~andrewrk/ziglang

2 2

Importance of copious release notes?

Details
Message ID
<dOUeNnmQWvd9U2GBV_O5E_gbqODz0GiW0KulMRpKi9q6RJlsGRVZoN8E-s4ncVYJggMB0eTrt6IYHtqX59kMEbVwxYA-ElBCpZtPtVjmXy0=@protonmail.com>
DKIM signature
pass
Download raw message
Anecdotally, it seems like a large proportion of people are tracking
master. The ziglearn site recommends to use master. And when people
ask a question (e.g. on IRC) and it turns out they are using a tagged
release, the first thing is always to get them on master, even if it's
not relevant to their problem.

So why spend so much time on release notes?

I've seen how much work Andrew puts into them (the page up now
mentions that it takes over a week), the kind of work which must be at
least a little soul-destroying. And it ends up being a very, very long
page to read. Maybe useful mainly as a reference, but how useful is a
version-by-version changelog when most people are tracking master.

I realize the benefit of release notes in crediting people who have
contributed, and in surfacing Zig in HN and other communities.
That said, the nature of HN discussion is usually people reacting the
headline or gist of the post, not the full content. You might get
equal exposure with a more abridged summary.

So I guess I have two questions:

1. At what point should the stance change to recommend users use
tagged versions instead of master?

2. Until that time, would the Zig maintainers consider spending less
of their valuable time on release notes?
Felix Quei├čner
Details
Message ID
<b23a31e5-49fb-bae5-5036-22e2109c1fdc@ib-queissner.de>
In-Reply-To
<dOUeNnmQWvd9U2GBV_O5E_gbqODz0GiW0KulMRpKi9q6RJlsGRVZoN8E-s4ncVYJggMB0eTrt6IYHtqX59kMEbVwxYA-ElBCpZtPtVjmXy0=@protonmail.com> (view parent)
DKIM signature
pass
Download raw message
Heya!

> 1. At what point should the stance change to recommend users use
> tagged versions instead of master?
Imho that point is 1.0. Until then, Zig is a ever-evolving piece of
software and not tracking the master makes your software outdated more
on a daily basis, making ports and adjustments between two tagged
versions a lot more work

> 2. Until that time, would the Zig maintainers consider spending less
> of their valuable time on release notes?
Nobody can keep track of all the changes and release notes document what
has happened since the last tagged release. This means that we can
verify that progress happens and what changes were done to the language
or stdlib, even if you lost track of the language for months

I usually track master with one to two days delay, but even i can't keep
track of all changes and i miss a lot of important stuff. So the release
notes get everyone on the same level of knowledge.

I don't consider the time andrew spents on release notes wasted. I can
imagine it helps reflecting all the work done and check if there were
mistakes or suboptimal solutions you might find in retrospective

Best regards
- xq
Details
Message ID
<CAEnbY+fkqzsTh74PsabiWjx8O3bgVk-kk38gmmP_3WZJLNWO_w@mail.gmail.com>
In-Reply-To
<dOUeNnmQWvd9U2GBV_O5E_gbqODz0GiW0KulMRpKi9q6RJlsGRVZoN8E-s4ncVYJggMB0eTrt6IYHtqX59kMEbVwxYA-ElBCpZtPtVjmXy0=@protonmail.com> (view parent)
DKIM signature
pass
Download raw message
On Tue, 10 Nov 2020 at 08:10, dbandstra <dbandstra@protonmail.com> wrote:
>
> I realize the benefit of release notes in crediting people who have
> contributed, and in surfacing Zig in HN and other communities.
> That said, the nature of HN discussion is usually people reacting the
> headline or gist of the post, not the full content. You might get
> equal exposure with a more abridged summary.


I for one got interested in zig after reading all the awesome release
notes, it showed a language going in a direction I wanted.
Reply to thread Export thread (mbox)