clear, killer feature or killer argument in our favour to get the end users' attention to the desktop
  • Nate: How about we reach to vendors (i.e. Lenovo) and aim for preinstalls?
  • Gerald: it's not that easy, this is a very demanding to create and maintain relationship with vendors and cloud service providers, and the latter prefer to have just 2 Linux distros; the strongest lever would be demand (schools, institutions, etc), you need customers asking for it.
  • Neal: we can use SLE prepare the ground to introduce Leap, but we don't have any marketing built around this idea.
  • Ariez: there is a perception is "for geeks" while Leap is your go-to platform; there are many program versions antique on Leap however (next cloud desktop: zypper gives a ~1yo version)
  • Axel: Leap is quite behind as far as packages updates are concerned
  • Gerald: being behind is not always a consequence of using SLE software
  • Neal: the out-of-dateness depends on whether we want to create more work for ourselves; updating software on Leap requires a lot of effort
  • Axel: we should discuss with maintenance team on how to make the process (updating on Leap) easier)
  • Ivo: we can / should create promotional contents centered on concrete use cases, to attract people who need to look up to someone or to a workflow they are particularly interested in.

Flagshipping

  • Doug: We might consider the idea of "flashipping" a distro
  • Neal: TW = "rolling-release with a safety net", we don't promote this aspect very much, we can make the point for this and promote around this; that something no competitor has, and that makes us special; TW might stay / be flaship for the reason is it more specific to us.
  • Doug: differentiators?
  • Ariez: have a painful experience running on ext4
  • Andreas: TW needs to update frequently, so me need to improve the user experience, and we need to reboot for installing a new kernel
  • Neal: two solutions -- force offline updates and go for transactional updates

Demographics

  • Andreas: we ought to try to keep track of our comparable results change from one survey to the next; we don't really know how the numbers differ comparatively with other distros or with regard to our own past
  • Neal: some people we are not aware of the survey
  • Andreas: desktop notification about community things?
  • Attila: the welcome app has this rss feed already
  • Axel: additional mail to project@ and factory@ is probably justified
  • Adrien: web push for browsers, per category feeds to avoid opting-out
  • LCP: per category feed is doable
  • Andreas: present the results during the oS conference (around May-June), which means that the survey should be done in February
  • Neal: we could have a release party where we talk about the survey, to celebrate the release separately from celebrating the community; the latest month would be end of nov)
  • Gerald: survey should be tied to a significant event to make sense of it
  • Neal: we should make a survey with a path for outsiders
  • Adrien: requires filters and fail-safe to make sure people know what they're talking about, otherwise it's a just a matter of perception, and it's difficult to ask informative question about perception if you don't know the vantage point of people; cost-efficiency is difficult to achieve
  • Ariez: we need to know better why people switch, and what would make them switch
  • Neal: having press kit ready would make anyone involved in promotion much easier
  • Adrien : make the community fun -> get people to form dedicated teams -> get featured and have more visibility
  • Neal: we already have many volunteers taking care of the channels and engaging with the community; parties and celebrations are in range; the bar was already a move in that direction
  • Ariez: are we making sure local communities / non-English native groups get to participate to things?
  • Neal: they are many local teams and they don't work in coordination, we should make sure there is more coordination between several teams;
  • Adrien: we should discipline ourselves to systematically translate important stuff into "most used languages difficult to get to English from", (Spanish Arabic Russian Chinese)?
Retrieved from "https://en.opensuse.org/index.php?title=End-of-year-surveys/2020/Discussion&oldid=152746"