Master/slave trolling pull request accepted to django master branch

2,835 views
Skip to first unread message

Meira

unread,
May 27, 2014, 8:14:43 AM5/27/14
to django-d...@googlegroups.com
As some of you may have notice, a hot discussion is happening in the comments of this pull request: https://github.com/django/django/pull/2692
Essentially, this pull request suggests that all occurences of master/slave be replaced with leader/follower. While this is clearly insane, a less jaw-dropping, but still weird change was made in commit https://github.com/django/django/commit/beec05686ccc3bee8461f9a5a02c607a02352ae1

Many users in comments to the original pull request agreed that primary/replica is not a good word choice, is vague and misleading. Current django docs compensate for the confusion by referring to "master/slave" in parentheses after mentioning "primary/replica". Of course, this change is nothing more than cosmetical, but it still carries more downsides than upsides.
Master/slave is immediately obvious for the experienced users, and easily googleable for the newbies.

I reverted the change and sent a pull request https://github.com/django/django/pull/2720. In the corresponding ticket, I was told to "wait 6 months" and then resubmit the ticket. (https://code.djangoproject.com/ticket/22707), and the pull request was closed immediately with an advice to start a discussion on mailing list. So that's what I'm doing here :)

I sum up my personal point of view in this comment: https://github.com/django/django/pull/2692#issuecomment-44265422

Of course, it'll be hard for the django maintainers to admit their mistake and revert the change. It's always hard to admit mistakes, but it's better than leaving it how it is.

Kai

unread,
May 27, 2014, 8:26:52 AM5/27/14
to django-d...@googlegroups.com
What is so bad about removing terms like master/slave that are related to or even originate from so much suffering and injustice and replacing it with neutral terms? Primary/Replica is used in many DB systems too.

alTus

unread,
May 27, 2014, 8:29:12 AM5/27/14
to django-d...@googlegroups.com
It took 7 minutes and 23 seconds to merge this troll PR without any discussion and now Meira is suggested to wait 6 months? But what can happen?

вторник, 27 мая 2014 г., 16:14:43 UTC+4 пользователь Meira написал:

Daniele Procida

unread,
May 27, 2014, 8:35:26 AM5/27/14
to django-d...@googlegroups.com
On Tue, May 27, 2014, Meira <poo...@gmail.com> wrote:

>As some of you may have notice, a hot discussion is happening in the
>comments of this pull request: https://github.com/django/django/pull/2692

If by "hot discussion" you mean silly pictures and noisy accusations...

There is a discussion to be had about this subject, which is why you were invited to start one.

You're clearly very angry about it, I don't know why, but if you'd like other people to participate in your discussion I suggest you stop making uncivil remarks in it - like "trolling" and "insane" - and state your case in a way that makes people feel like discussing it with you.

Please understand that if you're not able to moderate your own tone, you will not be permitted to post to this email list.

Daniele

Meira

unread,
May 27, 2014, 9:29:35 AM5/27/14
to django-d...@googlegroups.com
Sorry, I accidentally sent a private reply :) I'll try to repeat it here for others.

Those silly pictures are the community's emotional reply to an issue that they care about. I don't think calling the contributors "silly" is exactly politically correct, too, since we are on that level now :)

I requested to revert the changes for the exact reason that you're giving me: the discussion should happen before any changes are made.

I call the original change "trolling", because that's what I believe it is. This is not the first case of political correctness trolling on github, FSF in particular has a history of doing that with other projects. I do not believe any of the "political correct" changes were made with the goal of bettering the project, I think the only goal of such is to start a holywar discussion and have fun. Multiple comments on github assert that I'm not the only one aware of the spreading trolling, particularly from 4chan users.
The community is trying to protect the django project from the attack of people who seek no good for django. Please stop the confrontation, this will not help anyone. The feedback from the community is the most valuable thing django has, do not ignore it.

Offtopic:
Calling me angry, uncivil, referring to my way of putting things as inappropriate, suggesting that my tone might disengage people - those are signs of your own anger, if I may suggest. I made no assumptions about you personally or any other of the django maintainers, and I would appreciate if you adopt a similar approach.
I argue that stating my opinion honestly and persistently is not a sign of me being angry. You can call me harsh if that's what you think. I see no point in avoiding words that pinpoint exactly what I mean.

Daniele Procida

unread,
May 27, 2014, 9:44:13 AM5/27/14
to django-d...@googlegroups.com
On Tue, May 27, 2014, Meira <poo...@gmail.com> wrote:

>Sorry, I accidentally sent a private reply :) I'll try to repeat it here
>for others.

I have replied, privately, but I wanted to add publicly:

>The community is trying to protect the django project from the attack of
>people who seek no good for django. Please stop the confrontation, this
>will not help anyone. The feedback from the community is the most valuable
>thing django has, do not ignore it.

We won't - I promise. The feedback - even when it's expressed testily or in exasperation - is not just appreciated, but received gratefully. The same goes for patches or contributions, *including the ones that are not accepted*.

The Django Project and the Django community are not in my mind two different things.

>Calling me angry, uncivil, referring to my way of putting things as
>inappropriate, suggesting that my tone might disengage people - those are
>signs of your own anger, if I may suggest. I made no assumptions about you
>personally or any other of the django maintainers, and I would appreciate
>if you adopt a similar approach.

You are right and I apologise for making those remarks.

Daniele

Meira

unread,
May 27, 2014, 10:24:51 AM5/27/14
to django-d...@googlegroups.com
I appreciate your reply very much! And sure it's not wise to rename things every time someone asks for it, even when it's a lot of people. But same applies to the original renaming commit, doesnt it?

I would suggest that leaving names the way they have been since a long time is the best option. I.e. revert the changes made, give the community a little time to think about it, and then incorporate the change most people can agree on.
Saying the change has been made and now it's too late defies the very concept of version control. Reminder: we are using git. It can never be "too late" :)

Florian Apolloner

unread,
May 27, 2014, 11:13:43 AM5/27/14
to django-d...@googlegroups.com


On Tuesday, May 27, 2014 4:24:51 PM UTC+2, Meira wrote:
I appreciate your reply very much! And sure it's not wise to rename things every time someone asks for it, even when it's a lot of people. But same applies to the original renaming commit, doesnt it?

Yes and no, we trust our committers to make decisions about individual patches; if we don't agree on it we either revert it later on or adjust it so it matches our expectations (which we did in a follow up).
 
Saying the change has been made and now it's too late defies the very concept of version control. Reminder: we are using git. It can never be "too late" :)

Nobody is saying it's too late change it; we are aware that we could change it if we wanted to do so.

Florian

Aymeric Augustin

unread,
May 27, 2014, 11:24:37 AM5/27/14
to django-d...@googlegroups.com
In the interest of giving the full story to those who're genuinely worried that core devs don't give a fuck about the community — community being defined as the people who discovered this change on django-updates, not on 4chan or Hacker News...

> Le 27 mai 2014 à 16:24, Meira <poo...@gmail.com> a écrit :
>
> I would suggest that leaving names the way they have been since a long time is the best option. I.e. revert the changes made, give the community a little time to think about it, and then incorporate the change most people can agree on.

That's what happened just after the initial commit. A few days later the naming was changed to "primary / replica".

This second commit was discussed in a Trac ticket and everyone (even you!) was welcome to give their opinion.

IIRC 7 core devs weighed in on the Trac ticket or on IRC, making it the most discussed change in Diango this year.

Whatever happened after the trolling began doesn't matter. At this point it's impossible to tell concern from trolling.

Given that the current wording reflects consensus in the core team, you can assume that it won't be changed.

That's why your pull request was rejected.

--
Aymeric.




Meira

unread,
May 27, 2014, 11:38:23 AM5/27/14
to django-d...@googlegroups.com
This second commit was discussed in a Trac ticket and everyone (even you!) was welcome to give their opinion.

That's all nice and good, but why is the discussion taking the course of whether or not we're accepting the second commit? It is clearly better than the first. The question is, why the first commit was accepted.

Whatever happened after the trolling began doesn't matter. At this point it's impossible to tell concern from trolling.
Well, I'm claiming that the first commit that proposed the leader/follower pair was trolling.
 
Given that the current wording reflects consensus in the core team, you can assume that it won't be changed.
Given that the current wording is less clear than "master/slave" and many community member are against this change, I think that the issue is worth discussing a bit more.
(Here is a comment confirming the assumption: https://github.com/django/django/pull/2694#discussion_r12865261)

Alex Gaynor

unread,
May 27, 2014, 11:40:58 AM5/27/14
to django-d...@googlegroups.com
On Tue, May 27, 2014 at 10:38 AM, Meira <poo...@gmail.com> wrote:
This second commit was discussed in a Trac ticket and everyone (even you!) was welcome to give their opinion.

That's all nice and good, but why is the discussion taking the course of whether or not we're accepting the second commit? It is clearly better than the first. The question is, why the first commit was accepted.

Whatever happened after the trolling began doesn't matter. At this point it's impossible to tell concern from trolling.
Well, I'm claiming that the first commit that proposed the leader/follower pair was trolling.
 

It's not, the person who submitted that PR has been a contributing member of the Django community for years.
 
Given that the current wording reflects consensus in the core team, you can assume that it won't be changed.
Given that the current wording is less clear than "master/slave" and many community member are against this change, I think that the issue is worth discussing a bit more.
(Here is a comment confirming the assumption: https://github.com/django/django/pull/2694#discussion_r12865261)

Many people from 4chan are trolling, please don't confuse them. 

--
You received this message because you are subscribed to the Google Groups "Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.
To view this discussion on the web visit https://groups.google.com/d/msgid/django-developers/aaee8850-e135-44fd-8821-d0c3bd8fe423%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Alex

--
"I disapprove of what you say, but I will defend to the death your right to say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
"The people's good is the highest law." -- Cicero
GPG Key fingerprint: 125F 5C67 DFE9 4084

Florian Apolloner

unread,
May 27, 2014, 11:46:07 AM5/27/14
to django-d...@googlegroups.com


On Tuesday, May 27, 2014 5:38:23 PM UTC+2, Meira wrote:
This second commit was discussed in a Trac ticket and everyone (even you!) was welcome to give their opinion.

That's all nice and good, but why is the discussion taking the course of whether or not we're accepting the second commit? It is clearly better than the first. The question is, why the first commit was accepted.

Cause a committer (Alex in that case) agreed with the rationale outlined in the PR.
 
Given that the current wording reflects consensus in the core team, you can assume that it won't be changed.
Given that the current wording is less clear than "master/slave" and many community member are against this change, I think that the issue is worth discussing a bit more.
(Here is a comment confirming the assumption: https://github.com/django/django/pull/2694#discussion_r12865261)

To be honest, looking at the PR the "many community members" probably reduce to a number countable with all of my fingers. And even if not, "many" are okay with the change or like it -- and I strongly doubt that we will find a real consensus here…

Cheers,
Florian

Daniele Procida

unread,
May 27, 2014, 11:47:02 AM5/27/14
to django-d...@googlegroups.com
On Tue, May 27, 2014, Meira <poo...@gmail.com> wrote:

>> This second commit was discussed in a Trac ticket and everyone (even you!)
>> was welcome to give their opinion.
>>
>
>That's all nice and good, but why is the discussion taking the course of
>whether or not we're accepting the second commit? It is clearly better than
>the first. The question is, why the first commit was accepted.

A human being saw the patch, made the judgement in good faith that it should be accepted into core, and merged it.

I don't really see why you say that "why" is the question - it doesn't seem a very important one.

Daniele

Meira

unread,
May 27, 2014, 12:07:42 PM5/27/14
to django-d...@googlegroups.com
It seems to be, there are enough reasonable people leaving comments like this one: https://github.com/django/django/pull/2720#issuecomment-44296843

James Bennett

unread,
May 27, 2014, 12:12:57 PM5/27/14
to django-d...@googlegroups.com
On Tue, May 27, 2014 at 12:07 PM, Meira <poo...@gmail.com> wrote:
It seems to be, there are enough reasonable people leaving comments like this one: https://github.com/django/django/pull/2720#issuecomment-44296843

We'll just get the databases to change their terminology before we change ours!

Of course, the DB folks will get reasonable comments saying they should wait until the tools that integrate with the DBs change *their* documentation...

And thus we have an example of a mutually-recursive function. *Somebody* has to move first, and actually move rather than say "I'll move when those other people do". We decided to go ahead and move on this one.

For the record I am strongly against reverting at this point -- and I'm saying that as someone whose initial reaction to the pull request was "oh, a troll" until I looked more into who had submitted it. With well-written documentation, any confusion that's produced should be slight and momentary, and I'm OK with that as the tradeoff of using less-charged language.

Hannu Krosing

unread,
May 27, 2014, 12:20:30 PM5/27/14
to django-d...@googlegroups.com
On 05/27/2014 06:07 PM, Meira wrote:
It seems to be, there are enough reasonable people leaving comments like this one: https://github.com/django/django/pull/2720#issuecomment-44296843
Hi Meira

Unfortunately I have to agree, that calling some people "primaries" and some "replicas" is a serious insult to both.

Cheers
Hannu

On Tuesday, May 27, 2014 10:47:02 PM UTC+7, Daniele Procida wrote:
On Tue, May 27, 2014, Meira <poo...@gmail.com> wrote:

>> This second commit was discussed in a Trac ticket and everyone (even you!)
>> was welcome to give their opinion.
>>
>
>That's all nice and good, but why is the discussion taking the course of
>whether or not we're accepting the second commit? It is clearly better than
>the first. The question is, why the first commit was accepted.

A human being saw the patch, made the judgement in good faith that it should be accepted into core, and merged it.

I don't really see why you say that "why" is the question - it doesn't seem a very important one.

Daniele

--
You received this message because you are subscribed to the Google Groups "Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.

Andromeda Yelton

unread,
May 27, 2014, 12:44:58 PM5/27/14
to django-d...@googlegroups.com
On Tue, May 27, 2014 at 11:46 AM, Florian Apolloner <f.apo...@gmail.com> wrote:
To be honest, looking at the PR the "many community members" probably reduce to a number countable with all of my fingers.

Of the first 150 distinct commenters, 120 support the change (including everyone who is recognizably a person of color), 23 oppose, and 7 take unclear stances.  (After that point it starts degenerating into too much 4chan for me to bother counting.)

Which is a little beside the point as the process for merging PRs is not, in fact, democracy.  But is also fantastic, because I've spent the last week reading TRAC and hanging out here and talking to lots of people trying to figure out if Django will be a safe place for me to contribute.  And when I see that large a fraction of commenters come down on the side of inclusivity, I feel like "django developer" is a hat I can put on.

Cheers. \o/ \o/ \o/

Andromeda

Daniele Procida

unread,
May 27, 2014, 12:54:16 PM5/27/14
to Django Developers
On Tue, May 27, 2014, Andromeda Yelton <andromed...@gmail.com> wrote:

>Which is a little beside the point as the process for merging PRs is not,
>in fact, democracy. But is also fantastic, because I've spent the last
>week reading TRAC and hanging out here and talking to lots of people trying
>to figure out if Django will be a safe place for me to contribute. And
>when I see that large a fraction of commenters come down on the side of
>inclusivity, I feel like "django developer" is a hat I can put on.

If we get a single more person contributing to Django as a result, I would consider this whole episode as being entirely worth it.

Not that I think it's a sustainable strategy in the long term, of course.

Daniele

Meira

unread,
May 27, 2014, 1:14:49 PM5/27/14
to django-d...@googlegroups.com
I think it makes more sense to count reasonable arguments of both sides, not the people who thumb up in the comments (by the way, those who thumb up are mostly Americans, isn't that discrimination?)
If using the word "slave" is immediately associated with racism, it's a sign that we might have too many Americans in the topic, because for me, "slave" is not equal to "black slave". Maybe it's because we have too many bears and too few black folks in my country, but we did have (practically) slavery, too.

I also have a problem with the phrase "inclusive language". Who exactly was "included" by this change? I highly doubt that there was a slave who started using django after the change. It seems to me, it's the American historical guilt playing a huge role here.

It's an old misconception, it seems that if we change the words, we'll change the reality. By banning the word "slave", you cannot cancel the fact that for many years, black people in the US were treated worse than animals. I don't think that an attempt to forget that fact by aggressively labeling words as "racist" is "inclusive" or "positive".
I actually think that remembering bad things that are now history should encourage people to be a little nicer to each other at the moment.

We had slavery, and now we don't. It has nothing to do with databases :)

Alex Gaynor

unread,
May 27, 2014, 1:16:27 PM5/27/14
to django-d...@googlegroups.com
On Tue, May 27, 2014 at 12:14 PM, Meira <poo...@gmail.com> wrote:
I think it makes more sense to count reasonable arguments of both sides, not the people who thumb up in the comments (by the way, those who thumb up are mostly Americans, isn't that discrimination?)
If using the word "slave" is immediately associated with racism, it's a sign that we might have too many Americans in the topic, because for me, "slave" is not equal to "black slave". Maybe it's because we have too many bears and too few black folks in my country, but we did have (practically) slavery, too.

I also have a problem with the phrase "inclusive language". Who exactly was "included" by this change? I highly doubt that there was a slave who started using django after the change. It seems to me, it's the American historical guilt playing a huge role here.

It's an old misconception, it seems that if we change the words, we'll change the reality. By banning the word "slave", you cannot cancel the fact that for many years, black people in the US were treated worse than animals. I don't think that an attempt to forget that fact by aggressively labeling words as "racist" is "inclusive" or "positive".
I actually think that remembering bad things that are now history should encourage people to be a little nicer to each other at the moment.

We had slavery, and now we don't. It has nothing to do with databases :)





On Tuesday, May 27, 2014 11:54:16 PM UTC+7, Daniele Procida wrote:
On Tue, May 27, 2014, Andromeda Yelton <andromed...@gmail.com> wrote:

>Which is a little beside the point as the process for merging PRs is not,
>in fact, democracy.  But is also fantastic, because I've spent the last
>week reading TRAC and hanging out here and talking to lots of people trying
>to figure out if Django will be a safe place for me to contribute.  And
>when I see that large a fraction of commenters come down on the side of
>inclusivity, I feel like "django developer" is a hat I can put on.

If we get a single more person contributing to Django as a result, I would consider this whole episode as being entirely worth it.

Not that I think it's a sustainable strategy in the long term, of course.

Daniele

--
You received this message because you are subscribed to the Google Groups "Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.

For more options, visit https://groups.google.com/d/optout.

Meira

unread,
May 27, 2014, 1:21:42 PM5/27/14
to django-d...@googlegroups.com
I meant legally, of course. It is illegal now. Should we ban the word "drugstore" too, maybe?

I previously pointed out that I'm aware of the fact that there still is slavery in one form or another. I also mentioned that I don't believe this change made django more attractive for any of the current slaves.
Not even single slave's life was in any way improved by a free person complaining about a server being called "slave".

Meira

unread,
May 27, 2014, 1:36:46 PM5/27/14
to django-d...@googlegroups.com
I think a lot of recent changes in the language are harmful. Many common, short, clear, and concise words and phrases are being replaces with long, vague, sterile versions. Not only in the IT field, but everywhere.
Black people refer to themselves "a person of color", and then even shorten it: PoC. Isn't that a ridiculous acronym? If I were black, I would be deadly offended if someone dared to call me "a person of color"! What's the next step, "a person with skin color saturation level significantly higher than of the majority of people living in european and asian part of the world, as well as north and middle part of the north america, excluding those who live in the south closer to Mexico and in Italy and in Caucasia and/or spend a lot of time in the sun"?

Is it really so bad to state your skin color or your (unfortunate) social status by using the word created specifically for that?

It's bad when you subjectively judge people's abilities or personality. "Ugly", "stupid", "illiterate" - that's what you shouldn't call someone.
"Slave" is not a judging word. "Slave" describes someone's objective social status. Helping that person and millions of others in the same condition is a good thing, but banning the word is not going to help.

Jacob Kaplan-Moss

unread,
May 27, 2014, 1:46:44 PM5/27/14
to django-developers, Meira
Meira, your position has been made abundantly clear, and now your behavior is treading dangerously close to the line. I'll remind you and others of our community's code of conduct (https://www.djangoproject.com/conduct/), which specifically requires that we be welcoming, friendly, patient, and respectful. Meira, you're not doing a great job on any of these, specifically the respect.

Again you've made your point, and thank you for it. We all know where you stand. Now it's time for you to withdraw from this thread. 

I don't want to ban you from the list, but I will if I need to.

Jacob


Daniele Procida

unread,
May 27, 2014, 1:54:49 PM5/27/14
to Django Developers
On Tue, May 27, 2014, Meira <poo...@gmail.com> wrote:

>Black people refer to themselves "a person of color", and then even shorten
>it: PoC. Isn't that a ridiculous acronym? If I were black, I would be
>deadly offended if someone dared to call me "a person of color"!

You're not black though, and if some black people refer to themselves as "persons of colour", the very least you can do is not call that ridiculous.

That is not within the bounds of the courtesy you're expected to maintain while in this community.

Daniele

bartek

unread,
May 27, 2014, 3:58:02 PM5/27/14
to django-d...@googlegroups.com
On Tue, 27 May 2014 05:26:52 -0700 (PDT)
Kai <schn...@gmail.com> wrote:

> What is so bad about removing terms like master/slave that are
> related to or even originate from so much suffering and injustice and
> replacing it with neutral terms? Primary/Replica is used in many DB
> systems too.

Bad thing is that the whole logic of removing or changing every single
word that in theory might sound unpleasant to somebody is fundamentally
flawed, for a number of reasons:

First, to be consistent you should first find at least one person who
really feels personally offended by the term - but, from what I see,
nobody even tried; the assumption that in theory there might be somebody
somewhere is enough.

Second, there is no end to it - for nearly every word you can find
someone who dislikes it for one reason or another, whereas you have to
name things somehow or we would be unable to communicate.

And third, by going that way you are ignoring a quite large group of
people who care about their language, and their history too, and really
feed bad seing both of them tweaked and mangled. Those people deserve
some respect, too. Personally, I feel very strongly about it, and fully
understand and support Meira et al.

B.

>
> On Tuesday, May 27, 2014 2:14:43 PM UTC+2, Meira wrote:
> >
> > As some of you may have notice, a hot discussion is happening in
> > the comments of this pull request:
> > https://github.com/django/django/pull/2692 Essentially, this pull
> > request suggests that all occurences of master/slave be replaced
> > with leader/follower. While this is clearly insane, a less
> > jaw-dropping, but still weird change was made in commit
> > https://github.com/django/django/commit/beec05686ccc3bee8461f9a5a02c607a02352ae1
> >
> > Many users in comments to the original pull request agreed that
> > primary/replica is not a good word choice, is vague and misleading.
> > Current django docs compensate for the confusion by referring to
> > "master/slave" in parentheses after mentioning "primary/replica".
> > Of course, this change is nothing more than cosmetical, but it
> > still carries more downsides than upsides.
> > Master/slave is* immediately obvious* for the experienced users,
> > and *easily googleable* for the newbies.
> >
> > I reverted the change and sent a pull request
> > https://github.com/django/django/pull/2720. In the corresponding
> > ticket, I was told to "wait 6 months" and then resubmit the ticket.
> > ( https://code.djangoproject.com/ticket/22707), and the pull
> > request was closed immediately with an advice to start a discussion
> > on mailing list. So that's what I'm doing here :)
> >
> > I sum up my personal point of view in this comment:
> > https://github.com/django/django/pull/2692#issuecomment-44265422
> >
> > Of course, it'll be hard for the django maintainers to admit their
> > mistake and revert the change. It's always hard to admit mistakes,
> > but it's better than leaving it how it is.
> >
>



--
I don't want to belong to any club that accepts people like me as
members. (Groucho Marx)

Michael Manfre

unread,
May 27, 2014, 4:21:19 PM5/27/14
to django-d...@googlegroups.com
It was very clearly stated in the other email thread about this, by the no longer offensively titled BDFL :P, that the rename will not be reverted. It's nearly impossible to get a change in to core when there is a single core dev opposed to it and there have been many core devs who are -1 on reverting. 

The battle has been lost. The only potential outcome from continuing to press your argument is to anger core devs and flood inboxes of many people who would never have seen those doc pages if it were not for these controversial commits.

Regards,
Michael


--
You received this message because you are subscribed to the Google Groups "Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.

Hannu Krosing

unread,
May 27, 2014, 4:33:26 PM5/27/14
to django-d...@googlegroups.com
On 05/27/2014 10:20 PM, Michael Manfre wrote:
It was very clearly stated in the other email thread about this, by the no longer offensively titled BDFL :P, that the rename will not be reverted. It's nearly impossible to get a change in to core when there is a single core dev opposed to it and there have been many core devs who are -1 on reverting. 

The battle has been lost. The only potential outcome from continuing
to press your argument is to anger core devs and flood inboxes of many
people who would never have seen those doc pages if it were not for
these controversial commits.
This battle is lost indeed, but voicing your objection may be remembered
when somebody tries to steamroll in the next "politically correct" change.

What if somebody thinks that his mom may be offended by the use of the
word "hash" and provides a patch for wholesale rename of it to something
less offensive?

As you say it takes just one a core dev who agrees and it's a done thing again ;)

Cheers
Hannu


Justin Holmes

unread,
Jun 4, 2014, 12:39:08 PM6/4/14
to django-developers
Although Meira's comments are disagreeable to me, and in at least one case clearly factually incorrect, she has not come close to violating the code of conduct.  Nor has she been particularly disrespectful.  To even talk abut banning her is absurd, particularly in a thread whose subject is developing inclusive language.



For more options, visit https://groups.google.com/d/optout.



--
Justin Holmes
Chief Chocobo Breeder, slashRoot

slashRoot: Coffee House and Tech Dojo
New Paltz, NY 12561
845.633.8330

Justin Holmes

unread,
Jun 4, 2014, 12:56:40 PM6/4/14
to django-developers
OK, I guess I'll weigh in on the substance.

The thing about "master" and "slave" is not (or at least not only) that they refer to a sociopolitical configuration that is objectionable (for example, the institutionally racist forms of slavery that have occurred in many parts of the world throughout history).  It's that the interpersonal dynamic of "master" and "slave" is itself abhorrent, absent its implications or reminders.

On one hand, I am OK with regarding a machine (say, a database server) as an utter servant.  I want a future in which machines are not granted the leeways that humans have in the names of liberty and justice.  It's not difficult to imagine a number of dystopic scenarios built atop a world in which machine sovereignty is a trojan horse.

On the other hand, the words "master" and "slave," as they refer to distributed database systems, are not actual a reference to the roles of machines, but instead are a metaphor to describe our belief about how this abstraction is to be regarded.  In this sense, "master" and "slave" are not particularly accurate.  The other suggestions on these two threads all communicate a more expressive metaphor for the way we want distributed databases to work.

Finally (albeit perhaps tangentially), it is telling that, almost without exception, detractors from this change regard slavery as either having ended, being now illegal, or being something other than a really big deal.  Slavery, even in the United States, is not illegal.   The 13th amendment to the US constitution specifically includes an exception to the ban on slavery, allowing it "as a punishment for crime whereof the party shall have been duly convicted."  There are currently more people - and more people of color - living under slavery of this type now than at any point prior to the civil war.  Of these, an abhorrently large number are subject to this condition as "punishment for crime" which many if not all of us can agree is not justly regarded as a crime at all.

If, in some small way, this shift in language can signal that we regard not only the historical implications of the word "slave" but in fact the very relationship structure described by "slave" and "master," we've done a good thing.

Russell Keith-Magee

unread,
Jun 4, 2014, 7:39:47 PM6/4/14
to Django Developers
Hi Justin,

This email thread was not Meira's only contribution to the debate. In my opinion, if you take the rest of her contributions into account, and the general direction the debate was taking at the time, a reminder to her and others about the existence and consequences of the code of conduct was entirely justified.

Yours,
Russ Magee %-)



Robert Grant

unread,
Aug 12, 2014, 11:17:11 AM8/12/14
to django-d...@googlegroups.com
I'd really, really like it if we were to stop saying a UI element is "disabled" and say "differently abled".

Thanks



On Tuesday, 27 May 2014 14:14:43 UTC+2, Meira wrote:
As some of you may have notice, a hot discussion is happening in the comments of this pull request: https://github.com/django/django/pull/2692
Essentially, this pull request suggests that all occurences of master/slave be replaced with leader/follower. While this is clearly insane, a less jaw-dropping, but still weird change was made in commit https://github.com/django/django/commit/beec05686ccc3bee8461f9a5a02c607a02352ae1

Many users in comments to the original pull request agreed that primary/replica is not a good word choice, is vague and misleading. Current django docs compensate for the confusion by referring to "master/slave" in parentheses after mentioning "primary/replica". Of course, this change is nothing more than cosmetical, but it still carries more downsides than upsides.
Master/slave is immediately obvious for the experienced users, and easily googleable for the newbies.

I reverted the change and sent a pull request https://github.com/django/django/pull/2720. In the corresponding ticket, I was told to "wait 6 months" and then resubmit the ticket. (https://code.djangoproject.com/ticket/22707), and the pull request was closed immediately with an advice to start a discussion on mailing list. So that's what I'm doing here :)


I sum up my personal point of view in this comment: https://github.com/django/django/pull/2692#issuecomment-44265422

Of course, it'll be hard for the django maintainers to admit their mistake and revert the change. It's always hard to admit mistakes, but it's better than leaving it how it is.

Stan

unread,
Aug 14, 2014, 1:58:36 AM8/14/14
to django-d...@googlegroups.com
Ok Jerry Lewis,
Everybody got the idea, now nove on.

Andre Terra

unread,
Aug 14, 2014, 12:56:01 PM8/14/14
to django-d...@googlegroups.com

That is one great suggestion. +1 and as long as nobody -1s it, we're good to go!

--
You received this message because you are subscribed to the Google Groups "Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.

Tim Graham

unread,
Aug 14, 2014, 2:18:50 PM8/14/14
to django-d...@googlegroups.com
This topic is closed and no replies will be tolerated. There are plenty of Trac tickets that could use attention. Thank-you!

Cal Leeming [iops.io]

unread,
Aug 14, 2014, 2:26:44 PM8/14/14
to django-d...@googlegroups.com
Ah yes, such a brilliant way to engage legitimate concerns from community. *slow caps*

I'm out, it's been a fun 5 years guys, take care.

Cal


Juan Pablo Martínez

unread,
Aug 14, 2014, 2:27:35 PM8/14/14
to django-d...@googlegroups.com
Master/Slave +0
The discusion of this terminology is only for trollers and non technical professionals.




For more options, visit https://groups.google.com/d/optout.



--
juanpex
Reply all
Reply to author
Forward
0 new messages