Contributors mailing list archives


Re: Should we use the new GitHub reactions?

Camptocamp SA, Yannick Payot
- 17/03/2016 18:40:52
Thanks Holger and Pedro for the tests and feedbacks.

Indeed as it is more a social feature we better not use it.
We would have some difficulties to count between +1 for I like this
comment and +1 for PR is ok to be merged.
Yannick Vaucher
Business Solutions Software Developer

Camptocamp SA
PSE A, CH-1015 Lausanne
Phone: +41 21 619 10 30
Office: +41 21 619 10 10

On 17 March 2016 at 18:08, Leonardo Pistone <> wrote:
> Thanks for your analysis Pedro!
> Not a good fit for our reviews, at least now. For us an approval is
> (maybe) the most important message, that needs a clear author,
> timestamp, and optional message. The use case there seems more like a
> "like" or a "star" which is not the voting system that we need.
> I read the post
> It is indeed intended reduce noise on popular projects when the make a
> release and a million randos say YAY! +1 which is useless.
> Maybe it can be useful for us outside voting: maybe Pedro says
> something and I like his thinking, I can use the reaction to just say
> that without polluting the thread. But I think a vote needs an author,
> timestamp and everything.
> On Thu, Mar 17, 2016 at 5:23 PM,  <> wrote:
>> I have been playing with the feature and this is my analysis:
>> Pros:
>> It can summarize the reviews in one place
>> Contras:
>> The visibility of the persons that click on the reaction is worst. You
>> have
>> to click on the icon to see the involved persons. I have found some PR
>> authors clicking in the reaction. This can be confused easily with one
>> review. If some author makes the same with a comment, I delete that
>> comment
>> as project administrator, and let clean the PR. I can't do the same with
>> reactions.
>> People can add reactions in every comment, so it can be confusing to see
>> +1
>> (or -1) across all the PR comments. Is a different reviewer that got
>> confused on where to click? Is the same?
>> PSC members or core contributors doesn't receive an email with the +1.
>> Although most of the contributors doesn't use this, for me it's vital,
>> because I check when I receive a +1 if there are enough approvals to
>> merge.
>> As pointed by Holger, you cannot search the PRs with a number of
>> reactions.
>> This problem also happens with the other method, but it's mitigated with
>> the
>> previous point.
>> Reactions don't have timestamp, so you can't follow the time progression
>> of
>> the approvals.
>> So for now my judge is to not using them for reviewing purposes. It can
>> have
>> the same social component as in Facebook, but nothing more.
>> Regards.
>> 2016-03-17 16:39 GMT+01:00 Holger Brunn <>:
>>> > What Holger meant by github reaction is this new feature:
>>> >
>>> >
>>> >
>>> > To answer Holger's question I think we can use them and commiter can
>>> > take them into account. It could help to find quickly which PR is
>>> > ready.
>>> > We could give it a try.
>>> What I mean is that currently, I look at
>>> to see what I'm supposed to review and choose whatever looks interesting
>>> or important.
>>> According to, there's
>>> no way to slip in also filtering for reaction count. If this is
>>> possible, I'd be all for switching to reactions, then we could have one
>>> bookmark 'can be merged' (same as above, but reaction count >= 3) and
>>> 'needs more reviews' (same as above, but reaction count
>>> _______________________________________________
>>> Mailing-List:
>>> Post to:
>>> Unsubscribe:
>> _______________________________________________
>> Mailing-List:
>> Post to:
>> Unsubscribe:
> _______________________________________________
> Mailing-List:
> Post to:
> Unsubscribe: