I think we should separate language barriers for dev@ (a channel for all
its developers to agree on what the project is doing) and users@, which
could be much more diverse, but follow more of a Questions and Answer
format.

It's clear that in the Apache Way, the dev@ list should use a language that
all developers can follow (directly or indirectly via translators);
otherwise you will form sub-communities which may not agree on the
direction on the project.

Sometimes it might make sense to do dual-language posting, e.g. include the
message text in both Chinese and English. Yes, it is more work, but perhaps
some project need a more "UN" approach with many languages, to not be too
exclusive to a large potential dev base.

But I don't see a problem with language-specific *users* lists under ASF,
e.g rocketmq-用户列表@apache.org ("users list" ?), some projects like
OpenOffice use such lists for translation and localisation effort.

Also I think it's good for an ASF project committers to join third-party
user engagements where they happen anyway, like on StackOverflow, Gitter,
BioStars and regional equivalents, as long as they are public and don't
require payment; (but we would need to be careful about how we link to
those).

It it however important that any project development (including bug
reports) are brought back into the official Apache dev list, issue tracker
etc; where it would need to include an English summary/translation, with a
link or quote of the original discussion.

If say QQ don't allow you to catch up with what happened earlier (like on
Gitter) then perhaps some effort is needed to set up a bot with logging,
like we already have for the oldie IRC channels for Apache projects.

On 11 Nov 2016 7:43 am, "Reynold Xin" <r...@apache.org> wrote:

> Background: I have no tie to RocketMQ. I didn't even know about it until
> today and I don't know any of the people associated with the project. I am
> Chinese but living in the US. I'm purely playing devil's advocate about a
> meta-point here and don't know if it applies to RocketMQ or not.
>
> I definitely agree with Jeff's point that "my thoughts about community
> would be getting as many people and users involved as possible".
>
> That said, for a project started in China, it is unclear switching the
> primary development language from Chinese to English would help with
> accomplishing that goal. While lowering the bar for non-Chinese speakers to
> participate, it will limit the efficacy of its original developers, and
> increases the bar for more Chinese developers, which are the more natural,
> immediate expansion targets for the community.
>
> If we as a community want to enforce the usage of English as the standard,
> we should just explicitly say that.
>
> I'd avoid using the argument that English will bring more users, as it is
> not defensible and risk being interpreted as western arrogance. Afterall,
> three out of the six largest Internet companies (by market cap) are
> currently in mainland China, and they all have enormous daily active users
> even though they are targeting primarily Chinese.
>
>
> On Thu, Nov 10, 2016 at 11:14 PM, Jeff Genender <jgenen...@apache.org>
> wrote:
>
> > I would think that English is generally used because its the most
> > international language, not because its the most used in the world.  Thus
> > it helps cross borders for communication.  At the end of the day, I think
> > you need to look at your community and ask if you want it to cross
> borders
> > or not.  Do you want worldwide contribution (and adoption)?  I can tell
> you
> > that I glean a lot of information from the mail lists when I run into
> > problems or issues using Apache software.  If the discussions are in
> > Chinese, you may miss a lot of people who can be a part of the discussion
> > from outside of China.  I think you really need to think about who you
> want
> > your users to be and how you want your product adopted.
> >
> > In addition, this is an incubated project.  AFAICT, the champion doesn’t
> > speak Chinese, and I am wild-guessing maybe 2 of the mentors do.  This
> > means the other mentors may have a difficult time steering the project
> when
> > they are needed.  It makes it difficult for the champion to asses any
> > problems without having someone notify him of a translated issue.  In the
> > unlikely event that the project requires input from the incubation PMC
> or,
> > the board for that matter, it would be very difficult to get a proper
> > insight into the issues without have solid knowledge of the language.
> >
> > I personally don’t know of any rule or regulation that locks down a
> > language and perhaps a board member can chime in on that.  But my .02 is
> > that if I were bringing a project to Apache, my thoughts about community
> > would be getting as many people and users involved as possible.  If you
> > don’t use a more cross-border/international language, then I believe that
> > you may ultimately be hindering your project beyond your borders.  I
> think
> > that would be a shame.  OTOH, maybe your desire is to keep RocketMQ a
> > Chinese piece of software.  I guess that is ok too… but I would be
> > interested in why.
> >
> > Just my usual .02.
> >
> > Jeff
> >
> > > On Nov 10, 2016, at 11:53 PM, Tom Barber <t...@spicule.co.uk> wrote:
> > >
> > > I believe I saw something the other day where someone was talking about
> > diverse languages on mailing lists. personally I think it's okay but
> > obviously it decreases the chance of participation of others.
> > >
> > > of course the old saying "if it wasn't discussed on the list it never
> > happened" didn't mention the language.
> > >
> > > Thought must be taken for jira and code comments as well. how would non
> > Chinese speaking people follow development?
> > >
> > >
> > > On 11 Nov 2016 06:45, "Reynold Xin" <r...@apache.org <mailto:
> > r...@apache.org>> wrote:
> > > Adding members@
> > >
> > > On Thu, Nov 10, 2016 at 10:40 PM, Reynold Xin <r...@apache.org
> <mailto:
> > r...@apache.org>> wrote:
> > >
> > > > To play devil's advocate: is it OK for Apache projects that consist
> > > > primarily of Chinese developers to communicate in Chinese? Or put it
> > > > differently -- is it a requirement that all communications must be in
> > > > English?
> > > >
> > > > I can see an inclusiveness argument for having to use English, as
> > English
> > > > is one of the most common languages. However, many talented software
> > > > developers in China don't have the sufficient level of proficiency
> > when it
> > > > comes to English, as the penetration rate of English in China is much
> > lower
> > > > than other countries. It is as hard for Chinese speakers to learn
> > English
> > > > as for English speakers to learn Chinese.
> > > >
> > > > One can certainly argue forcing everybody to use English will also
> > exclude
> > > > those Chinese developers, and from the perspective of the number of
> > native
> > > > speakers, Mandarin (a Chinese dialect) outnumbers English 3 to 1
> > according
> > > > to Wikipedia.
> > > >
> > > > Similar argument also applies to Japanese, and many other countries,
> > > > except the number of Chinese speakers is much larger.
> > > >
> >
>

Reply via email to