All of lore.kernel.org
 help / color / mirror / Atom feed
* Proposal: rename xen.git#master (to #trunk, perhaps)
@ 2020-06-24 16:13 Ian Jackson
  2020-06-24 17:38 ` Stefano Stabellini
  2020-06-24 23:10 ` Andrew Cooper
  0 siblings, 2 replies; 7+ messages in thread
From: Ian Jackson @ 2020-06-24 16:13 UTC (permalink / raw)
  To: xen-devel; +Cc: committers

I think it would be a good idea to rename this branch name.  This name
has unfortunate associations[1], even if it can be argued[2] that the
etymology is not as bad as in some uses of the word.

This is relativity straightforward on a technical level and will
involve a minimum of inconvenience.  Since only osstest ever pushes to
xen.git#master, we could easily make a new branch name and also keep
#master for compatibility as long as we like.

The effects[1] would be:

Users who did "git clone https://xenbits.xen.org/git-http/xen.git""
would find themselves on a branch called "trunk" which tracked
"origin/trunk", by default.  (Some users with old versions of git
using old protocols would still end up on "master".)

Everyone who currently tracks "master" would be able to switch to
tracking "trunk" at their leisure.

Presumably at some future point (a year or two from now, say) we would
abolish the name "master".

Comments ?  In particular, comments on:

1. What the new branch name should be called.  Suggestions I have seen
include "trunk" and "main".  I suggest "trunk" because this was used
by SVN, CVS, RCS, CSSC (and therefore probably SCCS) for this same
purpose.

2. Do we want to set a time now when the old name will be removed ?
I think 12 months would be good but I am open to arguments.

In any case in my capacity as osstest maintainer I plan to do
something like this.  The starting point is rather different.  There
will have to be an announcement about that, but I thought I would see
what people thought about xen.git before pressing ahead there.

Thanks,
Ian.

[1] It seems that for a significant number of people the word reminds
them of human slavery.  This seems undesirable if we can easily avoid
it, if we can.

[2] The precise etymology of "master" in this context is unclear.  It
appears to have come from BitKeeper originally.  In any case the
etymology is less important than the connotations.


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Proposal: rename xen.git#master (to #trunk, perhaps)
  2020-06-24 16:13 Proposal: rename xen.git#master (to #trunk, perhaps) Ian Jackson
@ 2020-06-24 17:38 ` Stefano Stabellini
  2020-06-24 18:40   ` Roman Shaposhnik
                     ` (2 more replies)
  2020-06-24 23:10 ` Andrew Cooper
  1 sibling, 3 replies; 7+ messages in thread
From: Stefano Stabellini @ 2020-06-24 17:38 UTC (permalink / raw)
  To: Ian Jackson; +Cc: xen-devel, committers

On Wed, 24 Jun 2020, Ian Jackson wrote:
> I think it would be a good idea to rename this branch name.  This name
> has unfortunate associations[1], even if it can be argued[2] that the
> etymology is not as bad as in some uses of the word.
> 
> This is relativity straightforward on a technical level and will
> involve a minimum of inconvenience.  Since only osstest ever pushes to
> xen.git#master, we could easily make a new branch name and also keep
> #master for compatibility as long as we like.
> 
> The effects[1] would be:
> 
> Users who did "git clone https://xenbits.xen.org/git-http/xen.git""
> would find themselves on a branch called "trunk" which tracked
> "origin/trunk", by default.  (Some users with old versions of git
> using old protocols would still end up on "master".)
> 
> Everyone who currently tracks "master" would be able to switch to
> tracking "trunk" at their leisure.
> 
> Presumably at some future point (a year or two from now, say) we would
> abolish the name "master".
> 
> Comments ?  In particular, comments on:
> 
> 1. What the new branch name should be called.  Suggestions I have seen
> include "trunk" and "main".  I suggest "trunk" because this was used
> by SVN, CVS, RCS, CSSC (and therefore probably SCCS) for this same
> purpose.

Github seems to be about to make a similar change. I wonder if we should
wait just a couple of weeks to see what name they are going to choose.

https://www.theregister.com/2020/06/15/github_replaces_master_with_main/


Of course I don't particulalry care one way or the other, but it would
be good if we end up using the same name as everybody else. It is not
that we have to choose the name Github is going to choose, but their
user base is massive -- whatever they are going to pick is very likely
going to stick.



> 2. Do we want to set a time now when the old name will be removed ?
> I think 12 months would be good but I am open to arguments.
> 
> In any case in my capacity as osstest maintainer I plan to do
> something like this.  The starting point is rather different.  There
> will have to be an announcement about that, but I thought I would see
> what people thought about xen.git before pressing ahead there.
> 
> Thanks,
> Ian.
> 
> [1] It seems that for a significant number of people the word reminds
> them of human slavery.  This seems undesirable if we can easily avoid
> it, if we can.
> 
> [2] The precise etymology of "master" in this context is unclear.  It
> appears to have come from BitKeeper originally.  In any case the
> etymology is less important than the connotations.
> 


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Proposal: rename xen.git#master (to #trunk, perhaps)
  2020-06-24 17:38 ` Stefano Stabellini
@ 2020-06-24 18:40   ` Roman Shaposhnik
  2020-06-25  8:14   ` Bertrand Marquis
  2020-07-20 15:04   ` Julien Grall
  2 siblings, 0 replies; 7+ messages in thread
From: Roman Shaposhnik @ 2020-06-24 18:40 UTC (permalink / raw)
  To: Stefano Stabellini; +Cc: Ian Jackson, committers, Xen-devel

On Wed, Jun 24, 2020 at 10:39 AM Stefano Stabellini
<sstabellini@kernel.org> wrote:
>
> On Wed, 24 Jun 2020, Ian Jackson wrote:
> > I think it would be a good idea to rename this branch name.  This name
> > has unfortunate associations[1], even if it can be argued[2] that the
> > etymology is not as bad as in some uses of the word.
> >
> > This is relativity straightforward on a technical level and will
> > involve a minimum of inconvenience.  Since only osstest ever pushes to
> > xen.git#master, we could easily make a new branch name and also keep
> > #master for compatibility as long as we like.
> >
> > The effects[1] would be:
> >
> > Users who did "git clone https://xenbits.xen.org/git-http/xen.git""
> > would find themselves on a branch called "trunk" which tracked
> > "origin/trunk", by default.  (Some users with old versions of git
> > using old protocols would still end up on "master".)
> >
> > Everyone who currently tracks "master" would be able to switch to
> > tracking "trunk" at their leisure.
> >
> > Presumably at some future point (a year or two from now, say) we would
> > abolish the name "master".
> >
> > Comments ?  In particular, comments on:
> >
> > 1. What the new branch name should be called.  Suggestions I have seen
> > include "trunk" and "main".  I suggest "trunk" because this was used
> > by SVN, CVS, RCS, CSSC (and therefore probably SCCS) for this same
> > purpose.
>
> Github seems to be about to make a similar change. I wonder if we should
> wait just a couple of weeks to see what name they are going to choose.
>
> https://www.theregister.com/2020/06/15/github_replaces_master_with_main/
>
>
> Of course I don't particulalry care one way or the other, but it would
> be good if we end up using the same name as everybody else. It is not
> that we have to choose the name Github is going to choose, but their
> user base is massive -- whatever they are going to pick is very likely
> going to stick.

<peanut gallery>
+1 to Stefano's way of thinking here -- that's exactly what we're
doing in some of the other LF projects I'm involved in. Of course all
the master/slave terminology has to be addresses at the project level
-- but I haven't come across much of that in my Xen hacking
</peanut gallery>

Thanks,
Roman.


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Proposal: rename xen.git#master (to #trunk, perhaps)
  2020-06-24 16:13 Proposal: rename xen.git#master (to #trunk, perhaps) Ian Jackson
  2020-06-24 17:38 ` Stefano Stabellini
@ 2020-06-24 23:10 ` Andrew Cooper
  2020-06-25 11:11   ` George Dunlap
  1 sibling, 1 reply; 7+ messages in thread
From: Andrew Cooper @ 2020-06-24 23:10 UTC (permalink / raw)
  To: Ian Jackson, xen-devel; +Cc: committers

On 24/06/2020 17:13, Ian Jackson wrote:
> I think it would be a good idea to rename this branch name.

The tech industry does use some problematic terms, and I fully agree
with taking steps to reduce their use.  However, I disagree that this is
a problematic context.

In the English language, context is paramount.

Terms such as master/slave have an obvious inequality bias in the
context in which they are used.  There are alternatives to these terms,
probably one of which is more suited to the specific scenario in question.

However, the word master is a very overloaded word in terms of its
different meanings.

Describing someone as a "master of their trade/skill/other", is a
totally different context, and it would be excessive to suggest changing
the language used in this way.  So too, in my opinion, is master as in
"master copy", a different context and connotation to master as in
master/slave.


A much more meaningful use of time would be to address:

xen.git$ git grep -i -e slave -e whitelist -e blacklist | wc -l
194

two thirds of which look fairly easy to address, and one third fairly
complicated due to external dependencies.

~Andrew


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Proposal: rename xen.git#master (to #trunk, perhaps)
  2020-06-24 17:38 ` Stefano Stabellini
  2020-06-24 18:40   ` Roman Shaposhnik
@ 2020-06-25  8:14   ` Bertrand Marquis
  2020-07-20 15:04   ` Julien Grall
  2 siblings, 0 replies; 7+ messages in thread
From: Bertrand Marquis @ 2020-06-25  8:14 UTC (permalink / raw)
  To: Stefano Stabellini; +Cc: Ian Jackson, nd, committers, Xen-devel



> On 24 Jun 2020, at 18:38, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Wed, 24 Jun 2020, Ian Jackson wrote:
>> I think it would be a good idea to rename this branch name.  This name
>> has unfortunate associations[1], even if it can be argued[2] that the
>> etymology is not as bad as in some uses of the word.
>> 
>> This is relativity straightforward on a technical level and will
>> involve a minimum of inconvenience.  Since only osstest ever pushes to
>> xen.git#master, we could easily make a new branch name and also keep
>> #master for compatibility as long as we like.
>> 
>> The effects[1] would be:
>> 
>> Users who did "git clone https://xenbits.xen.org/git-http/xen.git""
>> would find themselves on a branch called "trunk" which tracked
>> "origin/trunk", by default.  (Some users with old versions of git
>> using old protocols would still end up on "master".)
>> 
>> Everyone who currently tracks "master" would be able to switch to
>> tracking "trunk" at their leisure.
>> 
>> Presumably at some future point (a year or two from now, say) we would
>> abolish the name "master".
>> 
>> Comments ?  In particular, comments on:
>> 
>> 1. What the new branch name should be called.  Suggestions I have seen
>> include "trunk" and "main".  I suggest "trunk" because this was used
>> by SVN, CVS, RCS, CSSC (and therefore probably SCCS) for this same
>> purpose.
> 
> Github seems to be about to make a similar change. I wonder if we should
> wait just a couple of weeks to see what name they are going to choose.
> 
> https://www.theregister.com/2020/06/15/github_replaces_master_with_main/
> 
> 
> Of course I don't particulalry care one way or the other, but it would
> be good if we end up using the same name as everybody else. It is not
> that we have to choose the name Github is going to choose, but their
> user base is massive -- whatever they are going to pick is very likely
> going to stick.

+1 to stefano
Whatever choice is made here it would be better to follow standards to prevent
user from being lost. And github is definitely a bit actor :-)

Bertrand

> 
> 
> 
>> 2. Do we want to set a time now when the old name will be removed ?
>> I think 12 months would be good but I am open to arguments.
>> 
>> In any case in my capacity as osstest maintainer I plan to do
>> something like this.  The starting point is rather different.  There
>> will have to be an announcement about that, but I thought I would see
>> what people thought about xen.git before pressing ahead there.
>> 
>> Thanks,
>> Ian.
>> 
>> [1] It seems that for a significant number of people the word reminds
>> them of human slavery.  This seems undesirable if we can easily avoid
>> it, if we can.
>> 
>> [2] The precise etymology of "master" in this context is unclear.  It
>> appears to have come from BitKeeper originally.  In any case the
>> etymology is less important than the connotations.



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Proposal: rename xen.git#master (to #trunk, perhaps)
  2020-06-24 23:10 ` Andrew Cooper
@ 2020-06-25 11:11   ` George Dunlap
  0 siblings, 0 replies; 7+ messages in thread
From: George Dunlap @ 2020-06-25 11:11 UTC (permalink / raw)
  To: Andrew Cooper; +Cc: Ian Jackson, committers, xen-devel



> On Jun 25, 2020, at 12:10 AM, Andrew Cooper <andrew.cooper3@citrix.com> wrote:
> 
> On 24/06/2020 17:13, Ian Jackson wrote:
>> I think it would be a good idea to rename this branch name.

[snip]

> Describing someone as a "master of their trade/skill/other", is a
> totally different context, and it would be excessive to suggest changing
> the language used in this way.  So too, in my opinion, is master as in
> "master copy", a different context and connotation to master as in
> master/slave.

Ian already noted that there was a question about the etymology of the name, but argued that we should change the name anyway.  With my committer hat on, I agree.

We could have a long discussion about the origin of various terms, and how well each one applies to our main development branch.  (For instance, if “master /copy” is the real etymology of the git master branch, I would argue that it was used inappropriately: “master / copy” is most appropriate in a situation where the thing being copied is nearly perfect and is rarely changed.  That’s certainly not true of our master branch.)

But reading the weather, I have to conclude that our industry is going in this direction, whether we like it or not.  (See for instance Jonathan Corbet’s opinion piece on the subject [1], and the subsequent discussion.)  At that point, the continued use of the word ‘master’ is no longer neutral: it will become a deliberate choice which will begin to communicate something which we don’t want to communicate.

There are times when standing against the tide is the right thing to do; but I don’t think this is one of them.

[1] https://lwn.net/Articles/823224/

> A much more meaningful use of time would be to address:
> 
> xen.git$ git grep -i -e slave -e whitelist -e blacklist | wc -l
> 194

Well, an even *more* meaningful use of our time might be for each of us to look into the behavior of the police forces over which we have influence, and consider whether we need to advocate for any systematic changes in the way they’re run.

Luckily, we don’t have to choose — we can do all of them. :-)

 -George

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Proposal: rename xen.git#master (to #trunk, perhaps)
  2020-06-24 17:38 ` Stefano Stabellini
  2020-06-24 18:40   ` Roman Shaposhnik
  2020-06-25  8:14   ` Bertrand Marquis
@ 2020-07-20 15:04   ` Julien Grall
  2 siblings, 0 replies; 7+ messages in thread
From: Julien Grall @ 2020-07-20 15:04 UTC (permalink / raw)
  To: Stefano Stabellini, Ian Jackson; +Cc: xen-devel, committers

Hi,

On 24/06/2020 18:38, Stefano Stabellini wrote:
> On Wed, 24 Jun 2020, Ian Jackson wrote:
>> I think it would be a good idea to rename this branch name.  This name
>> has unfortunate associations[1], even if it can be argued[2] that the
>> etymology is not as bad as in some uses of the word.
>>
>> This is relativity straightforward on a technical level and will
>> involve a minimum of inconvenience.  Since only osstest ever pushes to
>> xen.git#master, we could easily make a new branch name and also keep
>> #master for compatibility as long as we like.
>>
>> The effects[1] would be:
>>
>> Users who did "git clone https://xenbits.xen.org/git-http/xen.git""
>> would find themselves on a branch called "trunk" which tracked
>> "origin/trunk", by default.  (Some users with old versions of git
>> using old protocols would still end up on "master".)
>>
>> Everyone who currently tracks "master" would be able to switch to
>> tracking "trunk" at their leisure.
>>
>> Presumably at some future point (a year or two from now, say) we would
>> abolish the name "master".
>>
>> Comments ?  In particular, comments on:
>>
>> 1. What the new branch name should be called.  Suggestions I have seen
>> include "trunk" and "main".  I suggest "trunk" because this was used
>> by SVN, CVS, RCS, CSSC (and therefore probably SCCS) for this same
>> purpose.
> 
> Github seems to be about to make a similar change. I wonder if we should
> wait just a couple of weeks to see what name they are going to choose.

I have just tried to create a new repo on github. It looks like the 
default is still 'master' so far.

> 
> https://www.theregister.com/2020/06/15/github_replaces_master_with_main/
> 
> 
> Of course I don't particulalry care one way or the other, but it would
> be good if we end up using the same name as everybody else. It is not
> that we have to choose the name Github is going to choose, but their
> user base is massive -- whatever they are going to pick is very likely
> going to stick.

+1

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2020-07-20 15:05 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-06-24 16:13 Proposal: rename xen.git#master (to #trunk, perhaps) Ian Jackson
2020-06-24 17:38 ` Stefano Stabellini
2020-06-24 18:40   ` Roman Shaposhnik
2020-06-25  8:14   ` Bertrand Marquis
2020-07-20 15:04   ` Julien Grall
2020-06-24 23:10 ` Andrew Cooper
2020-06-25 11:11   ` George Dunlap

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.