incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From TJ Frazier <>
Subject Re: <id>
Date Mon, 22 Aug 2011 18:56:00 GMT
I volunteer to help with the long-range problems (karma and fu assumed). 

On 8/22/2011 14:38, Dennis E. Hamilton wrote:
> Mathias, thank you for pointing out this significant case.
> Concerning<id> in issue-tracking forms:
> If the bugtracker is bugzilla, there is both the CC information and the submitter (as
well as anyone who is listed as having some role in handling the issue).  I know that where
I am submitter (on on bugzilla, the link that shows my user name as
submitter has a in it, the e-mail address that I am registered
with.  If I expand the CC list on a bug that has a number of folks following it, I see the
specific e-mails, not user IDs.  So even when user name/ID is shown on the form, it is a link
to a hard-coded e-mail address, not some indirect link to a profile.
> The only JIRA I use is not as current or customized as the one at Apache, and it is limited
to "committers" so I don't know how "watchers" are handled (since changes to issues are posted
to a common dev list anyhow).
> But since we are starting with bugzilla and will likely migrate to a public bugzilla,
it would seem that those<id> mailto values will be all over the place,
along with many e-mail addresses that are on neither nor  Just
like for public wikis.
> I'd say there are two conclusions to be reached around this:
>   1. bugzilla is the safe podling choice for preservation of the current issue tracking,
open issues, and issue history on  (Since having it side-by-side with some
alternative has no traction, I think keeping bugzilla on the podling is an inevitable least-that-can-possibly-work
choice.)  We should get going on that migration.
>    ** We will probably require people to reregister to submit bugs on the podling version
simply because of terms of use and the Apache contribution rules.
>    ** This transposition will need to be a migration because I suspect the form will
change to some degree, especially with regard to attachments being contributions or not.
>   2. The only way to avoid closing the<id> e-mail forwarding and
then find ourselves caught by an unintended consequence, such as losing ties in the issue
tracker, is to simply preserve the forwarding for<id> in the limited
way already proposed and deal with issues with simple, limited efforts as they come up and
there are folks with cycles to address them.
> We need this functioning for all of the positive reasons already raised.  We need this
functioning to get it off the critical path for initial migration staging and to have in place
the least that can possibly work without damaging anything.  The front-loading solutions speculated
as alternatives are both costly and potentially damaging.
>   - Dennis
> -----Original Message-----
> From: Mathias Bauer []
> Sent: Monday, August 22, 2011 10:36
> To:
> Subject: Re:<id>
> On 22.08.2011 19:22, Rob Weir wrote:
>> On Mon, Aug 22, 2011 at 1:00 PM, Mathias Bauer<>  wrote:
>>> On 21.08.2011 19:32, Andy Brown wrote:
>>>> Watching the discussions here I have a question.
>>>> How hard would it be to find out which forwarding addresses are in
>>>> active use, last six months, last year?
>>>> Seems to me if an address is "active' then it should be maintained for
>>>> that user as that is where some contacts expect to find that person.  I
>>>> have one of those addresses and in the last year I may have received two
>>>> or three messages, all spam, so I do not need the address or really want
>>>> it.  As has been indicated it was given when I initially registered
>>>> OO.o.  I think I have received two or three actual messages with that
>>>> address and that was over two years ago when I signed up for the
>>>> Marketing Project.  I see no need to maintain address forwarding for
>>>> addresses that are not used but do feel that those that use them should
>>>> have them maintained for the benefit of the community.  It seems that
>>>> most of us agree that no new address should be assigned.
>>> The id@ooo addresses most probably are useful only in one case: getting
>>> mail notifications about changes in the bug tracking system. If we want
>>> to continue working with the existing bug database (and why shouldn't
>>> we?) we might want to be able to reach anybody who has either added
>>> something to an issue or has registered himself as an observer for this
>>> issue. In both cases the ooo userid and so the mailing address
>>> "" has been used to establish the contact. Throwing
>>> away these mailing addresses would mean that we could no longer reach
>>> the people that contributed to the issues or registered themselves to it.
>> But we also know that there are hundreds of notices in OOo Bugzilla
>> that are spam.  Has there been any effort to remove the accounts of
>> spammers?
>> I see a lot of spam still there.  Search Google: viagra
>> Is there any way we can preserve the project's ability to contact an
>> issue submitter, without at the same time allowing illegitimate uses
>> of the trademark, uses which are detrimental to the
>> project?
>> For example, could we create a dump of the addresses
>> and their corresponding real addresses, store that in the PPMC's
>> private directory, only for use in contacting reporters of bugs?  This
>> would be analogous to how Apache treats its MailAlias.txt file.
> IMHO it's enough to prevent the creation of new "@ooo" addresses and
> disable all accounts that spammed the bug tracker (like we did in the
> past).
> If it is easier to do the mail forwarding by mapping mail addresses
> explicitly, why not?
> Two possible problems: people must allow us to use their "real" mail
> address and have it listed in a file and the bug tracker must be able to
> use that list before it sends out notifications. Maybe the latter is not
> a problem at all, I don't know how the bug tracker works.
> Regards,
> Mathias

View raw message