There are too many PRs

User avatar
Lone_Wolf
Member
Posts: 2576
Joined: Sun Apr 09, 2017 05:50
GitHub: LoneWolfHT
IRC: LandarVargan
In-game: LandarVargan

Re: There are too many PRs

by Lone_Wolf » Post

Zughy wrote:
Wed Jun 17, 2020 16:58
the point of this is core devs not having enough dev time. If you bring more bureaucracy, it won't do much good. It's basically the summary of this
What you linked is completely different from Wuzzy's suggestion, which I agree with unlike that issue
My ContentDB -|- Working on CaptureTheFlag -|- Minetest Forums Dark Theme!! (You need it)

CalebJ
Member
Posts: 407
Joined: Sat Dec 29, 2018 19:21
GitHub: KaylebJay
IRC: KaylebJay
In-game: CalebJ
Location: Tunnelers' Abyss

Re: There are too many PRs

by CalebJ » Post

Wuzzy's suggestion is brilliant. There are a *lot* of old PRs that haven't been touched in months or even years (from the author). Now, this would remove more PRs but wouldn't make other PRs get merged quicker because the devs already disregarded these. Because of this, I agree, high complexity PRs should be reviewed first, and maybe this actually would do something about the PR problem.

I would suggest something else: In cases where there is the 'One Approval' label, find a second dev to approve it and merge it! There are currently 23 PRs with 'One Approval' and they usually never go anywhere. Putting high priority on these would definitely be a step in the right direction.

As for Issue #9989, my comment on it is already there. Modularizing minetest would make it more complicated without a shadow of a doubt, harder for devs to do things, and more confusing. Secondly, it won't close any PRs or issues automagically, it will just diversify them into different repositories/issue-trackers. Ugh.

User avatar
Linuxdirk
Member
Posts: 3216
Joined: Wed Sep 17, 2014 11:21
In-game: Linuxdirk
Location: Germany
Contact:

Re: There are too many PRs

by Linuxdirk » Post

CalebJ wrote:
Wed Jun 17, 2020 18:41
There are currently 23 PRs with 'One Approval' and they usually never go anywhere. Putting high priority on these would definitely be a step in the right direction.
If it works like with issues then “high priority” means nothing as there are issues opened 6 years ago labeled as “high priority”.

User avatar
rubenwardy
Moderator
Posts: 6972
Joined: Tue Jun 12, 2012 18:11
GitHub: rubenwardy
IRC: rubenwardy
In-game: rubenwardy
Location: Bristol, United Kingdom
Contact:

Re: There are too many PRs

by rubenwardy » Post

I have gone through the list today and closed 4 PRs that can definitely be closed by this point. I tagged a few others as possible close

We already have a policy of closing PRs after 1 month if there's no response from the author

Please bump issues and PRs that you think should be closed
Renewed Tab (my browser add-on) | Donate | Mods | Minetest Modding Book

Hello profile reader

User avatar
runs
Member
Posts: 3225
Joined: Sat Oct 27, 2018 08:32

Re: There are too many PRs

by runs » Post

Maybe a Testcon. To be discussed in person.

Spain would be is a good place to the first event (after covid). We have the 3 s: Sun, Sangria and S... that :-D

CalebJ
Member
Posts: 407
Joined: Sat Dec 29, 2018 19:21
GitHub: KaylebJay
IRC: KaylebJay
In-game: CalebJ
Location: Tunnelers' Abyss

Re: There are too many PRs

by CalebJ » Post

rubenwardy wrote:
Wed Jun 17, 2020 20:24
I have gone through the list today and closed 4 PRs that can definitely be closed by this point. I tagged a few others as possible close

We already have a policy of closing PRs after 1 month if there's no response from the author

Please bump issues and PRs that you think should be closed
Nice, thanks! I think, the rest of them just need more core dev love :-) (And of course, 5.3.0 is already in feature freeze ...)
If it works like with issues then “high priority” means nothing as there are issues opened 6 years ago labeled as “high priority”.
Quite true, Linuxdirk, but I only see 6 High Priority issues (one from 2014, ouch). 23 to 6 wouldn't be so bad :)

User avatar
v-rob
Developer
Posts: 970
Joined: Thu Mar 24, 2016 03:19
GitHub: v-rob
IRC: v-rob
Location: Right behind you.

Re: There are too many PRs

by v-rob » Post

I think the "Low complexity" label would be very useful; there are many PRs that are actually very simple and would only take a little bit of reviewing to complete. Finishing a lot of small jobs could drastically reduce the amount of PRs. I personally am trying to make my PRs smaller with less features for this precise reason, and I can add the rest of the features in later PRs.
Core Developer | My Best Mods: Bridger - Slats - Stained Glass

User avatar
Linuxdirk
Member
Posts: 3216
Joined: Wed Sep 17, 2014 11:21
In-game: Linuxdirk
Location: Germany
Contact:

Re: There are too many PRs

by Linuxdirk » Post

rubenwardy wrote:
Wed Jun 17, 2020 20:24
We already have a policy of closing PRs after 1 month if there's no response from the author
So you say that in all the 154 open PRs for Minetest the authors are active at least once a month?

User avatar
rubenwardy
Moderator
Posts: 6972
Joined: Tue Jun 12, 2012 18:11
GitHub: rubenwardy
IRC: rubenwardy
In-game: rubenwardy
Location: Bristol, United Kingdom
Contact:

Re: There are too many PRs

by rubenwardy » Post

Response requires a reviewer to request or ask for something
Renewed Tab (my browser add-on) | Donate | Mods | Minetest Modding Book

Hello profile reader

User avatar
Zughy
Member
Posts: 301
Joined: Thu Mar 26, 2020 18:23
GitHub: belongs_to_microsoft
In-game: Zughy
Location: Italy
Contact:

Re: There are too many PRs

by Zughy » Post

rubenwardy wrote:
Thu Jun 18, 2020 08:41
Response requires a reviewer to request or ask for something
So we're back on dev time

User avatar
Wuzzy
Member
Posts: 4786
Joined: Mon Sep 24, 2012 15:01
GitHub: Wuzzy2
IRC: Wuzzy
In-game: Wuzzy
Contact:

Re: There are too many PRs

by Wuzzy » Post

What about inviting people other than core devs to do some of the “grunt work”?
That is, doing this “boring” pre-review work.
People who at least have shown a basic understanding of software development and Minetest itself while not being a core dev …
Thing is, there's a really tiny number of people who (can) actually decide things.

I suppose most of this pre-review work can be done without having perfect knowledge about all of the Minetest internals.

IIRC celeron55 or another coredev had a similar idea a while ago. However, for this to actually work, IMO it's not enough to just say “help us review stuff” and that's it. An instruction manual specifically for those people would help a lot. Because they can't read your mind. :P And the label “pre-pending” will definitely help a lot here.

Also: What is the official opinion regarding my suggestion from the previous post?

The “lack of dev time” is a myth. There IS plenty of dev time. Look at how many volunteer coders we have! It's just that the majority are not core devs, that's all. What we have is a lack of CORE dev time. Subtle difference.

Also, the way the system works is that contributors are still kinda treated as people who “want something from Minetest” (like a customer), rather than collaborators on equal footing that actually give something. Even if they put in a lot of work in their PR … I think this is also a problem culturally, if we ignore all the organizational stuff, and I bet this quickly burns out many would-be-long-term-contributors. :-(

User avatar
LMD
Member
Posts: 1386
Joined: Sat Apr 08, 2017 08:16
GitHub: appgurueu
IRC: appguru[eu]
In-game: LMD
Location: Germany
Contact:

Re: There are too many PRs

by LMD » Post

Wuzzy wrote:
Fri Jun 19, 2020 18:23
The “lack of dev time” is a myth. There IS plenty of dev time. Look at how many volunteer coders we have! It's just that the majority are not core devs, that's all. What we have is a lack of CORE dev time. Subtle difference.

Also, the way the system works is that contributors are still kinda treated as people who “want something from Minetest” (like a customer), rather than collaborators on equal footing that actually give something. Even if they put in a lot of work in their PR … I think this is also a problem culturally, if we ignore all the organizational stuff, and I bet this quickly burns out many would-be-long-term-contributors. :-(
This.
My stuff: Projects - Mods - Website

User avatar
Linuxdirk
Member
Posts: 3216
Joined: Wed Sep 17, 2014 11:21
In-game: Linuxdirk
Location: Germany
Contact:

Re: There are too many PRs

by Linuxdirk » Post

Wuzzy wrote:
Fri Jun 19, 2020 18:23
Also, the way the system works is that contributors are still kinda treated as people who “want something from Minetest” (like a customer), rather than collaborators on equal footing that actually give something.
Quoted for emphasis and agreement.

User avatar
Lone_Wolf
Member
Posts: 2576
Joined: Sun Apr 09, 2017 05:50
GitHub: LoneWolfHT
IRC: LandarVargan
In-game: LandarVargan

Re: There are too many PRs

by Lone_Wolf » Post

But they *do* want something

They want their PRs merged
My ContentDB -|- Working on CaptureTheFlag -|- Minetest Forums Dark Theme!! (You need it)

User avatar
Zughy
Member
Posts: 301
Joined: Thu Mar 26, 2020 18:23
GitHub: belongs_to_microsoft
In-game: Zughy
Location: Italy
Contact:

Re: There are too many PRs

by Zughy » Post

Lone_Wolf wrote:
Sat Jun 20, 2020 00:04
They want their PRs merged
without waiting aeons

Astrobe
Member
Posts: 570
Joined: Sun Apr 01, 2018 10:46

Re: There are too many PRs

by Astrobe » Post

I think the Pull Requests feature popularized by Git gave to some the wrong idea.

PRs actually consists in hundreds of people |i]pushing[/i] things on a dozen of people (at best). No wonder PRs eventually overflow. Just looking at them takes too much time already. It's like a brain DDOS. That's the bad side of "social coding".

PRs are a terrible feature in the large, because it does not "scale" for this very reason. Just like, before the advent of Github, maintainers of popular projects did ignore email requests from random people. Github did not really solve anything in this area.

One cannot manage hundreds of developers with just one layer of management. Just look at what the Lua team does (no PR) or what the Linux kernel team does (multiple layers of management). Both are old and popular projects that did stay away from the "social coding" delusions.

User avatar
runs
Member
Posts: 3225
Joined: Sat Oct 27, 2018 08:32

Re: There are too many PRs

by runs » Post

Look what the team of Veloren do in a recent interview:

https://www.gamingonlinux.com/2020/06/interviewed-veloren-an-upcoming-foss-multiplayer-voxel-rpg/

In short: Collaborative teamwork and objectives. All what I said...

User avatar
Linuxdirk
Member
Posts: 3216
Joined: Wed Sep 17, 2014 11:21
In-game: Linuxdirk
Location: Germany
Contact:

Re: There are too many PRs

by Linuxdirk » Post

Astrobe wrote:
Sat Jun 20, 2020 09:22
PRs are a terrible feature in the large, because it does not "scale" for this very reason. Just like, before the advent of Github, maintainers of popular projects did ignore email requests from random people. Github did not really solve anything in this area.
If you don’t want PRs, disable the permission to create them.

CalebJ
Member
Posts: 407
Joined: Sat Dec 29, 2018 19:21
GitHub: KaylebJay
IRC: KaylebJay
In-game: CalebJ
Location: Tunnelers' Abyss

Re: There are too many PRs

by CalebJ » Post

Linuxdirk wrote:
Sun Jun 21, 2020 03:48
Astrobe wrote:
Sat Jun 20, 2020 09:22
PRs are a terrible feature in the large, because it does not "scale" for this very reason. Just like, before the advent of Github, maintainers of popular projects did ignore email requests from random people. Github did not really solve anything in this area.
If you don’t want PRs, disable the permission to create them.
... I smell a new topic. "There are 0 PRs" ....

User avatar
Lejo
Member
Posts: 718
Joined: Mon Oct 19, 2015 16:32
GitHub: Lejo1
In-game: Lejo

Re: There are too many PRs

by Lejo » Post

Another thing about dev-time.
I think it would be useful to split lua and core development up.
C++ Minetest Developement is a lot more harder to understand than just modding.
There is no core_api.txt file like lua_api.txt informing about all the classes.

A lot more people could be minetest-mods "core developer".
Than "real" core-devs would also have more time for the core.

User avatar
rubenwardy
Moderator
Posts: 6972
Joined: Tue Jun 12, 2012 18:11
GitHub: rubenwardy
IRC: rubenwardy
In-game: rubenwardy
Location: Bristol, United Kingdom
Contact:

Re: There are too many PRs

by rubenwardy » Post

ANAND / ClobberXD is now a core developer. More people are being considered
Renewed Tab (my browser add-on) | Donate | Mods | Minetest Modding Book

Hello profile reader

User avatar
rubenwardy
Moderator
Posts: 6972
Joined: Tue Jun 12, 2012 18:11
GitHub: rubenwardy
IRC: rubenwardy
In-game: rubenwardy
Location: Bristol, United Kingdom
Contact:

Re: There are too many PRs

by rubenwardy » Post

Lejo wrote:
Tue Jun 23, 2020 19:23
Another thing about dev-time.
I think it would be useful to split lua and core development up.
C++ Minetest Developement is a lot more harder to understand than just modding.
There is no core_api.txt file like lua_api.txt informing about all the classes.

A lot more people could be minetest-mods "core developer".
Than "real" core-devs would also have more time for the core.
Here's the landing page for Engine documentation: https://dev.minetest.net/Engine

Remember that minetest-mods is not official. There is also a separate Minetest Game core development team to Minetest (Engine)
Renewed Tab (my browser add-on) | Donate | Mods | Minetest Modding Book

Hello profile reader

User avatar
Lejo
Member
Posts: 718
Joined: Mon Oct 19, 2015 16:32
GitHub: Lejo1
In-game: Lejo

Re: There are too many PRs

by Lejo » Post

ANAND / ClobberXD is now a core developer. More people are being considered
Hope this helps.

I currently also have...
...one pr which were only one time aproved
...one reviewed pr which weren't merged or denied afterwards
...one rebase-PR which were not even touched
...one non trival which could be closed(just don't care)

At least two of them could be merged or denied pretty fast.
rubenwardy wrote:
Tue Jun 23, 2020 19:26
Here's the landing page for Engine documentation: https://dev.minetest.net/Engine
It's nothing in comparison to lua_api.txt
It's a help but core-development is anyway let's say... not easy.
rubenwardy wrote:
Tue Jun 23, 2020 19:26
Remember that minetest-mods is not official. There is also a separate Minetest Game core development team to Minetest (Engine)
Didn't knew that, it seems to be cause it's also managed by some core devs :)

User avatar
Linuxdirk
Member
Posts: 3216
Joined: Wed Sep 17, 2014 11:21
In-game: Linuxdirk
Location: Germany
Contact:

Re: There are too many PRs

by Linuxdirk » Post

rubenwardy wrote:
Tue Jun 23, 2020 19:26
Here's the landing page for Engine documentation: https://dev.minetest.net/Engine
So the wiki now is the official source for documentation again?

User avatar
rubenwardy
Moderator
Posts: 6972
Joined: Tue Jun 12, 2012 18:11
GitHub: rubenwardy
IRC: rubenwardy
In-game: rubenwardy
Location: Bristol, United Kingdom
Contact:

Re: There are too many PRs

by rubenwardy » Post

Linuxdirk wrote:
Tue Jun 23, 2020 20:39
rubenwardy wrote:
Tue Jun 23, 2020 19:26
Here's the landing page for Engine documentation: https://dev.minetest.net/Engine
So the wiki now is the official source for documentation again?
The wiki is the official source for engine development guidelines and documentation, and always has been. It is not the official source for the Lua API, and never was

"official" is a strange term here. The Wiki is official - the Lua API stuff is just not canonical or maintained during development
Renewed Tab (my browser add-on) | Donate | Mods | Minetest Modding Book

Hello profile reader

Post Reply

Who is online

Users browsing this forum: No registered users and 10 guests