[xwiki-devs] [Proposal] JIRA strategy for issues we don't plan to fix
Hi devs, We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV: * We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point WDYT? Thanks -Vincent
On 01/26/2010 02:26 PM, Vincent Massol wrote:
Hi devs,
We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV:
* We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point
WDYT?
+1 -- Sergiu Dumitriu http://purl.org/net/sergiu/
+1 Thanks, Marius Vincent Massol wrote:
Hi devs,
We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV:
* We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point
WDYT?
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
+1 On 1/26/2010 3:26 PM, Vincent Massol wrote:
Hi devs,
We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV:
* We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point
WDYT?
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
On Tue, Jan 26, 2010 at 14:26, Vincent Massol <[email protected]> wrote:
Hi devs,
We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV:
* We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point
+1
WDYT?
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
-- Thomas Mortagne
On Tue, Jan 26, 2010 at 2:26 PM, Vincent Massol <[email protected]> wrote:
Hi devs,
We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV:
* We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point
WDYT?
+1, JV.
+1 - Asiri On Tue, Jan 26, 2010 at 6:56 PM, Vincent Massol <[email protected]> wrote:
Hi devs,
We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV:
* We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point
WDYT?
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
Since we all agreed, I've updated http://dev.xwiki.org/xwiki/bin/view/Community/DevelopmentPractices with this new rule. I've also rewritten other JIRA rules listed on the page to make them more easy to find/read. http://dev.xwiki.org/xwiki/bin/view/Community/DevelopmentPractices#HJIRABest... Thanks -Vincent On Jan 26, 2010, at 2:26 PM, Vincent Massol wrote:
Hi devs,
We've had a quick discussion with Thomas on IRC regarding the strategy we want to adopt for issues we don't plan to fix. Here's my POV:
* We should close issues that we know we won't fix. We might not want to fix them for several reasons but one of these reason is that the issue is for an old XE version. Especially since we said that we're only supporting the last 2 versions. We = xwiki committers, but if any committer wants to fix an old issue he's free to do it of course, it's just a general rule. * An option would be to leave them open and wait for a patch but I don't think it's a good idea, because: - it sends the wrong signals that we're going to fix the issue - the issue can still be found with a search even if closed - when we close it then the reporter can explain why it's so important for him, or not. Which we wouldn't know if we hadn't closed it - contributors could reopen them and attach a patch or just create a new issue - it means that for all other issues we plan to fix them at some point
WDYT?
Thanks -Vincent
Hi, Here's an excerpt from the set of rules about not creating issues for unreleased work: "Imagine that someone has been working on a given issue for a given version. Then someone else notices a problem related to the work done. If the version has not been released then you shouldn't create a new JIRA issue for reporting the problem. Instead you should reopen the issue (if it's been closed) and add a comment to it." There should be a mention about what to do when you encounter a problem for which there is no Jira issue. For example the way I report this is by creating a page on the incubator in the "Test" space. There I present the issue, attach printscreens and mention the version where I encountered the problem. If the problem is not fixed in the release I report it on Jira. Is there a "standard" procedure for this type of situation? ----- Silvia Rusu Tester & Documentation Writer - XWiki http://twitter.com/silviarusu -- View this message in context: http://n2.nabble.com/Proposal-JIRA-strategy-for-issues-we-don-t-plan-to-fix-... Sent from the XWiki- Dev mailing list archive at Nabble.com.
On Feb 2, 2010, at 2:12 PM, Silvia Rusu wrote:
Hi,
Here's an excerpt from the set of rules about not creating issues for unreleased work:
"Imagine that someone has been working on a given issue for a given version. Then someone else notices a problem related to the work done. If the version has not been released then you shouldn't create a new JIRA issue for reporting the problem. Instead you should reopen the issue (if it's been closed) and add a comment to it."
There should be a mention about what to do when you encounter a problem for which there is no Jira issue.
You mean when you don't know if there's already a jira issue or not? Yes sometimes it's hard to know (especially if you're not a developer and thus not following the devs that happen). In this case you can just create an issue and the person responsible for it will close it as duplicate for ex and thus set the "fix for" to "unknown". Is that what you mean? Thanks -Vincent
For example the way I report this is by creating a page on the incubator in the "Test" space. There I present the issue, attach printscreens and mention the version where I encountered the problem. If the problem is not fixed in the release I report it on Jira.
Is there a "standard" procedure for this type of situation?
Hi, I mean when I know there is no Jira issue at all, as the problem didn't arise in any of the previous releases. The problem only arises on a version that hasn't been released. In this case: - There is no Jira issue to reopen or comment on - I cannot open a new bug on Jira (the rules say if the version has not been released then I shouldn't create a new JIRA issue for reporting the problem) vmassol wrote:
On Feb 2, 2010, at 2:12 PM, Silvia Rusu wrote:
Hi,
Here's an excerpt from the set of rules about not creating issues for unreleased work:
"Imagine that someone has been working on a given issue for a given version. Then someone else notices a problem related to the work done. If the version has not been released then you shouldn't create a new JIRA issue for reporting the problem. Instead you should reopen the issue (if it's been closed) and add a comment to it."
There should be a mention about what to do when you encounter a problem for which there is no Jira issue.
You mean when you don't know if there's already a jira issue or not?
Yes sometimes it's hard to know (especially if you're not a developer and thus not following the devs that happen). In this case you can just create an issue and the person responsible for it will close it as duplicate for ex and thus set the "fix for" to "unknown".
Is that what you mean?
Thanks -Vincent
For example the way I report this is by creating a page on the incubator in the "Test" space. There I present the issue, attach printscreens and mention the version where I encountered the problem. If the problem is not fixed in the release I report it on Jira.
Is there a "standard" procedure for this type of situation?
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
----- Silvia Rusu Tester & Documentation Writer - XWiki http://twitter.com/silviarusu -- View this message in context: http://n2.nabble.com/Proposal-JIRA-strategy-for-issues-we-don-t-plan-to-fix-... Sent from the XWiki- Dev mailing list archive at Nabble.com.
On Feb 2, 2010, at 3:22 PM, Silvia Rusu wrote:
Hi,
I mean when I know there is no Jira issue at all, as the problem didn't arise in any of the previous releases. The problem only arises on a version that hasn't been released.
In this case: - There is no Jira issue to reopen or comment on - I cannot open a new bug on Jira (the rules say if the version has not been released then I shouldn't create a new JIRA issue for reporting the problem)
This cannot happen.... :) It would mean someone has made changes without creating a jira issue which is forbidden... Since we're following all commits the occurrence should be pretty rare. Thanks -Vincent
vmassol wrote:
On Feb 2, 2010, at 2:12 PM, Silvia Rusu wrote:
Hi,
Here's an excerpt from the set of rules about not creating issues for unreleased work:
"Imagine that someone has been working on a given issue for a given version. Then someone else notices a problem related to the work done. If the version has not been released then you shouldn't create a new JIRA issue for reporting the problem. Instead you should reopen the issue (if it's been closed) and add a comment to it."
There should be a mention about what to do when you encounter a problem for which there is no Jira issue.
You mean when you don't know if there's already a jira issue or not?
Yes sometimes it's hard to know (especially if you're not a developer and thus not following the devs that happen). In this case you can just create an issue and the person responsible for it will close it as duplicate for ex and thus set the "fix for" to "unknown".
Is that what you mean?
Thanks -Vincent
For example the way I report this is by creating a page on the incubator in the "Test" space. There I present the issue, attach printscreens and mention the version where I encountered the problem. If the problem is not fixed in the release I report it on Jira.
Is there a "standard" procedure for this type of situation?
Thanks Vincent, Sergiu & Marius for your answers. All clear now. ----- Silvia Rusu Tester & Documentation Writer - XWiki http://twitter.com/silviarusu -- View this message in context: http://n2.nabble.com/Proposal-JIRA-strategy-for-issues-we-don-t-plan-to-fix-... Sent from the XWiki- Dev mailing list archive at Nabble.com.
On Feb 2, 2010, at 4:14 PM, Silvia Rusu wrote:
Thanks Vincent, Sergiu & Marius for your answers. All clear now.
Np Silvia, please feel free to complement the jira rules on the wiki page to make it more clear. Thanks -Vincent
On 02/02/2010 03:22 PM, Silvia Rusu wrote:
Hi,
I mean when I know there is no Jira issue at all, as the problem didn't arise in any of the previous releases. The problem only arises on a version that hasn't been released.
In this case: - There is no Jira issue to reopen or comment on - I cannot open a new bug on Jira (the rules say if the version has not been released then I shouldn't create a new JIRA issue for reporting the problem)
Are you talking about stuff that is being developed, for example work in progress on the incubator? If yes, then the best thing is to just comment on the page(s) of the feature. Talking with the developer on IRC is good, too, for things that are small enough to just be mentioned in person.
vmassol wrote:
On Feb 2, 2010, at 2:12 PM, Silvia Rusu wrote:
Hi,
Here's an excerpt from the set of rules about not creating issues for unreleased work:
"Imagine that someone has been working on a given issue for a given version. Then someone else notices a problem related to the work done. If the version has not been released then you shouldn't create a new JIRA issue for reporting the problem. Instead you should reopen the issue (if it's been closed) and add a comment to it."
There should be a mention about what to do when you encounter a problem for which there is no Jira issue.
You mean when you don't know if there's already a jira issue or not?
Yes sometimes it's hard to know (especially if you're not a developer and thus not following the devs that happen). In this case you can just create an issue and the person responsible for it will close it as duplicate for ex and thus set the "fix for" to "unknown".
Is that what you mean?
Thanks -Vincent
For example the way I report this is by creating a page on the incubator in the "Test" space. There I present the issue, attach printscreens and mention the version where I encountered the problem. If the problem is not fixed in the release I report it on Jira.
Is there a "standard" procedure for this type of situation?
-- Sergiu Dumitriu http://purl.org/net/sergiu/
Hi Silvia, Silvia Rusu wrote:
Hi,
Here's an excerpt from the set of rules about not creating issues for unreleased work:
"Imagine that someone has been working on a given issue for a given version. Then someone else notices a problem related to the work done. If the version has not been released then you shouldn't create a new JIRA issue for reporting the problem. Instead you should reopen the issue (if it's been closed) and add a comment to it."
There should be a mention about what to do when you encounter a problem for which there is no Jira issue.
If you find a problem that is present in a snapshot but not in the latest release then that problem was introduced by a change made after the release. When we developers make a change we _usually_ refer to a JIRA issue in the commit log. This means you can face this scenarios: (1) A developer commits without opening a JIRA issue and his change produces a bug which you notice. In this case, IMO the developer either * reverts the commit * or opens a JIRA issue related to his/her change where you can comment the found bug (2) The developer refers to a JIRA issue in the commit log. In this case you find the JIRA issue either by searching or asking the developer. You can find the developer that works on a specific part of XWiki platform (the part where you discover the bug) by asking on the IRC/Jabber. There is also a special case where the developer is working on a new module which hasn't been release yet and it tells you to test it (i.e. there's no previous release for that module). In this case, IMO, the best is to ask the developer to release a first version, even if it's featureless, and to open JIRA issue against that that initial version for the features he/she's working on. This way you can comment on those issues or even report issues for the initial release (if it's the case). In case you don't find neither the JIRA issue nor the developer responsible for the problem (can happen) I think you should open a JIRA issue for the previous release. IMO it's better to close or correct a JIRA issue than to forget or ignore a problem. Thanks, Marius
For example the way I report this is by creating a page on the incubator in the "Test" space. There I present the issue, attach printscreens and mention the version where I encountered the problem. If the problem is not fixed in the release I report it on Jira.
Is there a "standard" procedure for this type of situation?
----- Silvia Rusu Tester & Documentation Writer - XWiki http://twitter.com/silviarusu
participants (8)
-
Asiri Rathnayake -
Florin Ciubotaru -
Jean-Vincent Drean -
Marius Dumitru Florea -
Sergiu Dumitriu -
Silvia Rusu -
Thomas Mortagne -
Vincent Massol