Board index » off-topic » Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

Re: Incorrectly closed reports: do they get reviewed if I commentonthem?


2007-03-29 07:16:12 AM
off-topic12
"Fernando Madruga" < XXXX@XXXXX.COM >wrote in message
Quote
Nick Hodges (CodeGear) wrote:
>Fernando Madruga wrote:
>
>>It's proving too much work for something that I'm doing for free and
>>apparently upsets so many people.
>
>Who is getting upset at your bug reports?
>

Well, just see the amount of "fuss" this simply thread generated. It
started as a simple question about some reports that were closed and I
felt they needed to be re-open and what could I do for that to happen.

Fernando,
I personally have greatly appreciated all the work you've
gone to in order to detail and submit QC reports.
You've put a lot of time and effort into it. You've created
a VM just for this purpose. You've installed and reinstalled.
But if you are going to put all that time into it, I want you
to get the most possible out of it.
So the shoe is on the other foot.
You submit QC reports because you care. And I really do
appreciate that. It's the same reason I do.
Now, other people, who also care, want those reports to get
opened and fixed. You think that Delphi should have had
more "time in the oven" before release and it could have been
that much better.
The sysops think a few of your reports could have used a few
more details and they could have been that much more useful.
Let's use an analogy of what happens in a busy hospital emergency
room. They do triage on incoming cases. Some need immediate
treatment, others need to be taken care of as time allows.
But the person that does the triage is not the same as the one who
finally treats. If the treating doctor thinks the triage person made
a mistake, there will definitely be feedback.
They could have sent a critical case on the slow track and risked
their life, or they could have sent a non-critical case on the fast track
and wasted valuable time that should have gone to a more critical case.
My point is that the sysops are not the ones that matter. They
are only doing triage for the developers.
It is the R&D developers that really count, in the sense of being
able to reproduce an issue.
If a sysop opens a report, and then R&D can't reproduce it,
then the sysop hears about it. They won't do that too often. 8:-)
That's why they are sticklers for reproducable steps.
But we are all on the same team, because we all care about
Delphi being the best it can be. And we don't want to be
submitting reports that are going to cost R&D a lot of time
figuring out what we are talking about.
That's why I really appreciate the job the sysops do, helping
me create clear clean reports, that can be acted on easily
and get me the results I want.
HTH,
Brad.
 
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

Brad White wrote:
Quote
HTH,
Brad.
Well, thanks for the input. I think my main problem was that when I
started reporting I made the (false) assumption that I would be
reporting for someone actually *using* the product in question, someone
for whom, for instance, RUN=RUN and not RUN=COMPILE, and that I would
not need to explicitly say "Press F9". If I need to make the reports
reproducible by someone who doesn't even know that, then I probably need
too much time to do it in such a way. As another example, saying that
"on my computer I have folders X,Y,Z under My Documents\RAD Studio", is
not that much help when I complain of not having them on mine! I found
out what was going on by someone else's comment later, that it was a
question of an installation choice of where to place "shortcuts" that
actually was used to "create folders" instead (or in addition to). I for
one surely am not expected to know that, but I *would* expect the person
on the other hand to know where/why/how those folders are created and
answer appropriately... Those are two of the type of mishandled reports
because the person on the other side does not know enough of the product
or, as I think was the first case, read the description/steps too
quickly and therefore misinterpreted what I was saying.
I think I'll switch to a middle term: I'll post the reports as best as I
can, but won't bother if they're closed/opened/deleted/whatever. If the
person reading the report does get the message, so fine, if not, too
bad. Maybe someone else will report it better. That's the best I can do
for now.
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

Fernando Madruga wrote:
Quote

Well, thanks for the input.
I would also like to thank you for the time you spent on reporting bugs
to QC. It's really appreciated (from my side).
Quote
I think my main problem was that when I
started reporting I made the (false) assumption that I would be
reporting for someone actually using the product in question, someone
for whom, for instance, RUN=RUN and not RUN=COMPILE, and that I would
not need to explicitly say "Press F9". If I need to make the reports
reproducible by someone who doesn't even know that, then I probably
need too much time to do it in such a way.
I think you're misinterpreting it. By stating they need accurate
stepping, sysops are only minimizing the possibility of 2 common
errors: assuming the other knows exactly what you are thinking, and
chaos (no joke here). I'll try to explain this as short as possible.
When we think something is easy, evident or trivial we tend to assume
that other people will see it the same way. This includes those who we
think have the same or higher "knowledge" level as we.
I'll give you an example: if you go to a bar in Spain and ask for a
coffee with milk the waiter may or may not assume you want sugar. If
you specify you want sugar the waiter may assume you want a bag of
sugar or you want him to bring you a cup of sugar with a spoon so you
can drop into the coffee the quantity of sugar you wish. You're being
ambiguous.
The same applies to bug reporting: if you say "it gives an error when
running my project" you are not specifying if you just do run (F9) or
first you are compiling (ctrl+F9) and then running (F9). While it may
seem trivial for us that both bring you to the same end, it is *not*
when working with programs, since they involve a different set of
actions.
And here is where chaos comes in. Basically, 2 different sets of
actions and options that apparently bring you to the same point do not
really need to do so. As an example, how many times have we seen posts
asking to delete the .dcu files and rebuild the application to see if
that solves the problem? Theoretically speaking, just doing run (F9)
would be the same as deleting the .dcu and then doing run (F9), since
the compiler would rebuild them, but it is *not* always the case.
Your programming environment and options, as well as your set of steps
used to provoque the bug are very important. That is why sysops will
sometimes ask for more accurate steps. Because they either are not
thinking the same as you when following the steps or their environment
and options are not the same as yours, thus making it impossible for
them to reproduce the bug.
humm... a little more lenghty than I wanted. Hope I have explained it
accurately. :-)
Just my 0.02 €,
--
Best regards :-)
Guillem Vicens Meier
Dep. Informática Green Service S.A.
www.clubgreenoasis.com
Contribute to the Indy Docs project: docs.indyproject.org
In order to contact me remove the -nospam
 

{smallsort}

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

Fernando Madruga wrote:
Quote
I made the (false) assumption that I would be reporting
for someone actually *using* the product in question
That assumption is not false. Sysops are chosen because they are users
who have demonstrated familiarity with the product and willingness to
volunteer their free time to try to help make the product better.
True, there is no provision to limit a sysop to working on reports for
one specific product, but if one were to begin acting inappropriately
the situation would be corrected.
Quote
for whom, for instance, RUN=RUN and not RUN=COMPILE
Please try to understand that while you may use these terms correctly,
that is not the case for all reports filed and there are a small
number of people trying to process a rather large number of reports.
Quote
and that I would not need to explicitly say "Press F9".
Please _do_ be that explicit. Not only is this the best way to avoid
misunderstanding, but often there are several different choices for
how to "run" your program (F9, Run | Run, Click run button on toolbar,
double-click on the exe file in explorer) and in some cases the exact
way you do it may affect the results.
Quote
if I need to make the reports reproducible by someone who doesn't
even know that, then I probably need too much time to do it
Including steps like:
F9 (Run)
should not be time consuming. This also allows you to try your exact
steps before filing the report - you may find that you have left out
an important detail. At least that often happens for me.
Quote
"on my computer I have folders X,Y,Z under My Documents\RAD Studio", is
not that much help when I complain of not having them on mine!
Nor is your statement that you don't have them on yours very much help
to the sysop. All that is known from either is that your systems are
different, and you gave no steps to show how this happened on your
system.
Quote
I for one surely am not expected to know that, but I *would* expect
the person on the other hand to know where/why/how those folders
are created
The person on the other end is a user, just like you, and the product
is new. As we become more familiar with the product, things like this
become known and the process is often a dialog among different users
and people form CodeGear.
Even the programmers at CodeGear do not always know the results of all
of the possible ways the products can be installed/used. If they did,
then there would be no bugs. To expect every one of the sysops to be
able to correctly diagnose a problem based only on a symptom seems
unreasonable to me.
Quote
I think I'll switch to a middle term: I'll post the reports as best
as I can, but won't bother if they're closed/opened/deleted/whatever.
I hope that you _will_ continue to pay attention. Often a dialog
between the user experiencing a problem and a sysop or CodeGear
personnel is the best way to get to the bottom of the issue.
By posting here or by commenting on the report or adding to the steps
you can keep the dialog active. I hope you can understand that we are
all interested in improving the products and tracking down bugs.
- Leo
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

"Fernando Madruga" < XXXX@XXXXX.COM >wrote in message
Quote
I think I'll switch to a middle term: I'll post the reports as best as I
can, but won't bother if they're closed/opened/deleted/whatever. If the
person reading the report does get the message, so fine, if not, too bad.
Maybe someone else will report it better. That's the best I can do for
now.
Well, that's better than nothing, but I hate to see you give up.
There needs to be some back and forth communication.
I submit a report that I think is clear.
They read it and think it is clear enough to submit
or ask for more info.
We shouldn't take that personally, as if they were
saying that our reports are no good or not important.
If I'm not going to follow up on my reports and shepherd
them through the process, what is the point of submitting them?
--
Thanks,
Brad.
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

JED wrote:
Quote
Sebastian Modersohn wrote:

>which has screenshots attached! Not sure if this is a web service or
>client issue. I've seen truncated reports in JED, QC too but those
>might be truncated because someone edited the report with the
>truncated description/steps.

I can assure everyone that use my client, it does not truncate text.
Either it is the updated CodeGear Win32 client or a WebService issue.

I believe it's the QC Client shipped with Delphi 2007: it did it to some
of my reports and that would not happen later with JED QC... I already
submitted a QC about it:
Report No: 43140 Status: Reported
Not all data retrieved when querying my reports
qc.codegear.com/wc/qcmain.aspx
I believe it posts ok, because I'd see the whole text in the web
interface, but retrieves a few bytes less, so upon editing, that cropped
text is sent resulting in cropped reports.
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

Sebastian Modersohn wrote:
Quote
which has screenshots attached! Not sure if this is a web service or
client issue. I've seen truncated reports in JED, QC too but those
might be truncated because someone edited the report with the
truncated description/steps.
I can assure everyone that use my client, it does not truncate text.
Either it is the updated CodeGear Win32 client or a WebService issue.
--
TJSDialog - TaskDialog for other operating systems:
www.jed-software.com/jsd.htm
Visual Forms IDE Add In: www.jed-software.com/vf.htm
Blog: jedqc.blogspot.com
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

JED wrote:
Quote
I can assure everyone that use my client, it does not truncate text.
What version of Delphi are you using to build your client, currently?
--
John Kaster blogs.codegear.com/johnk
CodeGear Developer Network: dn.codegear.com
Features and bugs: qc.codegear.com
Got source? cc.codegear.com
If it's not here, it's not happening: ec.codegear.com
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

John Kaster (CodeGear) wrote:
Quote
JED wrote:

>I can assure everyone that use my client, it does not truncate text.

What version of Delphi are you using to build your client, currently?
The version that most people use is built in Delphi 2005 I think.
Recent builds are done with Delphi 2006.
--
TJSDialog - TaskDialog for other operating systems:
www.jed-software.com/jsd.htm
Visual Forms IDE Add In: www.jed-software.com/vf.htm
Blog: jedqc.blogspot.com
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

John Kaster (CodeGear) wrote:
Quote
JED wrote:

>I can assure everyone that use my client, it does not truncate text.

What version of Delphi are you using to build your client, currently?
I can compile a version in D2007 if you think that might be the
problem.
I don't use data aware controls for editing a report so it might not be
a good test case. That is if the win client still uses data aware memos.
--
TJSDialog - TaskDialog for other operating systems:
www.jed-software.com/jsd.htm
Visual Forms IDE Add In: www.jed-software.com/vf.htm
Blog: jedqc.blogspot.com
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

JED wrote:
Quote
Recent builds are done with Delphi 2006.
I thought so. You would have had the same issue had you built with
Delphi 2007. Fortunately, a fix for it is being tested right now.
--
John Kaster blogs.codegear.com/johnk
CodeGear Developer Network: dn.codegear.com
Features and bugs: qc.codegear.com
Got source? cc.codegear.com
If it's not here, it's not happening: ec.codegear.com
 

Re:Re: Incorrectly closed reports: do they get reviewed if I commentonthem?

JED wrote:
Quote
I don't use data aware controls for editing a report so it might not
be a good test case.
It wasn't that.
--
John Kaster blogs.codegear.com/johnk
CodeGear Developer Network: dn.codegear.com
Features and bugs: qc.codegear.com
Got source? cc.codegear.com
If it's not here, it's not happening: ec.codegear.com