Bill Meyer No idea, but I'm very glad that they fixed my issues which did not have high ratings (as in, they weren't rated), because they were show-stoppers for me.
Dalija Prasnikar I realize that. But can't help wondering at the process through which they decide which to fix. I've never worked for a company that was quite so apparently arbitrary, nor so quick to disparage customers' concerns.
I think its a matter of signal to noise ratio. Any tools vendor will tell you other developers scream louder than any other group of end users because:
A. They understand the effort involved better than most. B. Think their requests should carry more weight because of A.
But let's face it. Bug fixes don't attract new customers (though ignoring them can certainly cost you existing ones). New features attract new customers.
Bill Meyer I know that QC is perceived as black hole, and I hope new bug tracking system will see the light of day sooner than later. But the more concerning issue is that number of bugs is inherently growing, and even serious bugs don't get fixed promptly. Lack of manpower and/or organizational skills in action.
Regardless, I just cannot let go, and if there is any chance that some pushing from my side can make the difference I will do it.
Dalija Prasnikar The number can only increase, when so many new features are being introduced in each new release. My concern now is that, just as with the features issue, defects in the new areas will command the majority of attention, and the desktop issues will remain unresolved.
Dalija Prasnikar I am sure that the folks making the decisions would tell us that our views are not representative of a majority or even a plurality of users.
Well I've mostly reported compiler defects and they've all been fixed (excep the one I reported last week). I do think though that they could be reminded of some of the older cases that's "always been there". The initiative Dalija made in non-tech is good I think. Get them up and in the open. Make Marco Cantù see them and pass them on.
They have built up a significat backlog, at this point it's probably hard to know what's important and what's no longer important.
Asbjørn Heid They do fix more serious bugs promptly. Compiler bugs, and regressions and reproducible crashes fall into that category.
Current problem is FireMonkey that has so many serious issues they don't know where to start, and compiler team has to chase new platforms instead of focusing on improvements.
Dalija Prasnikar True. Doesn't help that they often only fix immediate issues, and don't think a bit more about the issues. Today I noticed they've changed the TVector3D constructors to class functions but left the TQuarternion constructor, which is just a few lines below. Gah! Though why records can have constructors in the first place is another question...
In the last three releases, how many of the repaired defects which were in QC had high ratings? How many defects with high ratings went unrepaired?
ReplyDeleteBill Meyer No idea, but I'm very glad that they fixed my issues which did not have high ratings (as in, they weren't rated), because they were show-stoppers for me.
ReplyDeleteBill Meyer While you cannot count on issues being fixed, some eventually do get fixed because people were pointing them out.
ReplyDeleteAsbjørn Heid My concern is the limited correlation between real severity to customers and application of repairs.
ReplyDeleteDalija Prasnikar I realize that. But can't help wondering at the process through which they decide which to fix. I've never worked for a company that was quite so apparently arbitrary, nor so quick to disparage customers' concerns.
ReplyDeleteI think its a matter of signal to noise ratio. Any tools vendor will tell you other developers scream louder than any other group of end users because:
ReplyDeleteA. They understand the effort involved better than most.
B. Think their requests should carry more weight because of A.
But let's face it. Bug fixes don't attract new customers (though ignoring them can certainly cost you existing ones). New features attract new customers.
Kenneth Cochran The one I have a problem with is the notion that a longer a bug remains unfixed, the smaller becomes its footprint.
ReplyDelete"If we ignore it, eventually we won't have to fix it."
Bill Meyer "As Designed" :)
ReplyDeleteNicholas Ring I have begun to hate that phrase.
ReplyDeleteBill Meyer I know that QC is perceived as black hole, and I hope new bug tracking system will see the light of day sooner than later.
ReplyDeleteBut the more concerning issue is that number of bugs is inherently growing, and even serious bugs don't get fixed promptly.
Lack of manpower and/or organizational skills in action.
Regardless, I just cannot let go, and if there is any chance that some pushing from my side can make the difference I will do it.
Bill Meyer"Need More Info" and "Cannot Reproduce" :-D
ReplyDeleteNicholas Ring Yes, those, too. Especially on well documented and easily reproduced defects which many of us experience.
ReplyDeleteDalija Prasnikar "Wont Do" :-)
ReplyDeleteDalija Prasnikar The number can only increase, when so many new features are being introduced in each new release. My concern now is that, just as with the features issue, defects in the new areas will command the majority of attention, and the desktop issues will remain unresolved.
ReplyDeleteOr worse, resolved with "won't do" or equivalent.
Bill Meyer "Deferred to Next Release"
ReplyDeleteNicholas Ring There is nothing to be sorry about except the fact that we have need to post sarcastic comments like this.
ReplyDeleteDalija Prasnikar And after so many years, the sarcasm is surely warranted.
ReplyDeleteBill Meyer It keeps me alive ;)
ReplyDeleteDalija Prasnikar I am sure that the folks making the decisions would tell us that our views are not representative of a majority or even a plurality of users.
ReplyDeleteWell I've mostly reported compiler defects and they've all been fixed (excep the one I reported last week). I do think though that they could be reminded of some of the older cases that's "always been there". The initiative Dalija made in non-tech is good I think. Get them up and in the open. Make Marco Cantù see them and pass them on.
ReplyDeleteThey have built up a significat backlog, at this point it's probably hard to know what's important and what's no longer important.
Asbjørn Heid They do fix more serious bugs promptly. Compiler bugs, and regressions and reproducible crashes fall into that category.
ReplyDeleteCurrent problem is FireMonkey that has so many serious issues they don't know where to start, and compiler team has to chase new platforms instead of focusing on improvements.
Dalija Prasnikar True. Doesn't help that they often only fix immediate issues, and don't think a bit more about the issues. Today I noticed they've changed the TVector3D constructors to class functions but left the TQuarternion constructor, which is just a few lines below. Gah! Though why records can have constructors in the first place is another question...
ReplyDelete