SUMO community discussions

Help Desk, Issues & Policy: Bugs & Hard questions including Memory Use

  1. Introduction & background

    First of all welcome again to Patrick and the newer Rachel on the new helpdesk.

    • We now have paid staff on the team helping with AAQ
    • And a proper Escalation System; although that is in its infancy.
    • Another talk: unfortunately many contributors do not like these sorts of discussions,
      Same Same, I think we often have reoccurring discussions, reach similar discussions each time but fail to move forward with positive action.
    • Waste of time. Why talk it wastes our time when they we could be writing articles or answering questions.
      But Hey there have been many changes and improvements to Sumo over recent years. Let's as contributors get in on the discussions, as and before they are implemented. Otherwise it is crying over spilt milk.

    I am trying to get the ball rolling on this. I am posting my initial thoughts without fully thinking through and detailing them all. After all it is not my thoughts that count. It is the discussion and the agreed outcome. I can easily later on add more links to prior community discussion threads on these subjects and example threads in Sumo AAQ as that may helps some especially anyone joining recently.

    Bugs
    In some cases it will help to raise bugs from a question or a subject or crash report. That certainly is a topic for discussion. However we must avoid encouraging or filing bugs that we feel or know will almost certainly get closed as incomplete or invalid etc. Where I see the helpdesk being useful here is maybe they should monitor bugs raise directly from or relating to sumo questions. If the bugs lack information try to follow up from sumo or by bug comments to help the poster provide missing information. And/Or as we should currently do ask the bug poster to enquire (again?) on sumo AAQ instead.
    Do the helpdesk staff and others agree this is something needing to be done ?

    Memory problems or Slowness & CPU use etc.
    I am sure there are perennial press reports and blog discussions and articles on Firefox being slow or a memory hog etc. Whether true or not we often get such problems reported or ranted about on AAQ. We are not very successful in dealing with this, It does not help Firefox's image, It does not help Sumo.

    Other Old Chestnuts (in no particular order)

    1. Sync: IIRC development has stopped. It will be deprecated. The replacement will be a proper server backup.
    2. Panorama / Tab Groups : IIRC development has stopped. It will be deprecated. No idea about any replacement.
    3. Sumo Search : IIRC suggestions are to dumb it down. Integrate the advanced search with the standard one. Despite recent major changes because it was considered broken, IMHO it has the utility value of a chocolate fire-guard, but we need it to access parts not indexed with google. We will miss finding things because of the lack of boolean and string searches, probably a title search option would also help. Everyone wants a serch to do miracles. Contributors (& maybe helpdesk staff) may want it to narrow down aggressively to a particular thing, forget the '"'problem" of no hits on a boolean or string search. That proves a negative and may be important: i.e. a particular problem, software version, crash signature etc has not been reported.
      • Crash Stats hardcoded SUMO SEARCH link Try it. I need say no more than pointless waste of time.

    Forum Guidelines
    Some of this discussion may change practices, certainly the use of escalation and the often mentioned second tier. In fact it is one reason Madalina delayed discussion of escalation in the other thread.

    Lets use etherpads, to summarise this thread do you agree ?
    I sometimes tend to split threads into sub threads, but with our new Etherpads section and the knowledge that Madalina favoured using etherpads in the guidelines discussion maybe that is how this thread could be handled i.e. permit it to get unreadably long but summarise topics in an index Etherpad with subpads per topic.

    Policy: Escalating Hard Questions including the Long threads

    1. Escalation & Tags
      Currently I think escalation is merely picking off the low hanging fruit. Sure the last few percentage of questions get an answer. I do not understand about the zendesk tickets.
      (Someone could comment on that what are they ? )
      so may be escalation needs to use tags to generate these tickets. Otherwise the change was; from a forum point of view; redundant. Anyone and everyone could easily and effectively filter by unanswered questions (and for that matter the rather pointless nearly deprecated tags)
    2. Escalate the hard to solve questions
      I think the helpdesk and probably everyone agrees the target now should be solving more questions, not just answering them. That needs discussion.
      • Rants & Multi Questions.
        I tend to think, it is not worth closing the long threads and rants, they just startup again.
        What do others think ? Should we have Moderators and forum guidelines on this ?
    3. Current guidelines say single question per thread.
      • I know Kadir in a recent bug confirmed that is still intended policy
      • I remember in a Whats up with sumo video not long ago Michael agreed and reported back from an external event to say the one question per thread is the best use of sumo AAQ.
      • I think contributors should be strongly discouraged from feeding such threads by providing answers.
        They should encourage users to start a new thread for an answer. As long as responses and answers are posted the questions will will keep accumulating.
        The dilemma, is we sometimes need and encourage additional posts even from none contributors. Think breaking issues and suchlike where we may not have STR or know what systems or other software etc is involved.
    === Introduction & background=== First of all welcome again to Patrick and the newer Rachel on the new helpdesk. * We now have '''paid staff''' on the team helping with AAQ *And a proper '''Escalation System'''; although that is in its infancy. * Another talk: unfortunately many contributors do not like these sorts of discussions,<br /> Same Same, I think we often have reoccurring discussions, reach similar discussions each time but fail to move forward with positive action. * Waste of time. Why talk it wastes our time when they we could be writing articles or answering questions. <br /> But '''Hey''' there have been many changes and improvements to Sumo over recent years. Let's as contributors get in on the discussions, as and before they are implemented. Otherwise it is crying over spilt milk. I am trying to get the ball rolling on this. I am posting my initial thoughts without fully thinking through and detailing them all. After all it is not my thoughts that count. It is the discussion and the agreed outcome. I can easily later on add more links to prior community discussion threads on these subjects and example threads in Sumo AAQ as that may helps some especially anyone joining recently. <u>Bugs</u><br /> In some cases it will help to raise bugs from a question or a subject or crash report. That certainly is a topic for discussion. However we must avoid encouraging or filing bugs that we feel or know will almost certainly get closed as incomplete or invalid etc. Where I see the helpdesk being useful here is maybe they should monitor bugs raise directly from or relating to sumo questions. If the bugs lack information try to follow up from sumo or by bug comments to help the poster provide missing information. And/Or as we should currently do ask the bug poster to enquire (again?) on sumo AAQ instead.<br />'''Do the helpdesk staff and others agree this is something needing to be done ?''' <u>Memory problems or Slowness & CPU use etc.</u><br/> I am sure there are perennial press reports and blog discussions and articles on Firefox being slow or a memory hog etc. Whether true or not we often get such problems reported or ranted about on AAQ. We are not very successful in dealing with this, It does not help Firefox's image, It does not help Sumo. <u>Other Old Chestnuts</u> (in no particular order) # Sync: IIRC development has stopped. It will be deprecated. The replacement will be a proper server backup. # Panorama / Tab Groups : IIRC development has stopped. It will be deprecated. No idea about any replacement. # Sumo Search : IIRC suggestions are to dumb it down. Integrate the advanced search with the standard one. Despite recent major changes because it was considered broken, IMHO it has the utility value of a chocolate fire-guard, but we need it to access parts not indexed with google. We will miss finding things because of the lack of boolean and string searches, probably a title search option would also help. Everyone wants a serch to do miracles. Contributors (& maybe helpdesk staff) may want it to narrow down aggressively to a particular thing, forget the '"'problem''" of no hits on a boolean or string search. That proves a negative and may be important: i.e. a particular problem, software version, crash signature etc has not been reported. #* '''Crash Stats hardcoded SUMO SEARCH link''' Try it. I need say no more than ''pointless waste of time''. <u>Forum Guidelines</u><br /> Some of this discussion may change practices, certainly the use of escalation and the often mentioned second tier. In fact it is one reason Madalina delayed discussion of escalation in the other thread. <u>Lets use etherpads, to summarise this thread do you agree ?</u> <br /> I sometimes tend to split threads into sub threads, but with our new Etherpads section and the knowledge that Madalina favoured using etherpads in the guidelines discussion maybe that is how this thread could be handled i.e. permit it to get unreadably long but summarise topics in an index Etherpad with subpads per topic. <u>Policy: Escalating Hard Questions including the Long threads</u> # Escalation & Tags <br />Currently I think escalation is merely picking off the low hanging fruit. Sure the last few percentage of questions get an answer. I do not understand about the zendesk tickets.<br /> (Someone could comment on that what are they ? ) <br />so may be escalation needs to use tags to generate these tickets. Otherwise the change was; from a forum point of view; redundant. Anyone and everyone could easily and effectively filter by unanswered questions (and for that matter the rather pointless nearly deprecated tags) # Escalate the hard to solve questions <br />I think the helpdesk and probably everyone agrees the target now should be solving more questions, not just answering them. That needs discussion. #* Rants & Multi Questions. <br />I tend to think, it is not worth closing the long threads and rants, they just startup again. <br />What do others think ? Should we have Moderators and forum guidelines on this ? #Current guidelines say single question per thread. #*I know Kadir in a recent bug confirmed that is still intended policy #*I remember in a ''Whats up with sumo'' video not long ago Michael agreed and reported back from an external event to say the one question per thread is the best use of sumo AAQ. #*'''I think contributors should be strongly discouraged from feeding such threads by providing answers.''' <br /> They should encourage users to start a new thread for an answer. As long as responses and answers are posted the questions will will keep accumulating. <br />The dilemma, is we sometimes '''need''' and encourage additional posts even from none contributors. Think '''breaking issues''' and suchlike where we may not have STR or know what systems or other software etc is involved.
  2. more options

    Wow! This might be a record for the longest initial post covering the most topics! I'll just comment on the last point you raised:

    John99 said

    Current guidelines say single question per thread.
    • I know Kadir in a recent bug confirmed that is still intended policy
    • I remember in a Whats up with sumo video not long ago Michael agreed and reported back from an external event to say the one question per thread is the best use of sumo AAQ.
    • I think contributors should be strongly discouraged from feeding such threads by providing answers.
      They should encourage users to start a new thread for an answer. As long as responses and answers are posted the questions will will keep accumulating.

    I agree .... the rule should be enforced. This often happens in threads about common issues. Here's an example thread where I asked one user after another to start a new reply, and linked to the Mozilla Support rules and guidelines in each of my posts:
    /questions/977750 firefox 25.0.1 keeps randomly crashing, it never did this before the update.

    Here's the reply I posted to Hijacker #3:


    <Hijacker#3> and anyone else having crash issues:

    If you need more help please ask a new question and include troubleshooting info with your crash Report IDs, since this thread is for helping <OP>. You can use this link as a starting point: https://support.mozilla.org/questions/new

    See Mozilla Support rules and guidelines: ... For support requests, do not re-use existing threads started by others, even if they are seemingly on the same subject.


    I've been closing (locking) threads like this if it's a solved question but dealing with multiple users adding their own support requests to the same thread is a problem.

    Wow! This might be a record for the longest initial post covering the most topics! I'll just comment on the last point you raised: ''John99 [[#post-55689|said]]'' <blockquote> Current guidelines say single question per thread. *I know Kadir in a recent bug confirmed that is still intended policy *I remember in a ''Whats up with sumo'' video not long ago Michael agreed and reported back from an external event to say the one question per thread is the best use of sumo AAQ. *'''I think contributors should be strongly discouraged from feeding such threads by providing answers.''' <br /> They should encourage users to start a new thread for an answer. As long as responses and answers are posted the questions will will keep accumulating. </blockquote> I agree .... the rule should be enforced. This often happens in threads about common issues. Here's an example thread where I asked one user after another to start a new reply, and linked to the [[Forum rules and guidelines]] in each of my posts: <br>[/questions/977750] ''firefox 25.0.1 keeps randomly crashing, it never did this before the update.'' Here's the reply I posted to Hijacker #3: ------ <Hijacker#3> and anyone else having crash issues: If you need more help please ask a new question and include troubleshooting info with your crash Report IDs, since this thread is for helping <OP>. You can use this link as a starting point: https://support.mozilla.org/questions/new See [[Forum rules and guidelines]]: ... ''For support requests, do not re-use existing threads started by others, even if they are seemingly on the same subject.'' ----- I've been closing (locking) threads like this if it's a solved question but dealing with multiple users adding their own support requests to the same thread is a problem.
  3. Alice Wyman said

    Wow! This might be a record for the longest initial post covering the most topics! I'll just comment on the last point you raised: ....

    Yes I know its long ;-)

    I kinda like the idea of covering a lot of related topics at once rather than taking one but ignoring the others, also isn't this the sort of thing that is currently supposed to be getting priority: upping solved rates, and is something with changes being actively introduced right now.

    • New helpdesk / 2nd tier
    • Work In Progress: Change in thread status model

    I thought as we have a new helpdesk it makes sense to discuss how they are helping on the Forum. With an emphasis on what I see as current problems that will benefit from this 2nd tier approach. I suspect the helpdesk talents will be very much wasted if all they do is pick up the 5-10% questions that get no answers within 24 hours. The objective must be to increase the solved rate, lets help the 60-70% with unsolved threads. (That should improve dramatically anyhow just because of the proposed new thread status system)

    I dare say some or even most of the posts being picked up by the automated escalation tag fall into categories of

    • incomplete
    • off topic
    • mobile / Firefox OS <-- do they just take longer to be picked up initially ?

    Orchestrated Discussion

    • What do others think, maybe Madalina should get involved or lead this,or someone from the helpdesk ?
    • Is a set of etherpads like the other thread a good idea ?
    • Alice commented on the single question per thread should we do that as the first topic, and if so is it best actually discussed in the guidelines thread or in this one ? It almost certainly needs a policy decision confirmed at admin level, but that is effectively what revised guideline would be.

    Single Question Topic
    I have noticed you and some other contributors doing that already. It should be easy to get consensus agreement that we do not overturn the current guideline.

    I think to be effective we need to go even further and encourage, (by means of a reworded guideline) contributors not to answer most of the supplemental questions other than to say start your own thread etc.

    It is trivial to cross link threads, if that helps. Single question threads are easier to read, understand, find and solve (or close etc.).

    Escalate tag presumably automated escalation has kicked in now

    Rather hard to tell though to be honest as there are so few questions likely to be picked up each day.

    ''Alice Wyman [[#post-55690|said]]'' <blockquote> Wow! This might be a record for the longest initial post covering the most topics! I'll just comment on the last point you raised: .... </blockquote> ===Yes I know its long ;-)=== I kinda like the idea of covering a lot of related topics at once rather than taking one but ignoring the others, also isn't this the sort of thing that is currently supposed to be getting priority: upping solved rates, and is something with changes being actively introduced right now. * New helpdesk / 2nd tier * Work In Progress: Change in thread status model I thought as we have a new helpdesk it makes sense to discuss how they are helping on the Forum. With an emphasis on what I see as current problems that will benefit from this 2nd tier approach. I suspect the helpdesk talents will be very much wasted if all they do is pick up the 5-10% questions that get no answers within 24 hours. The objective must be to increase the solved rate, lets help the 60-70% with unsolved threads. (That should improve dramatically anyhow just because of the proposed new thread status system) I dare say some or even most of the posts being picked up by the automated escalation tag fall into categories of * incomplete * off topic * mobile / Firefox OS <-- do they just take longer to be picked up initially ? <u>Orchestrated Discussion</u><br /> *What do others think, maybe Madalina should get involved or lead this,or someone from the helpdesk ? * Is a set of etherpads like the other thread a good idea ? * Alice commented on the single question per thread should we do that as the first topic, and if so is it best actually discussed in the guidelines thread or in this one ? It almost certainly needs a policy decision confirmed at admin level, but that is effectively what revised guideline would be. <u>Single Question Topic</u><br /> I have noticed you and some other contributors doing that already. It should be easy to get consensus agreement that we do not overturn the current guideline. I think to be effective we need to go even further and encourage, (by means of a reworded guideline) contributors not to answer most of the supplemental questions other than to say start your own thread etc. It is trivial to cross link threads, if that helps. Single question threads are easier to read, understand, find and '''solve''' (or close etc.). ===Escalate tag presumably automated escalation has kicked in now=== Rather hard to tell though to be honest as there are so few questions likely to be picked up each day.
  4. Is there any progress on that notification system I discussed sometime ago?????

    There needs to be a way for users to get easy access to answers people have provided them, instead of sticking their head in the Inbox and maybe just not getting help at all.

    Example: Google's Google + Notifications.

    This is related to the solved/response rate.

    Is there any progress on that notification system I discussed sometime ago????? There needs to be a way for users to get easy access to answers people have provided them, instead of sticking their head in the Inbox and maybe just not getting help at all. Example: Google's Google + Notifications. <img src="http://1.bp.blogspot.com/-Abh0WR7LvbI/UcFdQhRPl0I/AAAAAAABKSY/tg3k_F7hyxI/s640/google-plus-new-notifications-2.png"/> This is related to the solved/response rate.

    Modified by kobe on

  5. Maybe dig up your old thread. Presumably you have noticed we have email and RSS subscriptions for those asking or replying to questions in AAQ.

    Other fora have options to use email subscriptions from the profile settings /users/settings

    Maybe dig up your old thread. Presumably you have noticed we have email and RSS subscriptions for those asking or replying to questions in AAQ. Other fora have options to use email subscriptions from the profile settings [/users/settings]
  6. Hi John, lot's of topics here. Thanks for moving the discussion forward, one way to make progress would be to get a common understanding of the priority of these issues. Once we have an agreement on priority we could focus on the highest priorities, come up with potential solutions and drive them to completion.

    Hi John, lot's of topics here. Thanks for moving the discussion forward, one way to make progress would be to get a common understanding of the priority of these issues. Once we have an agreement on priority we could focus on the highest priorities, come up with potential solutions and drive them to completion.
  7. Thanks for the comment Kadir.

    Hopefully others will also comment, but what would your suggestions be for priorities ?

    The first stage of course is discussion and some of that can be done in parallel with the use of etherpads. Probably very little of my proposals involve any development or other staff.

    I am also thinking about

    • What has already or will shortly be implemented and how to make best use of that.
    • Keeping the discussions in a logical sequence for instance helping with bugs may help with Memory etc issues.

    In my opinion the top priorities & suggested order would be as follows.

    1. Single question per thread
      But deal with it in the existing discussion on forum guidelines. Relatively quick and none controversial,other than the fact that I am pushing for a recommendation to not feed the threads by having contributors answer supplemental questions unnecessarily within the thread
    2. Orchestrated Discussion (First in this thread)
      Seems like you have started that off by asking for prioritises. And I think take advantage of etherpads for summaries and recording decisions.
    3. Advanced Search
      I think we need to improve this. It will then need development time. (or not if the idea is rejected) I know others have either commented or shown examples where an improved search would be an advantage. Should be easy enough to make a Story for this.
    4. Escalation policies maybe also discuss tags and topics
      Update See now Rachel's Thread
      (if not fully solved in the guidelines discussion thread)
    5. Bugs - subjects raised on AAQ
      (File or assist if necessary)
      Update See now Rachel's post downthread
      Could just be a discussion on policy. I am expecting it is going to need to involve Help Desk members, and presumably they are going to need time to settle down, understand the workload, their own priorities & possibly identify themselves what needs change and priority over here.
    6. Memory, CPU, slowness, CRASHES and other hard questions
      These are Just vague first suggestions for discussion here. The thread status will help as will escalation. We all have to improve and learn from one another and maybe get help from paid staff. Let's document some of that so the knowledge is passed on or available for reference.
    • Help users determine a solution, maybe file bugs
    • Educate us contributors better,
    • Cooperate with developers and bug triagers
    • Filebugs where appropriate
    • KB advanced article about, ( & or linking to MDN documentation) consoles, the profiler, about:memory
    • FHR, & telemetry results can we learn from or use those ?
    Thanks for the comment Kadir. Hopefully others will also comment, but what would your suggestions be for priorities ? The first stage of course is discussion and some of that can be done in parallel with the use of etherpads. Probably very little of my proposals involve any development or other staff. I am also thinking about * What has already or will shortly be implemented and how to make best use of that. *Keeping the discussions in a logical sequence for instance helping with bugs may help with Memory etc issues. In my opinion the top priorities & suggested order would be as follows. # '''Single question per thread''' <br> But deal with it in the existing discussion on forum guidelines. Relatively quick and none controversial,other than the fact that I am pushing for a recommendation to not feed the threads by having contributors answer supplemental questions unnecessarily within the thread #'''Orchestrated Discussion '''(First in this thread)<br> Seems like you have started that off by asking for prioritises. And I think take advantage of etherpads for summaries and recording decisions. # '''Advanced Search '''<br />I think we need to improve this. It will then need development time. (or not if the idea is rejected) I know others have either commented or shown examples where an improved search would be an advantage. Should be easy enough to make a ''Story'' for this. #'''Escalation policies ''' maybe also discuss tags and topics<br />'''Update''' See now Rachel's [https://support.mozilla.org/en-US/forums/contributors/709929 Thread]<br />(if not fully solved in the guidelines discussion thread) #'''Bugs - subjects raised on AAQ''' <br />(File or assist if necessary)<br /> '''Update''' See now Rachel's post [https://support.mozilla.org/en-US/forums/contributors/709812?last=56814#post-56810 downthread]<br />Could just be a discussion on policy. I am expecting it is going to need to involve Help Desk members, and presumably they are going to need time to settle down, understand the workload, their own priorities & possibly identify themselves what needs change and priority over here. # '''Memory, CPU, slowness, CRASHES and other hard questions''' <br />These are Just vague first suggestions for discussion here. The thread status will help as will escalation. We all have to improve and learn from one another and maybe get help from paid staff. Let's document some of that so the knowledge is passed on or available for reference. *Help users determine a solution, maybe file bugs *Educate us contributors better, *Cooperate with developers and bug triagers * Filebugs where appropriate * KB advanced article about, ( & or linking to MDN documentation) consoles, the profiler, about:memory * FHR, & telemetry results can we learn from or use those ?

    Modified by John99 on

  8. Hey John99,

    Thanks for the thoughtful post. There are a couple of goals that SUMO is trying to accomplish this quarter. One is accomplished with the escalate tag in getting every posters question replied to within 24 hours, as well as changes to the forum to increase the solved rate which is being tracked in this bug: https://bugzilla.mozilla.org/show_bug.cgi?id=920945.

    Zendesk is the tool that Rachel and I use to interact with Firefox OS partners. We thought it would also serve as a great way to get notification when something needs our attention in the forums which come in the form of escalated posts.

    I also appreciate your analysis of why questions go unanswered by contributors. I'm noticing the same pattern of cases escalated, but don't think we should avoid answering them because it's a Sync, memory or incomplete post. Our goal in SUMO should be to make sure everyone gets a reply to at least move them forward to a solution.

    I'm looking forward to more discussion and feedback, so please keep it coming!

    Cheers, Patrick

    Hey John99, Thanks for the thoughtful post. There are a couple of goals that SUMO is trying to accomplish this quarter. One is accomplished with the escalate tag in getting every posters question replied to within 24 hours, as well as changes to the forum to increase the solved rate which is being tracked in this bug: https://bugzilla.mozilla.org/show_bug.cgi?id=920945. Zendesk is the tool that Rachel and I use to interact with Firefox OS partners. We thought it would also serve as a great way to get notification when something needs our attention in the forums which come in the form of escalated posts. I also appreciate your analysis of why questions go unanswered by contributors. I'm noticing the same pattern of cases escalated, but don't think we should avoid answering them because it's a Sync, memory or incomplete post. Our goal in SUMO should be to make sure everyone gets a reply to at least move them forward to a solution. I'm looking forward to more discussion and feedback, so please keep it coming! Cheers, Patrick

    Modified by patrickmc on

  9. Hey Patrick,
    thanks for the comments.

    I did not know what zendesk was and had not noticed the tracker bug. We often have discussions on here that get some sort of consensus but do not get actioned, only for them to repeat months or years later.

    Every now and again things jump forward, as is happening now. Rather than try to reopen old discussions I thought it better to start a long thread and hope we get everyone together with ideas. Make the most of the changes we have been asking for, for so long.


    Replies Target (imho that should be two tier also !)
    The 100% replies gets achieved occasionally now. It is currently often 90-5% something undreamt of a couple of years ago. With usually less than 100/day that, at first glance only means answering about another 6-8 per day to hit the target.

    The last comment of course is an over simplification. In fact IMHO a better target would be in two bands say 95% within 18 Hours and 5% within 36 hours. Then you probably would only need to answer half a dozen posts a day. That leaves more time for the real work on increasing the solved rate.

    The problem with going for 100% in any time period is that the 2nd tier ie you need to pick the correct posts to reply to; and or reply within a very short time frame. In practice you will look at posts escalated after 12 hours and answer some that would have otherwise been answered anyway within 24hours by the 1st tier.


     don't think we should avoid answering them because it's a Sync, memory or incomplete post.
    

    Don't get me wrong I am not saying we should avoid answering any question, even if only to

    • say it is off topic and close it, including a useful link if possible, as per the current Guidelines
    • or to ask for more information.

    Upping the solved rate
    As for the solved rate. Well depending on how that is defined it will leap up just because we get rid of the solved| not binary classification status for something more sane and workable.

    What we really need to escalate
    The hard nut to crack is the stalled questions persons (user and contributor) still there. That is what gets missed and needs escalating.

    I see no need to auto-tag and escalate most of what is being done now. It is redundant. The forum already filters unanswered questions and sorts by date posted.

    At present I would say it is the ability to manually tag for escalation that is potentially most useful.

    • IMHO it needs agreement,
      then a guideline
      to say (? 1st tier ?) users may, and should, escalate questions when they are unable to proceed to a solution
      • Another tag escalateM for manual escalation maybe ?

    Cheers, John

    Hey Patrick, <br />thanks for the comments. I did not know what zendesk was and had not noticed the tracker bug. We often have discussions on here that get some sort of consensus but do not get actioned, only for them to repeat months or years later. Every now and again things jump forward, as is happening now. Rather than try to reopen old discussions I thought it better to start a long thread and hope we get everyone together with ideas. Make the most of the changes we have been asking for, for so long. ------- <u>Replies Target (imho that should be two tier also !)</u><br /> The 100% replies gets achieved occasionally now. It is currently often 90-5% something undreamt of a couple of years ago. With usually less than 100/day that, at first glance only means answering about another 6-8 per day to hit the target. The last comment of course is an over simplification. In fact IMHO a better target would be in two bands say 95% within 18 Hours and 5% within 36 hours. Then you probably would only need to answer half a dozen posts a day. That leaves more time for the real work on increasing the solved rate. The problem with going for 100% in any time period is that the 2nd tier ie '''you''' need to pick the correct posts to reply to; and or reply within a very short time frame. In practice you will look at posts escalated after 12 hours and answer some that would have otherwise been answered anyway within 24hours by the 1st tier. ------ '' don't think we should avoid answering them because it's a Sync, memory or incomplete post.'' Don't get me wrong I am not saying we should avoid answering any question, even if only to *say it is off topic and close it, including a useful link if possible, as per the current Guidelines *or to ask for more information. ------- <u>Upping the solved rate</u><br /> As for the solved rate. Well depending on how that is defined it will leap up just because we get rid of the ''solved| not'' binary classification status for something more sane and workable. <u>What we really need to escalate </u><br/> The hard nut to crack is the stalled questions persons (user and contributor) still there. That is what gets missed and needs escalating. I see no need to auto-tag and escalate most of what is being done now. It is redundant. The forum already filters unanswered questions and sorts by date posted. * [/questions?filter=no-replies] At present I would say it is the ability to manually tag for escalation that is potentially most useful. * IMHO '''it needs agreement, <br />then a guideline <br />to say (? 1st tier ?) users may, and should, escalate questions when they are unable to proceed to a solution''' **Another tag ''escalateM'' for manual escalation maybe ? Cheers, John
  10. patrickmc said

    Hey John99, .....
    One is accomplished with the escalate tag in getting every posters question replied to within 24 hours, .....

    Yes well done.

    I note

    • To see the 100% we now need to remember the metrics are split and select the en-us /questions/dashboard/metrics/en-US
    • Pity about the display bug. They finally got round to dealing with the last period of data for months and weeks (truncated if incomplete) I guess they neglected to fix the 'days so the last part of daily graph tends to dive down still.
    ''patrickmc [[#post-55709|said]]'' <blockquote> Hey John99, ..... <br /> One is accomplished with the escalate tag in getting every posters question replied to within 24 hours, ..... <br /> </blockquote> Yes well done. I note * To see the 100% we now need to remember the metrics are split and select the en-us [/questions/dashboard/metrics/en-US] *Pity about the display bug. They finally got round to dealing with the last period of data for ''months'' and ''weeks'' (truncated if incomplete) I guess they neglected to fix the '''days'' so the last part of daily graph tends to dive down still.
  11. Now that the new escalation system has started I posted this elsethread

    I thought they would only test new threads from the last few days; since the date of introduction of the new escalation method!! philipp said
    apparently all old questions fitting the criteria got auto-escalated, so that we now have ten-thousands of escalated threads
    - this is quite an ambitious task for our helpdesk i have to say ;-)

    Anyhow even if they just run as intended and test newer threads I think it is not the best system. After an initial surge when it tests all threads since introduction, I think it would continue to escalate too many, and not necessarily the threads most benefiting of an escalation.

    Now that the new escalation system has started I posted this [/forums/contributors/709759?page=2#post-55862 elsethread] <blockquote> I thought they would only test new threads from the last few days; since the date of introduction of the new escalation method!! ''philipp [[#post-55861|said]]'' <blockquote> apparently all old questions fitting the criteria got auto-escalated, so that we now have ten-thousands of escalated threads <br />- this is quite an ambitious task for our helpdesk i have to say ;-) </blockquote> Anyhow even if they just run as intended and test newer threads I think it is not the best system. After an initial surge when it tests all threads since introduction, I think it would continue to escalate too many, and not necessarily the threads most benefiting of an escalation. </blockquote>
  12. Waka_Flocka_Flame said

    Is there any progress on that notification system I discussed sometime ago????? There needs to be a way for users to get easy access to answers people have provided them, instead of sticking their head in the Inbox and maybe just not getting help at all. Example: Google's Google + Notifications. This is related to the solved/response rate.

    CROSS LINK also see now

    'Kadir Topal said

    Recently I realized that the email messages in the support forum could be improved and Bram came up with an amazing mock-up for them:
    One thing that we are currently not doing is labeling the original poster in some way. What do you think, would that be better, or just unnecessary information?
    ''Waka_Flocka_Flame [[#post-55693|said]]'' <blockquote> Is there any progress on that notification system I discussed sometime ago????? There needs to be a way for users to get easy access to answers people have provided them, instead of sticking their head in the Inbox and maybe just not getting help at all. Example: Google's Google + Notifications. <img src="http://1.bp.blogspot.com/-Abh0WR7LvbI/UcFdQhRPl0I/AAAAAAABKSY/tg3k_F7hyxI/s640/google-plus-new-notifications-2.png"/> This is related to the solved/response rate. </blockquote> CROSS LINK also see now 'Kadir Topal [https://support.mozilla.org/en-US/forums/contributors/709901?last=56686#post-56584 said]'' <blockquote> Recently I realized that the email messages in the support forum could be improved and Bram came up with an amazing mock-up for them: <img src="https://bugzilla.mozilla.org/attachment.cgi?id=8362329"> <br /> One thing that we are currently not doing is labeling the original poster in some way. What do you think, would that be better, or just unnecessary information? </blockquote>
  13. John99, your comment about the issues 4, 5, 6

    4 is in progress in these discussions: https://support.mozilla.org/en-US/for.../709929?last=56808 and this one I am working on some guidelines also by asking questions about how to file bugs etc and I would love to get a better idea of the audience. Currently there are guidelines in MDN and I would love to address this in 5.

    My concerns for bugs are this though:

    • audience
    • when you know you have enough information to file a bug, and knowing what component to put it in for each product, where can we make this easy for a new bug filer?
    • What can make is easy? and fun for contributors? a game? points for quality bugs? badges?

    For 6, are you asking how to gather troubleshooting info- are there any engineers or volunteers that work on bugs that want to start collaborating on this. The guidelines I have only found are the MDN how to write a good bug, and how to write a good first bug.

    We could even incorporate this and would love help from anyone who knows of any other needs.

    John99, your comment about the issues 4, 5, 6 4 is in progress in these discussions: [https://support.mozilla.org/en-US/forums/contributors/709929?last=56808] and this one I am working on some guidelines also by asking questions about how to file bugs etc and I would love to get a better idea of the audience. Currently there are guidelines in MDN and I would love to address this in 5. My concerns for bugs are this though: *audience *when you know you have enough information to file a bug, and knowing what component to put it in for each product, where can we make this easy for a new bug filer? *What can make is easy? and fun for contributors? a game? points for quality bugs? badges? For 6, are you asking how to gather troubleshooting info- are there any engineers or volunteers that work on bugs that want to start collaborating on this. The guidelines I have only found are the MDN how to write a good bug, and how to write a good first bug. We could even incorporate this and would love help from anyone who knows of any other needs.
  14. Hi Rachel

    5.) Bugs.

    All this is only personal opinion.
    Its a discussion better ideas & alternative views welcome.

    I was hoping you may be able to get others onboard to help with specific bug subjects. The guidelines you have already found are I imagine these

    It is probably not vital we get everything perfect. Related bugs are a good source to crib from. Triagers and others will tweak bug titles or change components etc if it helps. The main problem I foresee is likely to be inappropriate bugs.

    • Unsupported versions. That is only OK if it is a regression.
      It would then help if we could use mozregression, but at the very least in a case that is difficult to reproduce we need the user to show the problem occurs in the Current Release but not in the he unsupported version needed as a workaround.
    • Not using SafeMode and a Clean profile with plugins disabled.
      Unless
      • It is one specific plugin causing the issue.
      • A profile / bookmarks etc problem needing a working profile to demonstrate
      • Or maybe something we are asking be blocked !
        (Compare the Babylon saga.)
    • Bugs are for Firefox problems. Not an excuse to not troubleshoot.
      If the problem is seen with addons then if it is an addon problem.
      • Bugs are not normally filed for addons. FlashPlayer is an exception.
        (The sooner we have an alternative for that the better)
        Again try all the troubleshooting steps first and not when other plugins are in use (especially Real Player related)
    • Most important is that we get some good STR.
      It would be good if is always such that someone else could easily reproduce the problem. In practice that may not be so easy.
    Hi Rachel ==5.) Bugs.== All this is only personal opinion. <br />Its a discussion better ideas & alternative views welcome. I was hoping you may be able to get others onboard to help with specific bug subjects. The guidelines you have already found are I imagine these * https://developer.mozilla.org/en-US/docs/Mozilla/QA/Bug_writing_guidelines * '''edit''' ''and https://quality.mozilla.org/docs/bugzilla/starter-kit/how-to-write-a-proper-bug/ <br /> https://quality.mozilla.org/docs/bugzilla/starter-kit/how-to-write-a-proper-bug-part-2/ '' It is probably not vital we get everything perfect. Related bugs are a good source to crib from. Triagers and others will tweak bug titles or change components etc if it helps. The main problem I foresee is likely to be inappropriate bugs. * '''Unsupported versions.''' That is only OK if it is a regression.<br/ /> It would then help if we could use mozregression, but at the very least in a case that is difficult to reproduce we need the user to show the problem occurs in the Current Release but not in the he unsupported version needed as a workaround. * '''Not using SafeMode and a Clean profile with plugins disabled.'''<br /> Unless ** It is one specific plugin causing the issue. ** A profile / bookmarks etc problem needing a working profile to demonstrate **Or maybe something we are asking be blocked ! <br />(Compare the Babylon saga.) * '''Bugs are for Firefox problems.''' Not an excuse to not troubleshoot.<br /> If the problem is seen with addons then if it is an addon problem. ** '''Bugs are not normally filed for addons.''' FlashPlayer is an exception. <br />(The sooner we have an alternative for that the better)<br /> Again try all the troubleshooting steps first and not when other plugins are in use (especially Real Player related) *'''Most important is that we get some good STR.''' <br /> It would be good if is always such that someone else could easily reproduce the problem. In practice that may not be so easy.

    Modified by John99 on

  15. 5.) Bugs

    guigs2 said

    .... My concerns for bugs are this though:
    • audience
    • when you know you have enough information to file a bug, and knowing what component to put it in for each product, where can we make this easy for a new bug filer?
    • What can make is easy? and fun for contributors? a game? points for quality bugs? badges?

    Bugs Audience

    Contributors
    There probably are not too many contributors interested in doing this, but if we have guidelines then maybe it will happen. Especially if the contributor is aware another contributor or HelpDesk is there if there is any problem.

    End Users
    We can always point a user at whatever guidelines we have. Possibly Users wishing to file bugs will fall into one of three main categories.

    1. Experienced users needing little or no help.
    2. The majority that would need help or wish us to file a bug for them.
    3. A large proportion that would just like to file a bug saying Firefox is now useless. Your problem. Please fix or I will use an alternative browser..

    New Bugs Filers

    Guidelines, guidelines, guidelines. We can put guidelines somewhere. As long as everyone - or at least most contributors - can find them then we are on our way. Then we need a bit of magic from HelpDesk. If a bug is filed from AAQ and valid then have it looked over and followed up if necessary by yourselves User Advocacy or whoever is appropriate.

    If contributors realise filing bugs gets results they will I am sure do this. In a similar manor Users will realise problems may get solved if bugs are filed, and we are able to point to examples.

    Fun

    Yes ideas to make it fun are going to help. I will leave that to you, Madalina & other Contributors. We do have badges.

    ==5.) Bugs== ''guigs2 [[#post-56810|said]]'' <blockquote> .... My concerns for bugs are this though: *audience *when you know you have enough information to file a bug, and knowing what component to put it in for each product, where can we make this easy for a new bug filer? *What can make is easy? and fun for contributors? a game? points for quality bugs? badges? </blockquote> === Bugs Audience === '''Contributors''' <br /> There probably are not too many contributors interested in doing this, but if we have guidelines then maybe it will happen. Especially if the contributor is aware another contributor or HelpDesk is there if there is any problem. '''End Users''' <br /> We can always point a user at whatever guidelines we have. Possibly Users wishing to file bugs will fall into one of three main categories. #Experienced users needing little or no help. #The majority that would need help or wish us to file a bug for them. #A large proportion that would just like to file a bug saying Firefox is now useless. Your problem. Please fix or I will use an alternative browser.. ===New Bugs Filers=== Guidelines, guidelines, guidelines. We can put guidelines somewhere. As long as everyone - or at least most contributors - can find them then we are on our way. Then we need a bit of magic from HelpDesk. If a bug is filed from AAQ and valid then have it looked over and followed up if necessary by yourselves User Advocacy or whoever is appropriate. If contributors realise filing bugs gets results they will I am sure do this. In a similar manor Users will realise problems may get solved if bugs are filed, and we are able to point to examples. ===Fun=== Yes ideas to make it fun are going to help. I will leave that to you, Madalina & other Contributors. We do have badges.