Re: [Qt-creator] Creator stability

2010-03-29 Thread Eike Ziller

On Mar 29, 2010, at 1:55 AM, King Bill (Nokia-D-Qt/Brisbane) wrote:

> On 03/26/2010 07:18 PM, Ziller Eike (Nokia-D-Qt/Berlin) wrote:
>> 
>> Well, except that we don't have windows binary builds for for various 
>> reasons at the moment, we have nightly binary builds of Qt Creator master on 
>> top of Qt 4.7 (or before a release the release branches), which you could 
>> use to find the last working thing for you. (Of course you should still bug 
>> us about issues you find, increasing the chances that it will be fixed in a 
>> timely manner.)
>> ftp://ftp.qt.nokia.com/qtcreator/snapshots
>> 
>> The known quality of the binary snapshots is obviously that it had minimal 
>> build testing ;)
>> 
> Maybe put this link up on the gitorious page, I (for one) was very
> unaware of the snapshots directory.

You can find this information and lots of other useful things under the 
gitorious Qt Creator Wiki link:
http://qt.gitorious.org/qt-creator/pages/Home

I now put a link to that page also in the project description, perhaps that 
helps people find it.

-- 
Eike Ziller
Software Engineer
Nokia, Qt Development Frameworks

Nokia gate5 GmbH
Firmensitz: Invalidenstr. 117, 10115 Berlin, Germany
Registergericht: Amtsgericht Charlottenburg, Berlin: HRB 106443 B
Umsatzsteueridentifikationsnummer: DE 812 845 193
Geschäftsführer: Dr. Michael Halbherr, Karim Tähtivuori




___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-28 Thread Bill King
On 03/26/2010 07:18 PM, Ziller Eike (Nokia-D-Qt/Berlin) wrote:
>
> Well, except that we don't have windows binary builds for for various reasons 
> at the moment, we have nightly binary builds of Qt Creator master on top of 
> Qt 4.7 (or before a release the release branches), which you could use to 
> find the last working thing for you. (Of course you should still bug us about 
> issues you find, increasing the chances that it will be fixed in a timely 
> manner.)
> ftp://ftp.qt.nokia.com/qtcreator/snapshots
>
> The known quality of the binary snapshots is obviously that it had minimal 
> build testing ;)
>   
Maybe put this link up on the gitorious page, I (for one) was very
unaware of the snapshots directory.


-- 
Bill King, Software Engineer
Qt Development Frameworks, Nokia Pty Ltd
Brisbane Office

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-28 Thread Bill King
On 03/26/2010 05:06 PM, Raggi Roberto (Nokia-D-Qt/Berlin) wrote:
>
> Hi,
>
> On Mar 26, 2010, at 12:15 AM, ext Bill King wrote:
>>>
>> In this case I got a "doesn't happen here" and then it was dropped
>> (maybe it was worked on internally), but the email responses from
>> roberto showed that it had pretty much been ignored (even after me
>> adding extra information, and then even a patch to fix the issue).
>
> I'm pretty sure the only email I sent about this topic was a short
> reply on our internal mailing list where I just said 1) we can't
> reproduce it 2) we don't have a working valgrind for OSX 10.6 and 3)
> the email ends with - I guess we have to investigate what's going
> wrong with the binder.
The original bug report was not transitioned from reported to accepted
for over 3 weeks. Considering our recent focus on doing at the very
least that, yeah, that's the issue I was trying to point out.
>
> We applied your patch (commit b0b95f88756dbdf4981c97a325734300a65d8268
> ) but unfortunately it was creating quite a number of regressions in
> our refactoring engine so we had to revert it
> (change 933e52888e8e69f876bdf4640379ecdddb2c2c9c ).
>
> And by the way, we tried to reproduce the crash on 4 different
> computers. We tried Linux, OSX and Windows and we also checked with
> memcheck so who knows, maybe we can get some help from the guys on
> this mailing list. Please, can someone try to reproduce this
> bug http://bugreports.qt.nokia.com/browse/QTCREATORBUG-807 ? I'm sure
> we will have a better understanding of the problem if we have more
> reports.
I had the crash on linux and 2 windows boxes, and both purify and
memcheck reported the issue. Very odd.
>
> About the quality of the C++ front-end. Well, i can tell you that we
> are testing it quite a lot. 
> We have 3 tools that we use daily to test the front-end with the gcc
> test suite and with the Qt source code (in creator/tests/manual,
> cplusplus-dump, cplusplus, plain-c++).
Testing's good :D
>
> Yes, it can happen that we introduce regressions but that's because an
> incremental front-end for C++ need to parse both correct and
> invalid(unfinished?) code and i'm sure you can see how many "wrong"
> configuration you can have when parsing a few million lines of C++
> code so the test suite needs to be a bit special and that's the reason
> why we're using gcc/gcc/testsuite.
>
> I think your attack to us is wrong and not fare.
Fair/not fair, at the end of the day, we're professionals, and as such,
we have to take the good with the bad. Part of that is maintaining
communication, part of it also is realising when things are going wrong
and buckling down to rectify that situation. I've been talking with Ed,
I'll leave it with that for now :)
>
> ciao robe
>


-- 
Bill King, Software Engineer
Qt Development Frameworks, Nokia Pty Ltd
Brisbane Office

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-26 Thread Kai Koehne
ext John Vilburn wrote:
> [...]
> Creator is working fine for general Qt development for me. My top 
> request is for Creator to include a way in Design mode to specify 
> actions and transitions for QML. I know that opens up a lot of 
> possibilities and a lot of work would be involved. But a good and very 
> useful start would be to allow the designer to specify a transition 
> between states and to specify that a mouse click or double click in a 
> certain mousearea will cause the transition from state A to state B.

Hi John,

Support for transitions is also high on our TODO list ... but we will 
not be able to work on it for 2.0, unfortunately. But I agree that it's 
something you really want to have.

Regards

Kai Koehne

-- 
Kai Koehne
Software Engineer
Nokia, Qt Development Frameworks

Nokia gate5 GmbH
Firmensitz: Invalidenstr. 117, 10115 Berlin, Germany
Registergericht: Amtsgericht Charlottenburg, Berlin: HRB 106443 B
Umsatzsteueridentifikationsnummer: DE 812 845 193
Geschäftsführer: Dr. Michael Halbherr, Karim Tähtivuori
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-26 Thread Eike Ziller

On Mar 26, 2010, at 12:20 AM, King Bill (Nokia-D-Qt/Brisbane) wrote:

> On 03/25/2010 07:40 PM, Ziller Eike (Nokia-D-Qt/Berlin) wrote:
>> 
>> Especially regarding Qt Creator's Qml support, a CI system for Qt Creator 
>> would be mostly pointless.
>> 
> Unless it produced a point of stability, say a branch in both qt and
> creator called qtcreator_stable, that's had minimal build and run
> testing, it's a known quality that people can understand, currently
> that's not even available, and you have to rewind one or both until you
> get something that works. (something that's both difficult and time
> consuming).

Well, except that we don't have windows binary builds for for various reasons 
at the moment, we have nightly binary builds of Qt Creator master on top of Qt 
4.7 (or before a release the release branches), which you could use to find the 
last working thing for you. (Of course you should still bug us about issues you 
find, increasing the chances that it will be fixed in a timely manner.)
ftp://ftp.qt.nokia.com/qtcreator/snapshots

The known quality of the binary snapshots is obviously that it had minimal 
build testing ;)

>> Because most of the time when Qt Creator builds break in Qml support, it is 
>> because of changes in Qt/QML that don't yet have corresponding changes in Qt 
>> Creator.
>> We already have to handle the problem that "synchronized changes" in Qt and 
>> Qt Creator first need the change in Qt/mainline, and only then the 
>> corresponding change in Qt Creator can be pushed. That wouldn't get better 
>> with yet another process in between for the Qt Creator changes.
>> Also the current CI system is barely able to handle Qt at the moment, 
>> resources-wise (that will hopefully be mended, but takes time).
>> 
>> We have nighlty builds btw, and there is a (admittedly slow) progress 
>> ongoing to move that to pulse, so we'll hopefully profit from the 
>> infrastructure there in the longer run.
>> 
>> 
>> 
> Can I help this somehow? Rohan's just around the corner, so I can pester
> him in person if it's that that's needed.

I don't think that pestering is what is needed, because afaik he's got the 
important Job of making the CI system (more) usable for Qt already,
and I wouldn't like to keep him from doing that first :)

-- 
Eike Ziller
Software Engineer
Nokia, Qt Development Frameworks

Nokia gate5 GmbH
Firmensitz: Invalidenstr. 117, 10115 Berlin, Germany
Registergericht: Amtsgericht Charlottenburg, Berlin: HRB 106443 B
Umsatzsteueridentifikationsnummer: DE 812 845 193
Geschäftsführer: Dr. Michael Halbherr, Karim Tähtivuori




___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-26 Thread Tobias Hunger
Hello Bill!

On 26.03.2010 00:15, King Bill (Nokia-D-Qt/Brisbane) wrote:
> No, in this case one of your team had a go at me for trying to bring to
> light the quality issues I'd seen.

Unfortunately we do get too passionate about defending our baby at times:-(

>> We are trying hard to keep the response time low. From my experience as
>> an developer working on Qt creator my impression is that most bug
>> reports are looked at in less than a day. We do not always update the
>> reports, so this might not always be visible to you as the reporter. We
>> might be able to improve here, but that is a issue of priority: Do we
>> want to keep implementing new features or update the bugtracker?
>>
> In this case I got a "doesn't happen here" and then it was dropped
> (maybe it was worked on internally), but the email responses from
> roberto showed that it had pretty much been ignored (even after me
> adding extra information, and then even a patch to fix the issue).

Since I happen to sit next to Roberto I can assure you that something 
was happening. Roberto already replied describing what he and the other 
C++ guys did, so I do not need to repeat this here.

My impression is that in your case the biggest issue is a communication 
failure: You are understandably upset about not getting any feedback and 
had to assume nothing happened. We on the other hand were surprised by 
your reaction and considered it to be out of proportion: After all we 
had seen what had been done to recreate the issue.

I will try to improve my communication in JIRA and do hope that others 
will do so too.

>> We all use Qt Creator daily, so we tend to notice crashes and it not
>> building. That is why they do get fixed pretty fast (at least in my
>> experience).
>>
> Next day tho :/

Brisbane is not really located in the best timezone to receive timely 
responses from our Germany-based team, that is true:-)

> This is why I suggested maybe a CI solution. minimals
> test of "does it build?" "does it start?" "does it open a c++ project?"
> (this is where mine was crashing) "does it open a qml project?". Even
> this would boost the quality perception through the roof.

We do nightly builds. It might be a good idea to document which 
revisions were build so that you and others can get a hint as to which 
recent versions at least passed this *very* basic test:-)

The other suggested tests do require user interaction at this time, so 
they are not really feasible right now for those daily snapshots:-( We 
might be able to do UI testing at some point, but that is not a simple 
thing to do for the master branch: We do polish the UI there all the 
time, so maintaining UI based tests is very labor-intensive.

>> I do understand your feelings and like the passion with which you are
>> fighting for a better creator!
>>
>>
> It's an incredible project, and yeah, it's a project you _can_ get
> passionate about ;)

Great that you still like it!

Best Regards,
Tobias

-- 
Tobias Hunger
Software Engineer
Nokia, Qt Development Frameworks

Nokia gate5 GmbH
Firmensitz: Invalidenstr. 117, 10115 Berlin, Germany
Registergericht: Amtsgericht Charlottenburg, Berlin: HRB 106443 B
Umsatzsteueridentifikationsnummer: DE 812 845 193
Geschäftsführer: Dr. Michael Halbherr, Karim Tähtivuori
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-26 Thread roberto.raggi

Hi,

On Mar 26, 2010, at 12:15 AM, ext Bill King wrote:

In this case I got a "doesn't happen here" and then it was dropped
(maybe it was worked on internally), but the email responses from
roberto showed that it had pretty much been ignored (even after me
adding extra information, and then even a patch to fix the issue).

I'm pretty sure the only email I sent about this topic was a short reply on our 
internal mailing list where I just said 1) we can't reproduce it 2) we don't 
have a working valgrind for OSX 10.6 and 3) the email ends with - I guess we 
have to investigate what's going wrong with the binder.

We applied your patch (commit b0b95f88756dbdf4981c97a325734300a65d8268 ) but 
unfortunately it was creating quite a number of regressions in our refactoring 
engine so we had to revert it (change 933e52888e8e69f876bdf4640379ecdddb2c2c9c 
).

And by the way, we tried to reproduce the crash on 4 different computers. We 
tried Linux, OSX and Windows and we also checked with memcheck so who knows, 
maybe we can get some help from the guys on this mailing list. Please, can 
someone try to reproduce this bug 
http://bugreports.qt.nokia.com/browse/QTCREATORBUG-807 ? I'm sure we will have 
a better understanding of the problem if we have more reports.

About the quality of the C++ front-end. Well, i can tell you that we are 
testing it quite a lot.
We have 3 tools that we use daily to test the front-end with the gcc test suite 
and with the Qt source code (in creator/tests/manual, cplusplus-dump, 
cplusplus, plain-c++).

Yes, it can happen that we introduce regressions but that's because an 
incremental front-end for C++ need to parse both correct and 
invalid(unfinished?) code and i'm sure you can see how many "wrong" 
configuration you can have when parsing a few million lines of C++ code so the 
test suite needs to be a bit special and that's the reason why we're using 
gcc/gcc/testsuite.

I think your attack to us is wrong and not fare.

ciao robe

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Bill King
On 03/25/2010 08:18 PM, Poenitz Andre (Nokia-D-Qt/Berlin) wrote:
>
> There is no promise (and never has been) that Creator master branch 
> works (including even the weirdest definitions of "works"...)
>
> If you want something that is known to compile, use a released version.
>
> We are happy about anybody actually using master and notifying us about
> any breakage but there's no guarantee whatsoever that it does anything 
> useful. We are doing our best to fix breakages quickly, though, and so far
> I had the impression that readers of this list and people on IRC find this
> concept both comprehensible and acceptable.
>   
See my "not a perception we should be portraying in a public codebase",
and my solution to this (CI build/tag/minimal run testing).
>   
>> I have reports both from internal and external users on a regular
>> basis of either creator not building, or creator crashing upon startup.
>> 
> Startup crashes are most of the time caused by unclean builds. 
> git clean -dxf && qmake -r && make && tons of coffee && be done.
>
> That's not the main problem here. The first problem is that people seem
> to send their reports about Qt Creator to you and not to us, and the 
> second is that even then those reports do not end up, say, here on the
> mailing list or in JIRA (barring the current incident).
>   
:) I'm a qt dev, these are all proffesional devs, this is the first
thing we tried, when the days it doesn't run outnumber the days it does
run, that's a quality issue.
>   
>> I have pushed the externals to submit bugreports,
>> 
> Good. Chances to get stuff fixed are way better with a decent bug report.
>
>   
>> but, again, my experience here has been less than glowing.
>> 
> Mind to elaborate? As in "your favourite bug has not been fixed the 
> next day"? In that case, honestly, I'd have a hard time to feign surprise.
>   
"Doesn't happen here" and then perceived as dropped (no response to
extra information, no transition to accepted, only now transitioned to
open/accepted and prioritsed after nearly 3 weeks).
>   
>> Being close to the front lines of the project, sometimes this can be missed.
>>
>> To fix these issues, can we implement some sort of staging/CI system
>> like we have for Qt?
>> 
> Is that what you really want? 
>
> No integrations for a fortnight because some test broke?
>   
If it takes you a fortnight to get a passing branch, then you have much
greater issues than we initially suspected. Pulse builds are down to 4
hours. A minimal set of testing shuould take 3 minutes, so 4 hours
turnaround if you haven't broken the build. If you have, then it won't
integrate until it's fixed. If that takes a fortnight, then yeah, time
to hit up HR for more staff.
> I guess that would quickly yield pretty dry blood at the bleeding edge.
>
> You can certainly setup a "really stable" branch guarded by CI if you
> think this helps, but we basically have that in form of releases.
>   
Not even pretty stable, just a "it compiles, it starts, it opens
projects". Most people can accept that, and it will increase the
stability perception multiple-fold.
>   
>> For qml usage, bleeding edge is the only choice currently, and the
>> perceived quality of creator from the bleeding edgers is that
>> creator's... not usable at all. That is not the Qt way, and not a
>> perception we should be having. Creator is an excellent product, and one
>> I use daily, and I'd like to go back to bleeding edge, as that's where
>> all the cool new features are :)
>> 
> It's not a perception I have either. I see Creator master branch used in 
> "production" outside the company basically daily, and while there are days
> when one regrets the pull and has to back up it usually work. This is not
> using QML, though, but application startup is most certainly covered.
>
> So back to business: If you have an issue, mention it on #qt-creator
> or file a bug.
>
> Andre'
>
> ___
> Qt-creator mailing list
> Qt-creator@trolltech.com
> http://lists.trolltech.com/mailman/listinfo/qt-creator
>   


-- 
Bill King, Software Engineer
Qt Development Frameworks, Nokia Pty Ltd
Brisbane Office

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Bill King
On 03/25/2010 07:40 PM, Ziller Eike (Nokia-D-Qt/Berlin) wrote:
>
> Especially regarding Qt Creator's Qml support, a CI system for Qt Creator 
> would be mostly pointless.
>   
Unless it produced a point of stability, say a branch in both qt and
creator called qtcreator_stable, that's had minimal build and run
testing, it's a known quality that people can understand, currently
that's not even available, and you have to rewind one or both until you
get something that works. (something that's both difficult and time
consuming).
> Because most of the time when Qt Creator builds break in Qml support, it is 
> because of changes in Qt/QML that don't yet have corresponding changes in Qt 
> Creator.
> We already have to handle the problem that "synchronized changes" in Qt and 
> Qt Creator first need the change in Qt/mainline, and only then the 
> corresponding change in Qt Creator can be pushed. That wouldn't get better 
> with yet another process in between for the Qt Creator changes.
> Also the current CI system is barely able to handle Qt at the moment, 
> resources-wise (that will hopefully be mended, but takes time).
>
> We have nighlty builds btw, and there is a (admittedly slow) progress ongoing 
> to move that to pulse, so we'll hopefully profit from the infrastructure 
> there in the longer run.
>
>   
>
Can I help this somehow? Rohan's just around the corner, so I can pester
him in person if it's that that's needed.

-- 
Bill King, Software Engineer
Qt Development Frameworks, Nokia Pty Ltd
Brisbane Office

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Bill King
On 03/25/2010 07:08 PM, Hunger Tobias (Nokia-D/Berlin) wrote:
>
> Hey Bill!
>
>   
>> I thought we were an open company, part of that is being open to
>> criticism when things are not working.
>> 
> I really do not understand this critique:
>   * We did investigate the issue
>   * We evaluated the patch
>   * We asked you to move this discussion from an internal list to a 
> public one.
>
> So where exactly were we unresponsive or not open?
>   
No, in this case one of your team had a go at me for trying to bring to
light the quality issues I'd seen.
>   
>> Creator mainline is not working. I have reports both from internal and
>> external users on a regular basis of either creator not building, or
>> creator crashing upon startup.
>> 
> Yes, there are issues with creator sometimes not building. Unfortunately 
> that can happen in a master branch.
>   
I think having a working snapshot may definitely help this. We used to
have a qtopia_stable tag in p4, for just this reason, to give us a
stable platform so that we could do our qtopia work for the day.
Something similar...
> We are trying hard to keep the response time low. From my experience as 
> an developer working on Qt creator my impression is that most bug 
> reports are looked at in less than a day. We do not always update the 
> reports, so this might not always be visible to you as the reporter. We 
> might be able to improve here, but that is a issue of priority: Do we 
> want to keep implementing new features or update the bugtracker?
>   
In this case I got a "doesn't happen here" and then it was dropped
(maybe it was worked on internally), but the email responses from
roberto showed that it had pretty much been ignored (even after me
adding extra information, and then even a patch to fix the issue).
>
> We all use Qt Creator daily, so we tend to notice crashes and it not 
> building. That is why they do get fixed pretty fast (at least in my 
> experience).
>   
Next day tho :/ This is why I suggested maybe a CI solution. minimals
test of "does it build?" "does it start?" "does it open a c++ project?"
(this is where mine was crashing) "does it open a qml project?". Even
this would boost the quality perception through the roof.
>   
>> For qml usage, bleeding edge is the only choice currently, and the
>> perceived quality of creator from the bleeding edgers is that
>> creator's... not usable at all.
>> 
> This does differ quite drastically from the experience we have here. I 
> am not saying that you are not having this trouble, just that I do 
> experience the quality of creator as very good overall.
>
> So now we need to find out where our use cases differ and how we can 
> improve the issue for your workload. I am afraid the best way to do this 
> is by reporting bugs, which you unfortunately seem to not agree with. Do 
> you have a suggestion on how we can improve the situation for you (and 
> others in this thread) short-term?
>   
See the bit above.
>   
>> That is not the Qt way, and not a
>> perception we should be having. Creator is an excellent product,
>> 
> Thanks! Finally something we can agree upon;-)
>
>   
>> and one
>> I use daily, and I'd like to go back to bleeding edge, as that's where
>> all the cool new features are :)
>> 
> I do understand your feelings and like the passion with which you are 
> fighting for a better creator!
>
>   
It's an incredible project, and yeah, it's a project you _can_ get
passionate about ;)

-- 
Bill King, Software Engineer
Qt Development Frameworks, Nokia Pty Ltd
Brisbane Office

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread John Vilburn

On Mar 24, 2010, at 10:10 PM,   
wrote:

> Well, you can stick to e.g. the Alpha, and even creator 1.3 has some limited 
> QML support. The QML text editor + runtime infrastructure has been pretty 
> stable for me anyhow. Regarding QmlDesigner - yeah, the quality is not yet up 
> to where it should be, but then again it's still in the Alpha stages, and one 
> part of the problem is also that we do have received rather minimal feedback 
> so far. So please, people - if you find something, report it! 
> 

Creator is working fine for general Qt development for me. My top request is 
for Creator to include a way in Design mode to specify actions and transitions 
for QML. I know that opens up a lot of possibilities and a lot of work would be 
involved. But a good and very useful start would be to allow the designer to 
specify a transition between states and to specify that a mouse click or double 
click in a certain mousearea will cause the transition from state A to state B.

Thanks,
John

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Victor Sardina
Frank:

I agree with most of what you wrote. I have used the master branch to 
build Creator from source for a while now, and have to regrets in that 
regard. Yes, sometimes that have forced me to revamp my Qt installation, 
but at the end nothing I wouldn't have ended up doing sooner or later. 
QML? I looked into it, but found very little documentation, or actual 
examples of how that would help me create a useful GUI (I know the 
concept, but not how to apply that to my needs). I don't want anyone to 
take the previous statement as a critic. Everyone posting here probably 
have different needs, programming habits (and vices), and even preferred 
tools.

Nobody can keep everybody happy 100% of the time. You try that and will 
certainly sink sooner than later. I would actually take some comments as 
what I call "appraisals in disguise", as they indicate that people with 
a very wide range of interests, views, needs, and habits, have started 
to regard Creator as a tool they want to use, or that they have to de 
facto consider the new kid in the block. I remember a while back that 
when git first came out it cause something similar...I think that people 
should perhaps have a little more patience, and let the team do its 
work: everything looks simpler from the outside.

When I have posted a question, a bug, or reported something weird going 
on, I have always received an answer in a very timely manner, either 
from the Trolls themselves, or some of the users in the community who 
have taken pity of my ignorance: thank you.

Victor

Frank Siegert wrote:
> André Pönitz, Thursday 25 March 2010:
>   
>> There is no promise (and never has been) that Creator master branch
>> works (including even the weirdest definitions of "works"...)
>>
>> If you want something that is known to compile, use a released version.
>>
>> We are happy about anybody actually using master and notifying us about
>> any breakage but there's no guarantee whatsoever that it does anything
>> useful. We are doing our best to fix breakages quickly, though, and so
>> far I had the impression that readers of this list and people on IRC
>> find this concept both comprehensible and acceptable.
>> 
>
> As a completely external user, I would like to support Andre's statement 
> here. I have for several periods been using master and found it incredibly 
> stable in comparison to all other projects' master/trunk/HEAD/... I have 
> used so far.
>
> Furthermore I am hugely impressed by the quick turn-around time on 
> bugfixes, feature implementations and also mailing list replies!
> (if you ignore the occasional "Nokia(!) should fix my favorite bug right 
> now" mail from external users, who rightfully have been pointed to the 
> open model of Creator development).
>
> I have to admit though, that I am not using anything related to QML.
>
> Thanks all for providing such a nice product in such an open way!
> Frank
>
> ___
> Qt-creator mailing list
> Qt-creator@trolltech.com
> http://lists.trolltech.com/mailman/listinfo/qt-creator
>   

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Frank Siegert
André Pönitz, Thursday 25 March 2010:
> There is no promise (and never has been) that Creator master branch
> works (including even the weirdest definitions of "works"...)
> 
> If you want something that is known to compile, use a released version.
> 
> We are happy about anybody actually using master and notifying us about
> any breakage but there's no guarantee whatsoever that it does anything
> useful. We are doing our best to fix breakages quickly, though, and so
> far I had the impression that readers of this list and people on IRC
> find this concept both comprehensible and acceptable.

As a completely external user, I would like to support Andre's statement 
here. I have for several periods been using master and found it incredibly 
stable in comparison to all other projects' master/trunk/HEAD/... I have 
used so far.

Furthermore I am hugely impressed by the quick turn-around time on 
bugfixes, feature implementations and also mailing list replies!
(if you ignore the occasional "Nokia(!) should fix my favorite bug right 
now" mail from external users, who rightfully have been pointed to the 
open model of Creator development).

I have to admit though, that I am not using anything related to QML.

Thanks all for providing such a nice product in such an open way!
Frank

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread André Pönitz
On Thursday 25 March 2010 01:50:12 ext ext Raul Fernandes wrote:
> Hi,
> 
> as user of different IDEs, I don't understand why qtcreator is trying
> to provide more advanced features (such as QML support) if more
> basic features are not implement/working properly. Important point: 
> how can the team adopts qtcreator as main development environment
> if it is not even getting more stable? 

Help fixing it? It's Open Source after all. If you are not happy with 
the priorities the payed developers set or have to set you are free 
to implement the missing basic features and file a merge request.

This has worked for a quite a few features in the past and there's
no reason to believe it would not work again.

Also, please note that what you are saying is already in contrast to
what Bill said: He wanted QML stable whereas you don't want it at all.
So making everybody completely happy is provably not possible.

Regards,
Andre
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread André Pönitz
On Thursday 25 March 2010 00:46:46 King Bill (Nokia-D-Qt/Brisbane) wrote:
> Bringing this over here as was requested ;)
> I was just criticized for criticizing the quality of creator ("on the
> basis of one bug report").
> 
> I thought we were an open company, part of that is being open to
> criticism when things are not working.
> 
> Creator mainline is not working. 

There is no promise (and never has been) that Creator master branch 
works (including even the weirdest definitions of "works"...)

If you want something that is known to compile, use a released version.

We are happy about anybody actually using master and notifying us about
any breakage but there's no guarantee whatsoever that it does anything 
useful. We are doing our best to fix breakages quickly, though, and so far
I had the impression that readers of this list and people on IRC find this
concept both comprehensible and acceptable.

> I have reports both from internal and external users on a regular
> basis of either creator not building, or creator crashing upon startup.

Startup crashes are most of the time caused by unclean builds. 
git clean -dxf && qmake -r && make && tons of coffee && be done.

That's not the main problem here. The first problem is that people seem
to send their reports about Qt Creator to you and not to us, and the 
second is that even then those reports do not end up, say, here on the
mailing list or in JIRA (barring the current incident).

> I have pushed the externals to submit bugreports,

Good. Chances to get stuff fixed are way better with a decent bug report.

> but, again, my experience here has been less than glowing.

Mind to elaborate? As in "your favourite bug has not been fixed the 
next day"? In that case, honestly, I'd have a hard time to feign surprise.

> Being close to the front lines of the project, sometimes this can be missed.
> 
> To fix these issues, can we implement some sort of staging/CI system
> like we have for Qt?

Is that what you really want? 

No integrations for a fortnight because some test broke?

I guess that would quickly yield pretty dry blood at the bleeding edge.

You can certainly setup a "really stable" branch guarded by CI if you
think this helps, but we basically have that in form of releases.

> For qml usage, bleeding edge is the only choice currently, and the
> perceived quality of creator from the bleeding edgers is that
> creator's... not usable at all. That is not the Qt way, and not a
> perception we should be having. Creator is an excellent product, and one
> I use daily, and I'd like to go back to bleeding edge, as that's where
> all the cool new features are :)

It's not a perception I have either. I see Creator master branch used in 
"production" outside the company basically daily, and while there are days
when one regrets the pull and has to back up it usually work. This is not
using QML, though, but application startup is most certainly covered.

So back to business: If you have an issue, mention it on #qt-creator
or file a bug.

Andre'

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Eike Ziller

On Mar 25, 2010, at 12:46 AM, King Bill (Nokia-D-Qt/Brisbane) wrote:

> Bringing this over here as was requested ;)
> 
> 
> I was just criticized for criticizing the quality of creator ("on the
> basis of one bug report").
> 
> I thought we were an open company, part of that is being open to
> criticism when things are not working.
> 
> Creator mainline is not working. I have reports both from internal and
> external users on a regular basis of either creator not building, or
> creator crashing upon startup. I have pushed the externals to submit
> bugreports, but, again, my experience here has been less than glowing.
> Being close to the front lines of the project, sometimes this can be missed.
> 
> To fix these issues, can we implement some sort of staging/CI system
> like we have for Qt?

Especially regarding Qt Creator's Qml support, a CI system for Qt Creator would 
be mostly pointless.
Because most of the time when Qt Creator builds break in Qml support, it is 
because of changes in Qt/QML that don't yet have corresponding changes in Qt 
Creator.
We already have to handle the problem that "synchronized changes" in Qt and Qt 
Creator first need the change in Qt/mainline, and only then the corresponding 
change in Qt Creator can be pushed. That wouldn't get better with yet another 
process in between for the Qt Creator changes.
Also the current CI system is barely able to handle Qt at the moment, 
resources-wise (that will hopefully be mended, but takes time).

We have nighlty builds btw, and there is a (admittedly slow) progress ongoing 
to move that to pulse, so we'll hopefully profit from the infrastructure there 
in the longer run.

> For qml usage, bleeding edge is the only choice currently, and the
> perceived quality of creator from the bleeding edgers is that
> creator's... not usable at all. That is not the Qt way, and not a
> perception we should be having. Creator is an excellent product, and one
> I use daily, and I'd like to go back to bleeding edge, as that's where
> all the cool new features are :)
> 
> -- 
> Bill King, Software Engineer
> Qt Development Frameworks, Nokia Pty Ltd
> Brisbane Office
> 
> ___
> Qt-creator mailing list
> Qt-creator@trolltech.com
> http://lists.trolltech.com/mailman/listinfo/qt-creator

-- 
Eike Ziller
Software Engineer
Nokia, Qt Development Frameworks

Nokia gate5 GmbH
Firmensitz: Invalidenstr. 117, 10115 Berlin, Germany
Registergericht: Amtsgericht Charlottenburg, Berlin: HRB 106443 B
Umsatzsteueridentifikationsnummer: DE 812 845 193
Geschäftsführer: Dr. Michael Halbherr, Karim Tähtivuori




___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread kai.koehne


From: qt-creator-boun...@trolltech.com [qt-creator-boun...@trolltech.com] On 
Behalf Of ext Brett Morgan [brett.mor...@gmail.com]
Sent: Thursday, March 25, 2010 9:29 AM
To: qt-creator@trolltech.com
Subject: Re: [Qt-creator] Creator stability

> Kai,
>
> I'm compiling and using qt-creator in multiple environments, primarily OSX 
> 10.6 and a variety of Linux distributions. I'm happy to give you details of 
> my environments and stack > traces.

Cool :) If it's really an 'obvious' thingy like stuff not compiling, the direct 
lane is to grap someone on #qt-creator (freenode IRC, developers are online 
primarily within European working time. You can of course grap me directly 
(kkoehne) if it's QML related). If you don't get any answer there, or the crash 
isn't easy to reproduce you shouldn't hesitate to file it to the bugtracker: 
http://bugreports.qt.nokia.com/browse/QTCREATORBUG . It's a bit more work for 
you, but it ensures that we don't forget about it, and you have a way to track 
progress :)

Thanks in advance,
Kai

> brett


--
Kai Koehne
Software Engineer
Nokia, Qt Development Frameworks

Nokia gate5 GmbH
Firmensitz: Invalidenstr. 117, 10115 Berlin, Germany
Registergericht: Amtsgericht Charlottenburg, Berlin: HRB 106443 B
Umsatzsteueridentifikationsnummer: DE 812 845 193
Geschäftsführer: Dr. Michael Halbherr, Karim Tähtivuori

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Tobias Hunger
Hey Bill!

On 25.03.2010 00:46, King Bill (Nokia-D-Qt/Brisbane) wrote:
> Bringing this over here as was requested ;)
> I was just criticized for criticizing the quality of creator ("on the
> basis of one bug report").

For reference: http://bugreports.qt.nokia.com/browse/QTCREATORBUG-807 is 
the bug I think Bill is talking about.

> I thought we were an open company, part of that is being open to
> criticism when things are not working.

I really do not understand this critique:
  * We did investigate the issue
  * We evaluated the patch
  * We asked you to move this discussion from an internal list to a 
public one.

So where exactly were we unresponsive or not open?

> Creator mainline is not working. I have reports both from internal and
> external users on a regular basis of either creator not building, or
> creator crashing upon startup.

Yes, there are issues with creator sometimes not building. Unfortunately 
that can happen in a master branch.

> I have pushed the externals to submit
> bugreports,

This is great, thanks! Please push the internals as well.

> but, again, my experience here has been less than glowing.

We are trying hard to keep the response time low. From my experience as 
an developer working on Qt creator my impression is that most bug 
reports are looked at in less than a day. We do not always update the 
reports, so this might not always be visible to you as the reporter. We 
might be able to improve here, but that is a issue of priority: Do we 
want to keep implementing new features or update the bugtracker?

> Being close to the front lines of the project, sometimes this can be missed.

We all use Qt Creator daily, so we tend to notice crashes and it not 
building. That is why they do get fixed pretty fast (at least in my 
experience).

> For qml usage, bleeding edge is the only choice currently, and the
> perceived quality of creator from the bleeding edgers is that
> creator's... not usable at all.

This does differ quite drastically from the experience we have here. I 
am not saying that you are not having this trouble, just that I do 
experience the quality of creator as very good overall.

So now we need to find out where our use cases differ and how we can 
improve the issue for your workload. I am afraid the best way to do this 
is by reporting bugs, which you unfortunately seem to not agree with. Do 
you have a suggestion on how we can improve the situation for you (and 
others in this thread) short-term?

> That is not the Qt way, and not a
> perception we should be having. Creator is an excellent product,

Thanks! Finally something we can agree upon;-)

> and one
> I use daily, and I'd like to go back to bleeding edge, as that's where
> all the cool new features are :)

I do understand your feelings and like the passion with which you are 
fighting for a better creator!

-- 
Tobias Hunger
Software Engineer
Nokia, Qt Development Frameworks

Nokia gate5 GmbH
Firmensitz: Invalidenstr. 117, 10115 Berlin, Germany
Registergericht: Amtsgericht Charlottenburg, Berlin: HRB 106443 B
Umsatzsteueridentifikationsnummer: DE 812 845 193
Geschäftsführer: Dr. Michael Halbherr, Karim Tähtivuori
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread Brett Morgan
Kai,

I'm compiling and using qt-creator in multiple environments, primarily OSX
10.6 and a variety of Linux distributions. I'm happy to give you details of
my environments and stack traces.

brett

On Thu, Mar 25, 2010 at 7:10 PM,  wrote:

> Hi,
>
> since this partly about Qml support which I'm personally involved in, I'd
> like to add my two cents :)
>
> > [...]
> > Creator mainline is not working. I have reports both from internal and
> > external users on a regular basis of either creator not building, or
> > creator crashing upon startup. I have pushed the externals to submit
> > bugreports, but, again, my experience here has been less than glowing.
> > Being close to the front lines of the project, sometimes this can be
> missed.
>
> It would be interesting to know which platform / compiler they are using.
> Cause I personally have been working with bleeding edge creator since the
> early beginning, and it was rare that I had to switch to a different version
> altogether. I'm working 8 hours a day with latest master. Sure, there are
> crashes from time to time, but that's probably to be expected with "bleeding
> edge", and we can normally sort them out rather quickly.
>
> > To fix these issues, can we implement some sort of staging/CI system
> > like we have for Qt?
>
> I don't see that a staging system would gain us much. First, the Qt
> staging/CI system is heavily based on autotests, but the creator autotest
> suite isn't up to really catch most issues.  I think we should have more
> autotests (you can always have more autotests!), but then again testing a UI
> program with good coverage is inherently difficult ... So, I don't think it
> would gain you that much in terms of stability. Second, the experiences we
> have with the Qt autotest system when it comes to integration delays aren't
> exactly encouraging ;) Right now we can quickly respond to a bug report with
> "fixed it, will be available to externals in 12 hours". This would change to
> "fixed it, might take weeks until it is be available to you". This is btw
> especially true for the Qml support, where we had to quickly react to
> changes in the QtDeclarative API in the past ...
>
> An IMO more viable alternative would be to have some sort of tested
> snapshots now and then.
>
> > For qml usage, bleeding edge is the only choice currently, and the
> > perceived quality of creator from the bleeding edgers is that
> > creator's... not usable at all.
>
> Well, you can stick to e.g. the Alpha, and even creator 1.3 has some
> limited QML support. The QML text editor + runtime infrastructure has been
> pretty stable for me anyhow. Regarding QmlDesigner - yeah, the quality is
> not yet up to where it should be, but then again it's still in the Alpha
> stages, and one part of the problem is also that we do have received rather
> minimal feedback so far. So please, people - if you find something, report
> it!
>
> > That is not the Qt way, and not a
> > perception we should be having. Creator is an excellent product, and one
> > I use daily, and I'd like to go back to bleeding edge, as that's where
> > all the cool new features are :)
>
> Understandable :) I think we have managed to get along pretty well so far
> with our very 'lean' development model, based on continuous feedback rather
> than any continuous integration system ... Now QtCreator is getting more and
> more features that we as core developers aren't using ourselves every day,
> so we will probably need a better QA. But I really don't want to see the
> momentum we have being killed by processes.
>
> Regards
>
> Kai Koehne
> ___
> Qt-creator mailing list
> Qt-creator@trolltech.com
> http://lists.trolltech.com/mailman/listinfo/qt-creator
>
> ___
> Qt-creator mailing list
> Qt-creator@trolltech.com
> http://lists.trolltech.com/mailman/listinfo/qt-creator
>



-- 
Brett Morgan
http://www.google.com/profiles/brett.morgan
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-25 Thread kai.koehne
Hi,

since this partly about Qml support which I'm personally involved in, I'd like 
to add my two cents :)

> [...]
> Creator mainline is not working. I have reports both from internal and
> external users on a regular basis of either creator not building, or
> creator crashing upon startup. I have pushed the externals to submit
> bugreports, but, again, my experience here has been less than glowing.
> Being close to the front lines of the project, sometimes this can be missed.

It would be interesting to know which platform / compiler they are using. Cause 
I personally have been working with bleeding edge creator since the early 
beginning, and it was rare that I had to switch to a different version 
altogether. I'm working 8 hours a day with latest master. Sure, there are 
crashes from time to time, but that's probably to be expected with "bleeding 
edge", and we can normally sort them out rather quickly.

> To fix these issues, can we implement some sort of staging/CI system
> like we have for Qt?

I don't see that a staging system would gain us much. First, the Qt staging/CI 
system is heavily based on autotests, but the creator autotest suite isn't up 
to really catch most issues.  I think we should have more autotests (you can 
always have more autotests!), but then again testing a UI program with good 
coverage is inherently difficult ... So, I don't think it would gain you that 
much in terms of stability. Second, the experiences we have with the Qt 
autotest system when it comes to integration delays aren't exactly encouraging 
;) Right now we can quickly respond to a bug report with "fixed it, will be 
available to externals in 12 hours". This would change to "fixed it, might take 
weeks until it is be available to you". This is btw especially true for the Qml 
support, where we had to quickly react to changes in the QtDeclarative API in 
the past ... 

An IMO more viable alternative would be to have some sort of tested snapshots 
now and then.

> For qml usage, bleeding edge is the only choice currently, and the
> perceived quality of creator from the bleeding edgers is that
> creator's... not usable at all.

Well, you can stick to e.g. the Alpha, and even creator 1.3 has some limited 
QML support. The QML text editor + runtime infrastructure has been pretty 
stable for me anyhow. Regarding QmlDesigner - yeah, the quality is not yet up 
to where it should be, but then again it's still in the Alpha stages, and one 
part of the problem is also that we do have received rather minimal feedback so 
far. So please, people - if you find something, report it! 

> That is not the Qt way, and not a
> perception we should be having. Creator is an excellent product, and one
> I use daily, and I'd like to go back to bleeding edge, as that's where
> all the cool new features are :)

Understandable :) I think we have managed to get along pretty well so far with 
our very 'lean' development model, based on continuous feedback rather than any 
continuous integration system ... Now QtCreator is getting more and more 
features that we as core developers aren't using ourselves every day, so we 
will probably need a better QA. But I really don't want to see the momentum we 
have being killed by processes.

Regards

Kai Koehne
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-24 Thread ext Raul Fernandes
Hi,

as user of different IDEs, I don't understand why qtcreator is trying to
provide more advanced features (such as QML support) if more basic features
are not implement/working properly. Important point: how can the team adopts
qtcreator as main development environment if it is not even getting more
stable? Is there any continuous integration environment which provides
nighly builds so we can see also testing results?

Regards, Raul.

On Thu, Mar 25, 2010 at 12:32 AM, Brett Morgan wrote:

> As a user in the field who is bringing a development team from web
> development across to embedded development using QML and Qt/Embedded I
> second Bill's call for qt-creator's git to be put under some form of
> automatic build testing. I need this stable so that my team will have
> confidence to move ahead on our project.
>
> brett
>
> On Thu, Mar 25, 2010 at 10:46 AM, Bill King  wrote:
>
>> Bringing this over here as was requested ;)
>>
>>
>> I was just criticized for criticizing the quality of creator ("on the
>> basis of one bug report").
>>
>> I thought we were an open company, part of that is being open to
>> criticism when things are not working.
>>
>> Creator mainline is not working. I have reports both from internal and
>> external users on a regular basis of either creator not building, or
>> creator crashing upon startup. I have pushed the externals to submit
>> bugreports, but, again, my experience here has been less than glowing.
>> Being close to the front lines of the project, sometimes this can be
>> missed.
>>
>> To fix these issues, can we implement some sort of staging/CI system
>> like we have for Qt?
>>
>> For qml usage, bleeding edge is the only choice currently, and the
>> perceived quality of creator from the bleeding edgers is that
>> creator's... not usable at all. That is not the Qt way, and not a
>> perception we should be having. Creator is an excellent product, and one
>> I use daily, and I'd like to go back to bleeding edge, as that's where
>> all the cool new features are :)
>>
>> --
>> Bill King, Software Engineer
>> Qt Development Frameworks, Nokia Pty Ltd
>> Brisbane Office
>>
>> ___
>> Qt-creator mailing list
>> Qt-creator@trolltech.com
>> http://lists.trolltech.com/mailman/listinfo/qt-creator
>>
>
>
>
> --
> Brett Morgan
> http://www.google.com/profiles/brett.morgan
>
> ___
> Qt-creator mailing list
> Qt-creator@trolltech.com
> http://lists.trolltech.com/mailman/listinfo/qt-creator
>
>


-- 
Raul Fernandes Herbster
Signove - http://www.signove.com
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


Re: [Qt-creator] Creator stability

2010-03-24 Thread Brett Morgan
As a user in the field who is bringing a development team from web
development across to embedded development using QML and Qt/Embedded I
second Bill's call for qt-creator's git to be put under some form of
automatic build testing. I need this stable so that my team will have
confidence to move ahead on our project.

brett

On Thu, Mar 25, 2010 at 10:46 AM, Bill King  wrote:

> Bringing this over here as was requested ;)
>
>
> I was just criticized for criticizing the quality of creator ("on the
> basis of one bug report").
>
> I thought we were an open company, part of that is being open to
> criticism when things are not working.
>
> Creator mainline is not working. I have reports both from internal and
> external users on a regular basis of either creator not building, or
> creator crashing upon startup. I have pushed the externals to submit
> bugreports, but, again, my experience here has been less than glowing.
> Being close to the front lines of the project, sometimes this can be
> missed.
>
> To fix these issues, can we implement some sort of staging/CI system
> like we have for Qt?
>
> For qml usage, bleeding edge is the only choice currently, and the
> perceived quality of creator from the bleeding edgers is that
> creator's... not usable at all. That is not the Qt way, and not a
> perception we should be having. Creator is an excellent product, and one
> I use daily, and I'd like to go back to bleeding edge, as that's where
> all the cool new features are :)
>
> --
> Bill King, Software Engineer
> Qt Development Frameworks, Nokia Pty Ltd
> Brisbane Office
>
> ___
> Qt-creator mailing list
> Qt-creator@trolltech.com
> http://lists.trolltech.com/mailman/listinfo/qt-creator
>



-- 
Brett Morgan
http://www.google.com/profiles/brett.morgan
___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator


[Qt-creator] Creator stability

2010-03-24 Thread Bill King
Bringing this over here as was requested ;)


I was just criticized for criticizing the quality of creator ("on the
basis of one bug report").

I thought we were an open company, part of that is being open to
criticism when things are not working.

Creator mainline is not working. I have reports both from internal and
external users on a regular basis of either creator not building, or
creator crashing upon startup. I have pushed the externals to submit
bugreports, but, again, my experience here has been less than glowing.
Being close to the front lines of the project, sometimes this can be missed.

To fix these issues, can we implement some sort of staging/CI system
like we have for Qt?

For qml usage, bleeding edge is the only choice currently, and the
perceived quality of creator from the bleeding edgers is that
creator's... not usable at all. That is not the Qt way, and not a
perception we should be having. Creator is an excellent product, and one
I use daily, and I'd like to go back to bleeding edge, as that's where
all the cool new features are :)

-- 
Bill King, Software Engineer
Qt Development Frameworks, Nokia Pty Ltd
Brisbane Office

___
Qt-creator mailing list
Qt-creator@trolltech.com
http://lists.trolltech.com/mailman/listinfo/qt-creator