Skip navigation

LR 3.4 RC: Adobe bug fixing policy

Mar 10, 2011 2:49 PM

  Latest reply: Keith_Reeder, May 30, 2011 4:36 AM
Replies 1 2 Previous Next
  • Currently Being Moderated
    Mar 14, 2011 7:38 PM   in reply to JayS In CT

    JayS In CT wrote:

     

    That said, within this forum there have been numerous times when large numbers of individuals have been able to confirm an error, at times on one OS or another, and other times on both..  There's never been a suggestion from the Adobe folks that we should each write up the same problem, and fact is, I didn't think they wanted that.

     

    Define "a large numbers of individuals"...is it to the point where a large percentage of a platform's users are reporting the same issue?

     

    I would suggest it is indeed more useful to report those problems than to assume that the issue has been reported "enough" and not add to the overall emphasis of a particular issue by adding to the collective voice.

     

    The phrase "squeaky wheel get the grease" comes to mind.

     

    Depending on the forcefulness and the accuracy of the reported issue do you really think it's in Adobe's best interest in ignoring a large % of the user base? No, it's not.

     

    A lot depends on the frequency of the reported issue and how mission critical the issue is deemed to be. There are of course other factors involved; is the issue easy or hard to deal with? Is the issue contributing to data loss? Is the issue impacting a large % of users (this is a much bigger factor than most people realize)?

     

    However, NOT reporting an issue, for whatever reason, is short changing the entire process.

     

    Again, I will tell you that the more people who report an issue (and the more important the issue is deemed to be in part based on the numbers of users impacted) the greater the likelihood the issue will be resolved.

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 14, 2011 7:55 PM   in reply to TK2142

    TK2142 wrote:

     

    Lee Jay wrote:

     

    I would want you to think that you probably have no idea of the complexities of code you have no access to. 

     

    Lee, I've written complex software in teams myself. I teach software engineering. I know exactly what I'm talking about.

     

    Not without the code you don't.

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 14, 2011 8:37 PM   in reply to clvrmnky

    All of this I understand - I develop software (and fix bugs) for a living...

     

    I find it totally reasonable that some bugs persist across multiple dot releases - Adobe has to have priorities...

     

    What I dont get is why certain bugs, where certain means: not all, but some that are particularly glaring, and in my guesstimation - relatively easy to fix) that started in Lr1 still exist in Lr3.4RC.

     

    I'm beginning to have some deja vu here - we've been through all of this before.

     

    Its not that some bugs aren't fixed that vexes me, its the ones that keep getting left behind that seem like they really shouldn't be, that is perplexing.

     

    And neither you nor anybody else is going to convince me nor Thomas that fixing certain keywords-with-spaces problems or stacking in collections, etc., might cause irreparable damage... - source code or none...

     

    I try to give Adobe the benefit of the doubt - but they make it really hard in some cases...

     

    Maybe Adobe has a bug in their website and most bug reports never make it to them - that must be it (how would we know? - how would they know?)...

     

    R

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 14, 2011 9:23 PM   in reply to Jeff Schewe

    Jeff Schewe wrote:

     

    JayS In CT wrote:

     

    That said, within this forum there have been numerous times when large numbers of individuals have been able to confirm an error, at times on one OS or another, and other times on both..  There's never been a suggestion from the Adobe folks that we should each write up the same problem, and fact is, I didn't think they wanted that.

     

    Define "a large numbers of individuals"...is it to the point where a large percentage of a platform's users are reporting the same issue?

     

    I would suggest it is indeed more useful to report those problems than to assume that the issue has been reported "enough" and not add to the overall emphasis of a particular issue by adding to the collective voice.

     

    The phrase "squeaky wheel get the grease" comes to mind.

     

    Depending on the forcefulness and the accuracy of the reported issue do you really think it's in Adobe's best interest in ignoring a large % of the user base? No, it's not.

     

    A lot depends on the frequency of the reported issue and how mission critical the issue is deemed to be. There are of course other factors involved; is the issue easy or hard to deal with? Is the issue contributing to data loss? Is the issue impacting a large % of users (this is a much bigger factor than most people realize)?

     

    However, NOT reporting an issue, for whatever reason, is short changing the entire process.

     

    Again, I will tell you that the more people who report an issue (and the more important the issue is deemed to be in part based on the numbers of users impacted) the greater the likelihood the issue will be resolved.

     

    Jeff,

     

    The point wasn't "how" large, but more the fact that I don't ever recall a significant "everyone send in a bug report" type effort ever suggested here in the forum.  There have been numerous times when the suggestion was made to a thread originator that they should file a bug report, but I don't recall ever seeing a single thread, even those where folks from Adobe participated, where a call went out for everyone with the problem to write up a report.  I'm well aware of the "squeakly wheel" phrase, and based on what you're suggesting, should be implemented now as the status quo anytime a number (whatever that number may be) of folks hit an issue..  The call should be, "everyone with this issue, here's the address for the bug report"..  As I said, perhap naively, I assumed that when a thread about a particular issue was embedded in a problem report (which I've done) that the length of the thread may carry some weight.. From what you're saying, it's the weight of all the problem reports written..  good to know for the future.

     

    Jay S.

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 15, 2011 12:02 AM   in reply to Rob Cole

    Sadly, I have to agree with some of both view points.

     

    I once got handed a bug that took me three weeks to find.  15 minutes to fix.  So yes, the bug was "easy to fix"...  finding it was the hard part, and I was the third engineer to get handed the bug, after the others gave up in disgust.

     

    On the other hand, not having a customer viewable bug tracking means lots of people stumble over the same problem, post here for help (or not, not even knowing this exists), call customer support (nope, we've never heard of that.  Must be an hardware/OS/plugin problem...), or just quit or stay frustrated.

     

    Having a place where we could see bug reports and status (reported/confirmed/assigned/future/circular-filed) would at least cut down on some of the frustration, and we would actually know they at least have the report...

     

    on the other hand, having the number of bugs visible is a negative marketing point.  So they are not going to make them visible.  Heck, they don't even ackowledge the bug report, much less post them!  I was absolutely AMAZED when I filed a bug report for Photoshop.  Not only did I get an acknowledgement of the bug, it was by a real person!   With a reasoned response to my bug report.  So I know they _can_ do it...

     

    So Jeff, the fact that the LR team can't get it together doesn't mean that it's Adobe policy, since the PS team can do it, it must mean they just don't...  Which doesn't make any of us feel better about it, squeeky wheels/sand in the gears or not.

     

    Cheers!

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 15, 2011 4:34 AM   in reply to tgutgu

    I believe the total number of bugs fixed in the SDK since Lr3.0 is zero. (and thats not because there aren't any). Nor has there been any update to the API doc nor SDK doc to mention the things Adobe knows about that dont work.

     

    And, other than a few howdy-doodies by Chet Drarvik in the beginning (Eric Scouten's successor), the total number of responses from Adobe on that forum since Eric Scouten left is 0.

     

    And some of these bugs would unquestionably be easy to fix - no doubt about it.

     

    Definitely lots of room for improvement in this department... - a little communique at a minimum, but some SDK bug fixes would be nice too...

     

    Rob

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 24, 2011 2:25 PM   in reply to tgutgu

    Hi tqutqu,

     

    We've fixed the bug with entering keywords with spaces.  It will be in the final Lightroom 3.4 release.

     

    Regards,

    Becky

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 24, 2011 2:45 PM   in reply to +becs

    +becs wrote:

     

    Hi tqutqu,

     

    We've fixed the bug with entering keywords with spaces.  It will be in the final Lightroom 3.4 release.

     

    Regards,

    Becky

    I hope that pleases tqutqu and it also shows Adobe are paying attention to these 'user-to-user' forums.

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 24, 2011 3:43 PM   in reply to +becs

    becs+

     

    That's good news (although I no longer use keywords with spaces...)

     

    In case you haven't noticed, "we" like hearing from Adobe employees .

     

    Hope y'all are striking a good balance between new Lr4 dev and bug fixes.

     

    R

     
    |
    Mark as:
  • TK2142
    407 posts
    Jan 20, 2010
    Currently Being Moderated
    Mar 24, 2011 4:50 PM   in reply to +becs

    +becs wrote:

     

    We've fixed the bug with entering keywords with spaces.  It will be in the final Lightroom 3.4 release.

    Thanks for letting us know Becky, much appreciated!

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 25, 2011 7:23 AM   in reply to Rob Cole
    function(){return A.apply(null,[this].concat($A(arguments)))}

    areohbee wrote:

     

    I believe the total number of bugs fixed in the SDK since Lr3.0 is zero. (and thats not because there aren't any). Nor has there been any update to the API doc nor SDK doc to mention the things Adobe knows about that dont work.

     

    And, other than a few howdy-doodies by Chet Drarvik in the beginning (Eric Scouten's successor), the total number of responses from Adobe on that forum since Eric Scouten left is 0.

     

    And some of these bugs would unquestionably be easy to fix - no doubt about it.

     

    Definitely lots of room for improvement in this department... - a little communique at a minimum, but some SDK bug fixes would be nice too...

     

    Rob

     

    I gave up working in javascript with the CS1/2/3/4/5 SDK (Photoshop and Bridge) after it became obvious to me adobe do not provide timely support and fixes, so this is nothing new.  The contrast to Microsoft was sharp, where I often saw a 2 week turn-around after reporting a bug for VBA.

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 25, 2011 3:00 PM   in reply to hillrg

    hillrg wrote:

     

    The contrast to Microsoft was sharp, where I often saw a 2 week turn-around after reporting a bug for VBA.

     

    Kinda surprises me that Microsoft would be any better. I try to give Adobe the benefit of the doubt (imagining them busting their humps for a killer Lr4 - no time for bug fixes or talk...), but they do make it a challenge sometimes... For me and some, just a little bit of comm. can go a long way - a reply on the forum or an updated doc or a quick web-page with some pertinent info... I would really love to see that "Pertinent Issues" page, but I'm guessing that wont be happening soon...

     

    Tah-tah,

    Rob

     
    |
    Mark as:
  • Currently Being Moderated
    May 28, 2011 8:12 AM   in reply to tgutgu

    nm

     
    |
    Mark as:
  • Currently Being Moderated
    May 28, 2011 8:27 AM   in reply to tgutgu

    tgutgu wrote:


    I can understand that bugs, which can lead to data corruption, are fixed quickly.

     

    Then why did you ask the question?

     
    |
    Mark as:
  • Currently Being Moderated
    May 28, 2011 8:30 AM   in reply to tgutgu

    tgutgu wrote:

     

    I want to raise this topic again, because it clearly shows that soemthing is wrong with Adobe's bug fixing policy.

     

    Here is an extract from an entry of the Lightroom Killer Tips web site by Matt Kloslowski.

     

    Hey folks. Last week I mentioned a rare bug in Lightroom 3.4 that had to  do with JPEG photos. It only affected a very small amount of people  with a certain camera though. Nonetheless, if you were one of those  people then it was definitely a problem. Anyway, Adobe has quickly  released an update that addresses the issue. It’s available for download  now in Lightroom under the Help menu. Also, you can read more about it  over at Tom Hogarty’s blog (Lightroom Product Manager).

     

     

    So a bug, which apparently only affects a very small amount of people with a certain camera, gets fixed within one week, but others, which affect everybody (!) (adjustment slider bugs, keywords with spaces bugs, just to give examples) take months if not years to become addressed or are simply ignored.

     

    Defect scheduling is different for different shops, but one thing is nearly always the same. Any feature that risks data corruption or data loss through normal operation is always considered serious, even if it could only affect a small subset of the user base. This is one of the cases where an emergency point release is warranted.

     

    So, you issue a public statement regarding the fault (as Adobe did -- it showed up on Twitter and my RSS feed) so the user base can determine how at risk they are, and you get a fix out the door ASAP.

     

    This defect was one of those cases. There is nothing sinister going on here. This is the way enterprise software development works.

     
    |
    Mark as:
  • Currently Being Moderated
    May 28, 2011 8:33 AM   in reply to tgutgu

    http://blogs.adobe.com/lightroomjournal/2011/05/potential-jpeg-bug-in- lightroom-3-4-and-camera-raw-6-4.html

     

    "... the bug could cause image corruption which falls within our highest severity category of bugs."

     

    "Rare bugs like this make it quite difficult to pinpoint the source of  the problem or even create a reproducible scenario but thanks to Doug’s  persistence we were able to track down this bug quickly."

     

    So, it's a high-priority because it causes data corruption, and it was handed to them by a user that could isolate a reproducable case of a rare bug.  High priority + reproducable case = fixed quickly.

     

    Another reason they might not fix a bug quickly is if fixing it has the potential to spawn a large number of new bugs, or even a new category of bugs, especially if there's an easy workaround, it affects very few people and/or it doesn't put data at risk.  Bugs like that tend to wait for a major release.

     
    |
    Mark as:
  • Currently Being Moderated
    May 28, 2011 5:11 PM   in reply to tgutgu

    Bugs causing less damage are fixed with more quality control and come out on a normal schedule.

     
    |
    Mark as:
  • Currently Being Moderated
    May 28, 2011 5:24 PM   in reply to tgutgu

    tgutgu wrote:


    Reproducible malfunctions (known since a long time), which affect every users (not only a small group owning a particular camera) aren't fixed. Why?

     

    First of all, this doesn't affect all users.  Second, like I said, there are other considerations such as the possibility that other bugs will be created if this one is fixed.

     

    Also, they've opened a new system to help them with both bugs and feature requests from the public, and it's already paying dividends.

     
    |
    Mark as:
  • Currently Being Moderated
    May 30, 2011 4:36 AM   in reply to tgutgu

    Assuming that Adobe's bug scheduling policy doesn't exist purely to annoy you, Thomas, isn't it perfectly reasonable that relatively trivial bugs like the sliders problem (which is surely trivial to me - I never even realised there was such a thing as "scrubby sliders", much less that there was a bug in the functionality) are going to be pushed back to the bottom of the pile as new, more serious, potential showstopper bugs emerge?

     

    In my own past experience (like everyone else here it would seem, I used to work in IT, as a system designer for some of the the largest UK Government departments) even "easy" code fixes required considerable resource expenditure at the testing stage - identifying and implementing a code fix is only a small part of the delivery of a solution, and Adobe's resources aren't endless.

     
    |
    Mark as:
1 2 Previous Next
Actions

More Like This

  • Retrieving data ...

Bookmarked By (0)

Answers + Points = Status

  • 10 points awarded for Correct Answers
  • 5 points awarded for Helpful Answers
  • 10,000+ points
  • 1,001-10,000 points
  • 501-1,000 points
  • 5-500 points