The WildFly team is moving from Slack to Zulip, because Zulip seems to be
the only solution that is free, provides unlimited history, and allows
unlimited users even in private rooms (for OSS projects, at least). Gitter
has all that, except unlimited users, as we are limited to 25 people per
private room.
You can join them here:
Back to our solution... We are now 71 days away from the decommissioning of
HipChat. *Is everyone happy with Gitter?* Do you see a strong reason to
keep looking for another solution?
For my part, I noticed problems with the web client, in particular with
notifications, which are sub-standard, but with the desktop client
everything seems to work fine. It's simple, but it does the job.
Yoann Rodière
Hibernate NoORM Team
yoann(a)hibernate.org
On Wed, 21 Nov 2018 at 14:40, Yoann Rodiere <yoann(a)hibernate.org> wrote:
On top of not being able to add more than 25 people to a private
room,
there's another limitation of Gitter that Fabio just noticed: the chat
history for 1-to-1 conversations is very limited. In our case, we can only
see 2 days back, and there's no concept of archives like there is in rooms.
Meanwhile, the WildFly team is giving up on Slack because of the very
limited size of history in free plans. They are investigating Zulip,
RocketChat and MatterMost in particular. Maybe let's see what they end up
choosing and why?
Yoann Rodière
Hibernate NoORM Team
yoann(a)hibernate.org
On Tue, 13 Nov 2018 at 11:33, Yoann Rodiere <yoann(a)hibernate.org> wrote:
> On Tue, 13 Nov 2018 at 08:49, Yoann Rodiere <yoann(a)hibernate.org> wrote:
>
>> > Assuming the new chat platform takes off, there's a risk it might be
>> too successful as well
>>
>> Ok. Well, I guess we'll see. As I mentioned above, I don't think forcing
>> people to have a GitHub account will be very effective, but I can't suggest
>> a perfect solution either. Bots answering with a few links (documentation,
>> etc.) to the first message of each user come to mind, but that could be
>> considered rude, so I wouldn't do that unless the traffic becomes
>> unmanageable. Other solutions include kicking out "spammers" (but that
>> doesn't work if it's many users asking a single question), or making the
>> -dev rooms invite-only and only checking the user rooms once in a while
>> (might work if Gitter sends emails when your are mentioned while offline).
>> So, yeah, in short: I don't really know.
>>
>> > More just accountability. But if some form of login in needed to use
>> Gitter, that's enough for me. Sounded like the other option was "allow
>> anonymous", which I wanted to avoid.
>>
>> Then it should be fine: anonymous access apparently only allows to read
>> messages. Login through GitLab, GitHub or Twitter is necessary in order to
>> start posting new messages.
>>
>> Yoann Rodière
>> Hibernate NoORM Team
>> yoann(a)hibernate.org
>>
>>
>> On Mon, 12 Nov 2018 at 19:34, Steve Ebersole <steve(a)hibernate.org>
>> wrote:
>>
>>> For me its not so much about "the right kind of people". More
just
>>> accountability. But if some form of login in needed to use Gitter,
that's
>>> enough for me. Sounded like the other option was "allow
anonymous", which
>>> I wanted to avoid.
>>>
>>> On Mon, Nov 12, 2018 at 11:41 AM Sanne Grinovero <sanne(a)hibernate.org>
>>> wrote:
>>>
>>>> On Mon, 12 Nov 2018 at 17:27, Yoann Rodiere <yoann(a)hibernate.org>
>>>> wrote:
>>>> >
>>>> > I don't see why we should force people to have a GitHub
account,
>>>> considering there are other means of logging into Gitter.
>>>>
>>>> Ok.
>>>>
>>>> >
>>>> > As to getting the right type of people, I'm not sure it's
relevant.
>>>> Most people are likely to have one, and those who don't are likely to
not
>>>> have one for political reasons (think free software extremists) rather
than
>>>> because they aren't tech savvy enough: while the
"hibernate" naming might
>>>> confuse users looking for information about grizzly bears, I doubt my
>>>> grandmother, my 7-year-old nephew or even my non-software-engineer of a
>>>> wife would end up on Gitter by mistake.
>>>>
>>>> Well since that's obvious, clearly I was referring to a different
way
>>>> of cathegorizing people joining@ not by age or expertise in technology
>>>> but in having reasonable expectations and willing to do some research
>>>> before bothering us all.
>>>>
>>>> You probably weren't around yet, but Hibernate has had hard times in
>>>> which it was "victim of its own success": just too many
>>>> kinda-interested people making a ton of basic questions that could be
>>>> easily solved otherwise.
>>>>
>>>> Some "barriers" we have in place have made it manageable; of
course I
>>>> can't tell if it's all merit of the barriers of entry or just
people
>>>> coming in lower volumes with better intentions, but I'm confident
that
>>>> some of the barriers we have have helped to keep some sanity (e.g.
>>>> login on #hibernate-dev on IRC requiring an account).
>>>>
>>>> Assuming the new chat platform takes off, there's a risk it might be
>>>> too successful as well. But I guess we'll see, or let's use a
very
>>>> bad chat platform so to keep people from coming :P
>>>>
>>>> >
>>>> >
>>>> > Yoann Rodière
>>>> > Hibernate NoORM Team
>>>> > yoann(a)hibernate.org
>>>> >
>>>> >
>>>> > On Mon, 12 Nov 2018 at 18:02, Sanne Grinovero
<sanne(a)hibernate.org>
>>>> wrote:
>>>> >>
>>>> >> On Mon, 12 Nov 2018 at 16:02, Steve Ebersole
<steve(a)hibernate.org>
>>>> wrote:
>>>> >> >
>>>> >> > What is it a conscious decision to not require a GitHub
account
>>>> to join these rooms? I just noticed that is a toggle-option in the
room's
>>>> settings also.
>>>> >>
>>>> >> I don't remember. We created these rooms as an experiment in
2014..
>>>> >> Yoann created some more rooms recently.
>>>> >>
>>>> >> Should we enforce people to have a Github account? I'd like
that, I
>>>> >> think it would better nudge towards getting the right type of
people
>>>> >> to join.
>>>> >>
>>>> >> Thanks,
>>>> >> Sanne
>>>> >>
>>>> >>
>>>> >> >
>>>> >> > On Mon, Nov 12, 2018 at 6:17 AM Guillaume Smet <
>>>> guillaume.smet(a)gmail.com> wrote:
>>>> >> >>
>>>> >> >> On Mon, Nov 12, 2018 at 11:35 AM Sanne Grinovero <
>>>> sanne(a)hibernate.org>
>>>> >> >> wrote:
>>>> >> >>
>>>> >> >> > If one wants a lot of features then clearly only
Slack is the
>>>> way to
>>>> >> >> > go. Not saying we should go with Slack, just that
we'll need
>>>> to be
>>>> >> >> > patient and we'll always be short of some
features - if that's
>>>> not
>>>> >> >> > acceptable then only Slack will make you happy.
>>>> >> >> >
>>>> >> >>
>>>> >> >> TBH, I don't care about fancy features. Gitter is
OK for me but
>>>> yeah not
>>>> >> >> having sound is really annoying.
>>>> >> >>
>>>> >> >> I might miss notifications from time to time.
>>>> >> >>
>>>> >> >> In any case, it will mostly be a problem for you all if
you ping
>>>> me :).
>>>> >> >>
>>>> >> >>
>>>> >> >> > BTW the issue you linked to suggests the native
clients don't
>>>> have
>>>> >> >> > this specific problem.. might want to try that?
>>>> >> >> >
>>>> >> >>
>>>> >> >> I prefer to have it in the browser where I do most of
my
>>>> interactions with
>>>> >> >> people.
>>>> >> >>
>>>> >> >> And AFAIK, Yoann wrote they were only packaged as deb
(and not
>>>> very excited
>>>> >> >> about compiling it).
>>>> >> >>
>>>> >> >> BTW, tbh, I'm a bit worried GitLab has only one dev
on it if
>>>> they want to
>>>> >> >> become a player in this area. They certainly have some
work to
>>>> do to catch
>>>> >> >> up with others.
>>>> >> >>
>>>> >> >> --
>>>> >> >> Guillaume
>>>> >> >> _______________________________________________
>>>> >> >> hibernate-dev mailing list
>>>> >> >> hibernate-dev(a)lists.jboss.org
>>>> >> >>
https://lists.jboss.org/mailman/listinfo/hibernate-dev
>>>> >> _______________________________________________
>>>> >> hibernate-dev mailing list
>>>> >> hibernate-dev(a)lists.jboss.org
>>>> >>
https://lists.jboss.org/mailman/listinfo/hibernate-dev
>>>>
>>>