hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Raghu Angadi <rang...@yahoo-inc.com>
Subject Re: [jira] Issue Comment Edited: (HADOOP-4108) FileSystem support for POSIX access method
Date Wed, 17 Sep 2008 01:15:13 GMT

+1 for not allowing edits. though I myself have been guilty many many 
times. Most of my edits tend to be fixes for typos. I should really be 
more careful before submitting a comment.

If disallowing edits is not feasible, we can check if Jira makes adding 
a reason for the edit mandatory.


Doug Cutting wrote:
> Chris Douglas wrote:
>> +1 to the overall sentiment (broadened to deleting attachments from 
>> JIRA), but I'm -1 to preventing editing of comments unless it has a 
>> grace period of at least 15 minutes.
> Can you retract email for 15 minutes?  Each comment generates an email, 
> and many people follow comments through email.  If you're going to 
> generate another email to everyone, why not make it intelligible, rather 
> than just a slightly modified regurgitation of your previous email?
> If you make a mistake in email, what do you do?  If it's serious, you 
> send another consise email correcting the mistake.  Editing a jira 
> comment does not achieve this.  Adding a new comment does.
>> Despite the preview (which many haven't found), unexpected formatting 
>> quirks and unintended assertions/offense are common; such a policy 
>> would harm clarity and generate noise. 
> But using edit to fix these generates even more noise!  If folks are 
> forced to live with their mistakes, then perhaps they'll find that 
> preview buttion and be more careful next time.
>> Even for attachments, if someone accidentally discloses private data, 
>> they shouldn't need to leave it exposed while they beg a project admin 
>> to remove it.
> If someone accidentally attaches private data in an email to a publicly 
> archived list then they've permanently disclosed their data.  Hadoop 
> emails are archived so many places that it is effectively impossible to 
> retract one.
>> Community standards don't always need automated controls.
> I've complained about this so many times over the years, and the 
> community control doesn't seem to work.  Moreover, I don't think there's 
> *ever* a reason to edit a comment: it generates noise, the information 
> is already archived all over and cannot be retracted.  If folks are not 
> comfortable with this then they should not send email to Hadoop lists 
> either.
> Doug

View raw message