Why base library changes are only discussed on GHC issue tracker and not on the libraries@ list?

newer
Why base library changes are only...

older
RFC: Add HasCallStack constraint...

Oleg Grenrus

7 Jul 2021 7 Jul '21

11:11 a.m.

For example - https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed. - Oleg

Show replies by date

Carter Schonwald

7 Jul 7 Jul

11:42 a.m.

Agreed. This is a problem. I’ve tried to help make current clc folks aware of this privately a time or two this past year :( The only private part should be public coms on their discussion group to record voting on stuff that isn’t unanimous. Anything beyond that fails to align with healthy collaborative discourse norms More concerningly, only ~2-3 members of the current clc seem to be actively involved in public discussions on the library list or GitHub. And a deep misunderstanding that they need be maintainers of stuff that falls under the umbrella of core libraries now. Which is a fiction invented only on the past two years. Clc was formed to help guide decisions on base and be a suport for core libraries authors/maintainers. Not as an authority over those maintainers. There’s def problems with stuff and a lot of folks have privately expressed a lot of frustration about this over the past 12 months. On Wed, Jul 7, 2021 at 11:12 AM Oleg Grenrus wrote:

...

For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

- Oleg

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Henning Thielemann

11:49 a.m.

On Wed, 7 Jul 2021, Oleg Grenrus wrote:

...

For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

I think so, too, and I missed them as well.

Sandy Maguire

12:41 p.m.

At risk of being the messenger who gets shot.... As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this. I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes. And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through. Best, Sandy On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann < lemming@henning-thielemann.de> wrote:

...

On Wed, 7 Jul 2021, Oleg Grenrus wrote:

...
For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Taylor Fausak

1:29 p.m.

As another outsider, I agree with everything that Sandy said.

...

On Jul 7, 2021, at 12:41 PM, Sandy Maguire wrote:

At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann mailto:lemming@henning-thielemann.de> wrote:

On Wed, 7 Jul 2021, Oleg Grenrus wrote:

...
For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Oleg Grenrus

1:32 p.m.

Thanks for this reply. It made me reread https://wiki.haskell.org/Library_submissions page. In the Guide to proposers section it says: - All library proposals should start on the relevant issue tracker. - At this point, the library maintainer is responsible for taking next steps. - ... or decide that this is a controversial decision that must be discussed with the CLC. - If the CLC decides that the discussion must be discussed with the libraries@ mailing list, the original proposer may be asked to moderate the libraries@ mailing list discussion So do I understand right: it's up to the base-library maintainer to decide whether a change is controversial and must to be discussed with CLC, which in can elevate it to wider discussion or not. The page however lists Edward Kmett and Ryan Scott as base-maintainers, which I'm pretty sure is not right. Who are the base maintainers? I'm sorry for my misunderstanding, it seems you are right Sandy, the issues should be discussed in the issue trackers first, and only elevated to libraries@ list if CLC decides it needs to! That is much more reasonable then going to the libraries@ directly for every issue. - Oleg On 7.7.2021 19.41, Sandy Maguire wrote:

...

At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann mailto:lemming@henning-thielemann.de> wrote:

On Wed, 7 Jul 2021, Oleg Grenrus wrote:

> For example > > - https://gitlab.haskell.org/ghc/ghc/-/issues/20044 https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration > from primitive to base > - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String > behavior > > Why they are discussed "in private", I thought libraries@ list is where > such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Carter Schonwald

1:37 p.m.

It’s probably worth looking at the wiki edit history too. I’m pretty sure some edits were done to the policy in the past 18 months without community feedback or discussion. On Wed, Jul 7, 2021 at 1:33 PM Oleg Grenrus wrote:

...

Thanks for this reply. It made me reread https://wiki.haskell.org/Library_submissions page. In the Guide to proposers section it says:

- All library proposals should start on the relevant issue tracker. - At this point, the library maintainer is responsible for taking next steps. - ... or decide that this is a controversial decision that must be discussed with the CLC.

- If the CLC decides that the discussion must be discussed with the libraries@ mailing list, the original proposer may be asked to moderate the libraries@ mailing list discussion

So do I understand right: it's up to the base-library maintainer to decide whether a change is controversial and must to be discussed with CLC, which in can elevate it to wider discussion or not.

The page however lists Edward Kmett and Ryan Scott as base-maintainers, which I'm pretty sure is not right. Who are the base maintainers?

I'm sorry for my misunderstanding, it seems you are right Sandy, the issues should be discussed in the issue trackers first, and only elevated to libraries@ list if CLC decides it needs to! That is much more reasonable then going to the libraries@ directly for every issue.

- Oleg

On 7.7.2021 19.41, Sandy Maguire wrote:

At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann < lemming@henning-thielemann.de> wrote:

...
On Wed, 7 Jul 2021, Oleg Grenrus wrote:

...
For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Oleg Grenrus

1:41 p.m.

There aren't. You are the last to edit that page: https://wiki.haskell.org/index.php?title=Library_submissions&action=history - Oleg On 7.7.2021 20.37, Carter Schonwald wrote:

...

It’s probably worth looking at the wiki edit history too.

I’m pretty sure some edits were done to the policy in the past 18 months without community feedback or discussion.

On Wed, Jul 7, 2021 at 1:33 PM Oleg Grenrus mailto:oleg.grenrus@iki.fi> wrote:

Thanks for this reply. It made me reread https://wiki.haskell.org/Library_submissions https://wiki.haskell.org/Library_submissions page. In the Guide to proposers section it says:

- All library proposals should start on the relevant issue tracker. - At this point, the library maintainer is responsible for taking next steps. - ... or decide that this is a controversial decision that must be discussed with the CLC.

- If the CLC decides that the discussion must be discussed with the libraries@ mailing list, the original proposer may be asked to moderate the libraries@ mailing list discussion

So do I understand right: it's up to the base-library maintainer to decide whether a change is controversial and must to be discussed with CLC, which in can elevate it to wider discussion or not.

The page however lists Edward Kmett and Ryan Scott as base-maintainers, which I'm pretty sure is not right. Who are the base maintainers?

I'm sorry for my misunderstanding, it seems you are right Sandy, the issues should be discussed in the issue trackers first, and only elevated to libraries@ list if CLC decides it needs to! That is much more reasonable then going to the libraries@ directly for every issue.

- Oleg

On 7.7.2021 19.41, Sandy Maguire wrote:

...
At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann mailto:lemming@henning-thielemann.de> wrote:

On Wed, 7 Jul 2021, Oleg Grenrus wrote:

> For example > > - https://gitlab.haskell.org/ghc/ghc/-/issues/20044 https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration > from primitive to base > - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String > behavior > > Why they are discussed "in private", I thought libraries@ list is where > such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Carter Schonwald

1:35 p.m.

Respectfully: some of us like mailing lists for complicated discussions :) The issue here is deviation from processs that support a wider range of participants with different accessibility needs and varying levels of volunteer time. More over, a mailing list is easier for curious parties to subscribe to and filter to a dedicated inbox. Have you ever tried doing robust mailbox filters for GitHub or gitlab? It’s pretty tricky without creating a firehose of all repo events unless you’re in a group that’s always tagged. Any ticket tagged core libraries on gitlab needs to have a corresponding email to this list for folks who aren’t administratively in that notification group. Or needs to link to a thread here motivating it. I’m on that notification list for gitlab because current clc folks wanted me to continue helping out, but that’s not a scalable open participation model. On Wed, Jul 7, 2021 at 12:41 PM Sandy Maguire wrote:

...

At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann < lemming@henning-thielemann.de> wrote:

...
On Wed, 7 Jul 2021, Oleg Grenrus wrote:

...
For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Oleg Grenrus

1:41 p.m.

As I understood from the https://wiki.haskell.org/Library_submissions page (to which GHC wiki links from https://gitlab.haskell.org/ghc/ghc/-/wikis/contributing#contributing-to-ghc), this job of elevating issue to the CLC is the job of a library maintainer. So I repeat my question: who are the current maintainer(s) of base library to make these calls. I'd like to know whom comments I can and cannot ignore when proposing changes, as everyone has an opinion on how base should be (myself including). - Oleg On 7.7.2021 20.35, Carter Schonwald wrote:

...

Respectfully: some of us like mailing lists for complicated discussions :)

The issue here is deviation from processs that support a wider range of participants with different accessibility needs and varying levels of volunteer time.

More over, a mailing list is easier for curious parties to subscribe to and filter to a dedicated inbox. Have you ever tried doing robust mailbox filters for GitHub or gitlab? It’s pretty tricky without creating a firehose of all repo events unless you’re in a group that’s always tagged.

Any ticket tagged core libraries on gitlab needs to have a corresponding email to this list for folks who aren’t administratively in that notification group. Or needs to link to a thread here motivating it.

I’m on that notification list for gitlab because current clc folks wanted me to continue helping out, but that’s not a scalable open participation model.

On Wed, Jul 7, 2021 at 12:41 PM Sandy Maguire mailto:sandy@sandymaguire.me> wrote:

At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann mailto:lemming@henning-thielemann.de> wrote:

On Wed, 7 Jul 2021, Oleg Grenrus wrote:

> For example > > - https://gitlab.haskell.org/ghc/ghc/-/issues/20044 https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration > from primitive to base > - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String > behavior > > Why they are discussed "in private", I thought libraries@ list is where > such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

chessai

1:45 p.m.

I (chessai) am the current maintainer for base. On Wed, Jul 7, 2021, 12:41 Oleg Grenrus wrote:

...

As I understood from the https://wiki.haskell.org/Library_submissions page (to which GHC wiki links from https://gitlab.haskell.org/ghc/ghc/-/wikis/contributing#contributing-to-ghc), this job of elevating issue to the CLC is the job of a library maintainer. So I repeat my question: who are the current maintainer(s) of base library to make these calls.

I'd like to know whom comments I can and cannot ignore when proposing changes, as everyone has an opinion on how base should be (myself including).

- Oleg On 7.7.2021 20.35, Carter Schonwald wrote:

Respectfully: some of us like mailing lists for complicated discussions :)

The issue here is deviation from processs that support a wider range of participants with different accessibility needs and varying levels of volunteer time.

More over, a mailing list is easier for curious parties to subscribe to and filter to a dedicated inbox. Have you ever tried doing robust mailbox filters for GitHub or gitlab? It’s pretty tricky without creating a firehose of all repo events unless you’re in a group that’s always tagged.

Any ticket tagged core libraries on gitlab needs to have a corresponding email to this list for folks who aren’t administratively in that notification group. Or needs to link to a thread here motivating it.

I’m on that notification list for gitlab because current clc folks wanted me to continue helping out, but that’s not a scalable open participation model.

On Wed, Jul 7, 2021 at 12:41 PM Sandy Maguire wrote:

...
At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann < lemming@henning-thielemann.de> wrote:

...
On Wed, 7 Jul 2021, Oleg Grenrus wrote:

...
For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing listLibraries@haskell.orghttp://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Oleg Grenrus

2:08 p.m.

Could the library submissions page be updated to reflect that. I think that maintainers of random, template-haskell, primitive are not correct either. I'm not sure whether text and bytestring should be on the list as well. Also GHC Trac issue-tracker links are dead. - Oleg On 7.7.2021 20.45, chessai wrote:

...

I (chessai) am the current maintainer for base.

On Wed, Jul 7, 2021, 12:41 Oleg Grenrus mailto:oleg.grenrus@iki.fi> wrote:

As I understood from the https://wiki.haskell.org/Library_submissions https://wiki.haskell.org/Library_submissions page (to which GHC wiki links from https://gitlab.haskell.org/ghc/ghc/-/wikis/contributing#contributing-to-ghc https://gitlab.haskell.org/ghc/ghc/-/wikis/contributing#contributing-to-ghc), this job of elevating issue to the CLC is the job of a library maintainer. So I repeat my question: who are the current maintainer(s) of base library to make these calls.

I'd like to know whom comments I can and cannot ignore when proposing changes, as everyone has an opinion on how base should be (myself including).

- Oleg

On 7.7.2021 20.35, Carter Schonwald wrote:

...
Respectfully: some of us like mailing lists for complicated discussions :)

The issue here is deviation from processs that support a wider range of participants with different accessibility needs and varying levels of volunteer time.

More over, a mailing list is easier for curious parties to subscribe to and filter to a dedicated inbox. Have you ever tried doing robust mailbox filters for GitHub or gitlab? It’s pretty tricky without creating a firehose of all repo events unless you’re in a group that’s always tagged.

Any ticket tagged core libraries on gitlab needs to have a corresponding email to this list for folks who aren’t administratively in that notification group. Or needs to link to a thread here motivating it.

I’m on that notification list for gitlab because current clc folks wanted me to continue helping out, but that’s not a scalable open participation model.

On Wed, Jul 7, 2021 at 12:41 PM Sandy Maguire mailto:sandy@sandymaguire.me> wrote:

At risk of being the messenger who gets shot....

As an outsider, it seems very reasonable to me to file a bug against the issue tracker for a project whose code I think should be changed. For better or worse, this is the way that 99% of software projects work. Expecting everyone in the community to know that they _shouldn't_ be filing bugs against the issue tracker is a losing battle. I'm more hooked in than most, and even I didn't know this.

I can empathize with things not being done the way you'd like to be, but the claim that things happening on the GHC tracker are done "in private" is silly. The gitlab tracker is 10x more accessible, and the lack of community engagement on the mailing lists speaks volumes.

And besides, nobody wants to be on a mailing list anyway. It's a terrible experience with weird branching and no persistence, and while there are archives, it's an extremely unpleasant thing to try to spelunk through.

Best, Sandy

On Wed, Jul 7, 2021 at 8:52 AM Henning Thielemann mailto:lemming@henning-thielemann.de> wrote:

On Wed, 7 Jul 2021, Oleg Grenrus wrote:

> For example > > - https://gitlab.haskell.org/ghc/ghc/-/issues/20044 https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration > from primitive to base > - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String > behavior > > Why they are discussed "in private", I thought libraries@ list is where > such changes should be discussed.

I think so, too, and I missed them as well. _______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

chessai

1:52 p.m.

FWIW, I do believe that changing Show @String needs to be discussed on the list and I pinged the proposer to do so. The ByteArray change was in my/Andrew Martin's opinion that it was straightforward and sensible enough to not need the mailing list. As has been pointed out by Oleg, on the library submissions page, not all changes to any core library need to be discussed on the mailing list, as that would be a gross inefficiency and waste of time. Issue trackers are better, with escalation when necessary. On Wed, Jul 7, 2021, 10:12 Oleg Grenrus wrote:

...

For example

- https://gitlab.haskell.org/ghc/ghc/-/issues/20044 ByteArray migration from primitive to base - https://gitlab.haskell.org/ghc/ghc/-/issues/20027 Changing Show String behavior

Why they are discussed "in private", I thought libraries@ list is where such changes should be discussed.

- Oleg

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

1413

Age (days ago)

1413

Last active (days ago)

List overview

Download

12 comments

6 participants

participants (6)

Carter Schonwald
chessai
Henning Thielemann
Oleg Grenrus
Sandy Maguire
Taylor Fausak