Q to Chris: improving our onboarding process

[Lucas Nussbaum]

Hi,

First, thanks a lot for running for DPL. I'm a bit disappointed by the
amount of discussion so far. DPL campaigns used to be the time of the
year when we take a step back and look at what we are doing (and should
do). It's a pity that we seem to be losing that tradition (even if it's
probably a good thing for DPL candidates :P)

My first question is about "improving our onboarding process".

DPL platforms are often accused of including many vague promises to be
"transparent", "increase participation", "promote non-packaging work"
and so forth. Such overly-general promises are ultimately
counterproductive; lacking a concrete metric of success or failure,
they can easily lead to burnout and a lack of confidence in the role
in general.
Secondly, I will improve our onboarding process for new users and
developers. As suggested above, we are losing a large number of
newcomers with a mixture of missing information in places and overly
complex detail in others. I would experiment with practices such as
usability testing to identify our biggest bottlenecks to, for example,
a new user hearing about Debian and them trying it on their system.

It seems that you are mixing two different issues here: new users, new
developers. Why is it relevant to merge them together?

Also, you give usability testing as one example of things you would like
to try, and quality of documentation as something that should be
improved. Are there other examples of things you would like to try that
we aren't doing yet, or are not using up to their full potential?

As DPL, how do you plan to be involved in all that? (Which can range
from "saying that it's important and hoping that people will follow" to
"I'm prepared to do all the work myself if necessary")

Specifically about "missing infomation in places and overly complex
detail in others", how do you think that we should transform our
developers documentation? (which is in a poor state, I agree, with a mix
of not-so-maintained and partially outdated semi-official documents like
dev-ref, and many wiki pages)

Last, what is your "concrete metric of success or failure" for this
task?

Lucas

[Chris Lamb]

Hey Lucas,

Post by Lucas Nussbaum

I will improve our onboarding process for new users and developers

[…]

Post by Lucas Nussbaum
It seems that you are mixing two different issues here: new users, new
developers. Why is it relevant to merge them together?

I merged them mostly for brevity within my platform - you are right to
imply that attracting "pure" users and developers are distinct cases.

On the hand, Debian attracts a relatively bigger share of technically-
inclined folks where the distinction between terms such as "user" and
"contributor" is especially blurred.

I would tackle these cases somewhat independently, bearing in mind that
it will always be easier to convert users into developers than to attract
them from scratch.

Post by Lucas Nussbaum
Also, you give usability testing as one example of things you would like
to try, and quality of documentation as something that should be
improved. Are there other examples of things you would like to try [..]?

Definitely. As a caveat, I always try to avoid examples here as much as
possible in this area — the discussion can switch from the abstract to its
specific details, often bringing in some unnecessary baggage and history
that distracts from the real topic at hand.

However, as you asked (!) one example I would be especially interested in
improving is the "pre-Debian Installer" installation experience,
particularly in terms of getting Debian onto a USB stick after hearing about
it in passing.

Whilst it is *grossly* unfair and misleading, my go-to tease/factoid on this
topic is that on my laptop only 0.69% of <https://www.debian.org/> is used
for a download link.

Post by Lucas Nussbaum
As DPL, how do you plan to be involved in all that? (Which can range
from "saying that it's important and hoping that people will follow" to
"I'm prepared to do all the work myself if necessary")

Somewhere closer to "doing some of the work" than the former; I'm sure we
can all say that it's important! I'm quite interested in and motivated by
this area in general & believe we have substantial low-hanging fruit.

Post by Lucas Nussbaum
how do you think that we should transform our developers documentation?

This is indeed a situation where improvements lack a concrete metric of
success; simply counting the number of new developers would be highly
misleading, for example.

In addition, the idea of transforming our developer documentation in one
fell swoop is incredibly daunting and likely impossible. Therefore, my
idea would be to first identify the areas causing the most friction and
prioritise efforts around that, even if (!) that results in yet more
development documentation. Removing existing documentation is, as you
might have experienced yourself, incredibly difficult both from a
technical and and political point of view.


Regards,