Sonntag, Juni 20, 2010

More Results from the Debian Community Poll

I have announced the Debian Community Poll and published first results in former blog posts. I'll publish my analysis of the remaining questions about changes to Debian now.

Should Debian remove its non-free component?


  • 80.8% answered no
  • 10.7% answered yes
  • 6.3% answered: I don't know or don't care
  • 2.1% did not answer

Should Debian spend more money?


  • 28.9% choose answer #5: I don't know or don't care.
  • 22.8% choose answer #3: Debian should pay people having important positions in Debian and doing important work.
  • 21.2% choose answer #1: Debian should spend more money on organizing developer conferences and team meetings.
  • 11.4% choose answer #4: Debian should not spend more money.
  • 8.0% choose other (see below) or didn't answer
  • 7.7% choose answer #2: Debian should spend more money on free merchandizing, free DVDs, having a sexy web site, and being present on IT events.
There were quite a number of other answers. One participant missed the information about how the money is currently spend. Several participants didn't want to choose one of the provided answers. They either wanted to choose multiple answers or various combinations of them. Most other answers fall into one of the following categories:
  • various marketing suggestions with different focus than answer #2
  • Debian hosted hardware, infrastructure, services
  • funding upstream development
  • QA and work on release-blocking issues
  • partners and commercial support
  • developing important features
  • security support for oldstable
  • education of prospective developers and contributors
  • documentation for users
  • improving usability and accessibility
  • certifications like LPIC
  • getting supported by hardware and non-free software vendors
  • beer
  • promoting debian in developing countries
  • help contributors running a business on Debian
  • Bounty system
  • updating stable to avoid becoming stale
  • developing multimedia codecs
  • getting compliant with FSF guidelines for a free system distribution
  • developing free replacements to non-free software
  • maintaining a database of debian-friendly hardware
  • lobbying and politics
  • visibility to wider society, even non-IT
  • hardware for driver developers
  • a more sexy DVD/CD set with graphics (like Fedora, Ubuntu)
  • membership to boards of W3C, TEI Consortium, OASIS, etc.

Do you prefer time based releases instead of the "it's ready when it's ready" releases?


  • 73.1% answered no
  • 19.8% answered yes
  • 5.1% anwered: I don't know or don't care
  • 2.0% didn't answer

Which release interval do you prefer?


  • 38.7% choose answer #2: about 12 months
  • 36.9% choose answer #3: 18 - 24 months
  • 10.0% choose answer #5: I don't know or don't care.
  • 5.9% choose answer #1: about 6 months
  • 5.5% choose answer #4: more than 2 years
  • 3.0% didn't answer

Kommentare:

Anon1egYTaGU hat gesagt…

The question on "time based releases" doesn't distinguish between time-based releases (setting the release date long in advance) and time-based freezes (setting the freeze date long in advance but only releasing "when it's ready"). It's possible that some/many respondents who said no to time-based releases would be in favour of time-based freezes.

Torsten Werner hat gesagt…

I doubt that a question about freezes is that interesting to the average Debian user as it is for the developer. But only 6% of all participants have DD or DM status. It is a community poll, not a developer poll.

Anon7tENyHUN hat gesagt…

I (different anon than above) think it actually is relevant - as far as I am concerned, "when it's ready" is mostly about not having to deal with bugs on production servers that would have been found and squashed were it not for the fixed release date, which is why I am strongly against fixed release dates. However, it's generally only of minor importance to me which new generation of which software package gets included in which version of the distribution, as long as new versions do trickle through after some finite amount of time, security problems do get fixed, and there is a testing version that I can draw from in the cases where I need newer versions.

PS: this platform sucks - apparently needs javascript (bad enough in itself) and doesn't even tell you, just refuses to function.

Anon7RavUDe5 hat gesagt…

... and apparently it also needs cookies , just silently refusing to work if you don't allow those. Maybe it doesn't even need JS? Well, summary: it sucks.