From Newsgroup: muc.lists.freebsd.stable

Am 23.01.26 um 10:21 schrieb Tomoaki AOKI:

Upcoming (hopefully) moving from GitHub to Forgejo managed by
FreeBSD project could change the situation, but it's still
not official. I'm thinking about how my workflow should be
on the switch.

I contributed to Forgejo for a while (tried to improve the HTTPS setup instructions but was frustrated by a slow/haphazard review process and
gave up), so FWIW my impressions:

It's a project that it constrained by developer time, which has
consequences:
* Not all parts of the code have a maintainer. This is partly due to the
fact that Forgejo is a fork of a fork.
* The CVE track record shows that they do care (and that they do have a working CVE process), but also that they did have two remotely
exploitable vulnerabilities in the past three years and needed roughly a
month to fix each one.

Sadlys, I don't know an alternative that I can recommend.
I did a quick search and dug up some web sites with promising leads, but
I can't do the heavy lifting and evaluate them, so I'm just listing my
results in case anybody can follow up on these:

https://www.stackhero.io/en-US/#services
Open source managed services, preinstalled, on a VServer.
GitLab would be the code collaboration tool, but they do offer more
services, many of which could serve any open-source project with server
needs.
They do cost money.

https://www.wearedevelopers.com/en/magazine/298/top-github-alternatives
There are too many useless Top Ten Alternatives lists on the web, but
this one is actually useful and has interesting entries.
My take on each:
GitLab is good but requires a big server and administration.
BitBucket - yeah, Atlassian products give you everything you need but
are a pain to use.
OneDev - no idea. Having CI/CD and Kanban sounds useful.
Gog - tried it once, the setup was a bit too painful and I went to
Forgejo. YMMV.
Codeberg - that's Forgejo plus community and issue tracking. Not sure
how much beyond Forgejo there is.
Radicle - peer-to-peer collaboration based on git. That means the
development workflow does not require a hosting provider anymore, which
I find intriguing, but YMMV.
Gitea - Forgejo was forked from there. Prioritizes features over
security or reliability.
Google Cloud, AWS CodeCommit - I guess these are too commercial for Open-Source collaboration.
Launchpad - no idea. I don't like it because I don't like Canonical, but
YMMV.

Hope this helps somebody.

Regards, and back to lurking,
Jo


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

--000000000000fd861f0649385e5b
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sun, Jan 25, 2026 at 6:10=E2=80=AFAM <joachim@durchholz.org> wrote:

I contributed to Forgejo for a while (tried to improve the HTTPS setup instructions but was frustrated by a slow/haphazard review process and
gave up), so FWIW my impressions:

It's a project that it constrained by developer time, which has
consequences:
* Not all parts of the code have a maintainer. This is partly due to the
fact that Forgejo is a fork of a fork.
* The CVE track record shows that they do care (and that they do have a working CVE process), but also that they did have two remotely
exploitable vulnerabilities in the past three years and needed roughly a month to fix each one.

FWIW, those of us involved with the Glasgow Haskell Compiler have been considering alternatives to GitLab because of various non-technical issues.
We rejected ForgeJo in large part because of that developer time constraint=
:
we don't have enough active developers to contribute to both it and GHC,
and they indeed don't appear to have enough on their own.

I personally contribute to a project on CodeBerg, and my impression of it i=
s
that it seems to be mostly branding and points you to ForgeJo for
documentation and contribution. Also, its CI appears to be pretty
underpowered and large projects would probably need to self-host it.

--=20
brandon s allbery kf8nh
allbery.b@gmail.com

--000000000000fd861f0649385e5b
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr">On Sun, Jan 25, 2026 at 6:10=E2=80=AFAM &= lt;<a href=3D"mailto:joachim@durchholz.org">joachim@durchholz.org</a>&gt; w= rote:</div><div class=3D"gmail_quote gmail_quote_container"><blockquote cla= ss=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid = rgb(204,204,204);padding-left:1ex">I contributed to Forgejo for a while (tr= ied to improve the HTTPS setup <br>
instructions but was frustrated by a slow/haphazard review process and <br> gave up), so FWIW my impressions:<br>

It&#39;s a project that it constrained by developer time, which has <br> consequences:<br>
* Not all parts of the code have a maintainer. This is partly due to the <b=

fact that Forgejo is a fork of a fork.<br>
* The CVE track record shows that they do care (and that they do have a <br=

working CVE process), but also that they did have two remotely <br>
exploitable vulnerabilities in the past three years and needed roughly a <b=

month to fix each one.<br></blockquote><div><br></div><div>FWIW, those of u=
s involved with the Glasgow Haskell Compiler have been</div><div>considerin=
g alternatives to GitLab because of various non-technical issues.</div><div= >We rejected ForgeJo in large part because of that developer time constrain= t:</div><div>we don&#39;t have enough active developers to contribute to bo=
th it and GHC,</div><div>and they indeed don&#39;t appear to have enough on=
their own.</div><div><br></div><div>I personally contribute to a project o=
n CodeBerg, and my impression of it is</div><div>that it seems to be mostly=
branding and points you to ForgeJo for</div><div>documentation and contrib= ution. Also, its CI appears to be pretty</div><div>underpowered and large p= rojects would probably need to self-host it.</div></div><div><br></div><spa=
n class=3D"gmail_signature_prefix">-- </span><br><div dir=3D"ltr" class=3D"= gmail_signature"><div dir=3D"ltr"><div><div dir=3D"ltr"><div>brandon s allb= ery kf8nh</div><div><a href=3D"mailto:allbery.b@gmail.com" target=3D"_blank= ">allbery.b@gmail.com</a></div></div></div></div></div></div>

--000000000000fd861f0649385e5b--


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

Past experience with proposed commercial software for FreeBSD
(in this particular case, Jira) convinced me that the community
would not accept anything that was not open-source.

mcl


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

Am 25.01.26 um 18:29 schrieb Mark Linimon:

Past experience with proposed commercial software for FreeBSD
(in this particular case, Jira) convinced me that the community
would not accept anything that was not open-source.

Actually, I'd avoid Atlassian products in general.
Not because they aren't open source. That's certainly a minus, but the elephant in the room is that they are complicated and time-consuming to use. Plus, I have yet to see a single organisation that is able to set up a
Jira workflow that is actually appropriate for what people were trying
to do with it.

I attribute this to the fact that Atlassian, like any commercial
software vendor, cannot sell to developers but sells to management, and
this skews its capabilities towards what managers want rather than what developers want.
I believe this is actually one of the major points why commercial
software tends to suck for us developers, and FOSS is much more fun to
work with :-D


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

Plus, I have yet to see a single organisation that is able to set
up a Jira workflow that is actually appropriate for what people
were trying to do with it.

Many years ago, I actually had a working prototype but it was so
complicated that I was the only one who could possibly have used
it :-)

At which point, they changed their codebase, and the whole thing
went down the toilet.

mcl


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

i'm reading this and...
completely new idea!
if it's so difficult to choose from existing projects, why not write entirely new one from scratch?
--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

--0000000000008f5ee006493c919c
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sun, Jan 25, 2026 at 3:50=E2=80=AFPM Sulev-Madis Silber < freebsd-stable-freebsd-org730@ketas.si.pri.ee> wrote:

if it's so difficult to choose from existing projects, why not write
entirely new one from scratch?

Are there enough people to pull that off in a reasonable timeframe _and_ maintain it afterward?

--=20
brandon s allbery kf8nh
allbery.b@gmail.com

--0000000000008f5ee006493c919c
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr">On Sun, Jan 25, 2026 at 3:50=E2=80=AFPM S= ulev-Madis Silber &lt;<a href=3D"mailto:freebsd-stable-freebsd-org730@ketas= .si.pri.ee">freebsd-stable-freebsd-org730@ketas.si.pri.ee</a>&gt; wrote:</d= iv><div class=3D"gmail_quote gmail_quote_container"><blockquote class=3D"gm= ail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,= 204,204);padding-left:1ex">if it&#39;s so difficult to choose from existing=
projects, why not write entirely new one from scratch?<br> </blockquote></div><div><br clear=3D"all"></div><div>Are there enough peopl=
e to pull that off in a reasonable timeframe _and_ maintain it afterward?</= div><div><br></div><span class=3D"gmail_signature_prefix">-- </span><br><di=
v dir=3D"ltr" class=3D"gmail_signature"><div dir=3D"ltr"><div><div dir=3D"l= tr"><div>brandon s allbery kf8nh</div><div><a href=3D"mailto:allbery.b@gmai= l.com" target=3D"_blank">allbery.b@gmail.com</a></div></div></div></div></d= iv></div>

--0000000000008f5ee006493c919c--


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

Am 25.01.26 um 21:50 schrieb Sulev-Madis Silber:> i'm reading this and...

completely new idea!

if it's so difficult to choose from existing projects, why not write

entirely new one from scratch?
Sounds nice. One project to rule them all, right?
Just like with standards?

Oh no: https://xkcd.com/927/


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

On 01/25/2026 2:50 PM CST Sulev-Madis Silber <freebsd-stable-freebsd-org730@ketas.si.pri.ee> wrote:

I doubt I still have the reference to the article, but when
we were looking to move off GNATS, I found that someone had
written a comprehensive survey of bug trackers as of that time.
(fwiw, he found all of them lacking in one way or the other).

The article ended with the line:

"do NOT try to write your own".

I took this as Words Of Wisdom from someone who had tried.

mcl


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

Sulev-Madis Silber <freebsd-stable-freebsd-org730@ketas.si.pri.ee>
writes:

if it's so difficult to choose from existing projects, why not write
entirely new one from scratch?

https://openhub.net/p/forgejo/estimated_cost
tl;dr: Open Hub estimates the effort required to reproduce Forgejo from
scratch to 120 developer-years.
DES
--
Dag-Erling Sm|+rgrav - des@FreeBSD.org
--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.stable

On 01/26/2026 5:24 AM CST Dag-Erling Sm|+rgrav <des@freebsd.org> wrote:
Open Hub estimates the effort required to reproduce Forgejo from
scratch to 120 developer-years.

Before any customizations.
mcl
--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.21b-Linux NewsLink 1.2