[67cat] Results from NetBeans 6.7 Community Acceptance survey

[hjuskewycz]

The big headline is: NetBeans Community Approves NetBeans 6.7 for Prime Time Release



After I seeing the results (http://qa.netbeans.org/processes/cat/67/ca_results.html), I was wondering if this is the headline is for another survey ...



The results for the most important features are shocking! Nearly every 1/10 thinks that NetBeans isn't ready for prime time:



Editor Improvements: 10,4%

Performance Improvements: 9,7%

Overall Quality: 10,4%



And what about the rest? It's notable that the rest is better regarding the show stopper, but instead 40-50% don't use the feature.

I would interpret the survey that the new release isn't what the user wanted. They got features which they don't use and the already existing features aren't the quality they expected.



Please NetBeans team don't lie to yourself and to the readers, the survey is mixed at best, but doesn't clearly speak for NetBeans 6.7. If my post is too harsh, I am sorry. But this kind of marketing is more MS style, and I really don't like it.

[Matthies, Niklas]

Yeah, I rolled my eyes too at that headline. It is in line with the
general evangelist-ish tone of the NB newsletter though, so no surprise
there. I wish they would drop that and be more matter-of-fact.
Newsletter subscribers want information, not glamorization.

-- Niklas Matthies
 

> -----Original Message-----
> From: hjuskewycz [mailto:hjuskewycz@...]
> Sent: Wed, 24. Jun 2009 19:52
> To: netcat@...; netcat@...
> Subject: [67cat] Results from NetBeans 6.7 Community Acceptance survey
>
>
> The big headline is: NetBeans Community Approves NetBeans 6.7
> for Prime Time Release
>
>
>
> After I seeing the results
> (http://qa.netbeans.org/processes/cat/67/ca_results.html), I
> was wondering if this is the headline is for another survey ...
>
>
>
> The results for the most important features are shocking!
> Nearly every 1/10 thinks that NetBeans isn't ready for prime time:
>
>
>
> Editor Improvements: 10,4%
>
> Performance Improvements: 9,7%
>
> Overall Quality: 10,4%
>
>
>
> And what about the rest? It's notable that the rest is better
> regarding the show stopper, but instead 40-50% don't use the feature.
>
> I would interpret the survey that the new release isn't what
> the user wanted. They got features which they don't use and
> the already existing features aren't the quality they expected.
>
>
>
> Please NetBeans team don't lie to yourself and to the
> readers, the survey is mixed at best, but doesn't clearly
> speak for NetBeans 6.7. If my post is too harsh, I am sorry.
> But this kind of marketing is more MS style, and I really
> don't like it.
>
>
>
>
>
>

[Glenn Holmer]

On Wed, 2009-06-24 at 17:51 +0000, hjuskewycz wrote:
> the survey is mixed at best

I wasn't going to say anything, but I don't know if "79% of 144
respondents agree that NetBeans 6.7 Release Candidate is stable enough
to be shipped" is exactly a ringing endorsement (which RC, anyway?).

There have been dramatic improvements since RC1, and even my "pet peeve"
bug was just fixed in dev branch (I am hoping it will make it to 6.7):

http://www.netbeans.org/issues/show_bug.cgi?id=115033

Developers are doing the right thing by not releasing yet.  When we have
what we really think is the final RC, we should repeat the survey ("The
NetBeans developers have listened, and here is the result: now what do
you think?").  Then we will have higher numbers and it will look better
in the news.

--
____________________________________________________________
Glenn Holmer                          gholmer@...
Software Engineer                        phone: [phone]
Weyco Group, Inc.                          fax: [phone]

[Jiri Kovalsky]

Hello Helmut,

   yes, the news is true. The CA survey is part of every release and rules are publicly available in advance. Please read "Community Acceptance (CA) survey" section [1] in the NetCAT 6.7 Plan which clearly defines what to do with results from this research. In particular it states that if one issue is selected by more than 20% of respondents it becomes automatically a showstopper which must be fixed prior to releasing FCS. If this does not happen, we interpret it as an approval for the release. Can you share with us your own proposal what numbers are good/bad enough to release/not release? However, I hope you understand we obviously can't fix every bug that some person feels strongly about otherwise we could end up in never ending stabilization phase. :-)

[1] http://wiki.netbeans.org/NB67NetCAT#section-NB67NetCAT-CommunityAcceptanceCASurvey

   In addition to that after this CA survey we decided to contact all respondents who were very critical about performance problems or quality in general as a follow-up and you were one of these people. Why didn't you answer our e-mail?

Best regards,
-Jirka

hjuskewycz wrote:

> The big headline is: NetBeans Community Approves NetBeans 6.7 for Prime Time Release
>
> After I seeing the results (http://qa.netbeans.org/processes/cat/67/ca_results.html), I was wondering if this is the headline is for another survey ...
>
> The results for the most important features are shocking! Nearly every 1/10 thinks that NetBeans isn't ready for prime time:
>
> Editor Improvements: 10,4%
> Performance Improvements: 9,7%
> Overall Quality: 10,4%
>
> And what about the rest? It's notable that the rest is better regarding the show stopper, but instead 40-50% don't use the feature.
> I would interpret the survey that the new release isn't what the user wanted. They got features which they don't use and the already existing features aren't the quality they expected.
>
> Please NetBeans team don't lie to yourself and to the readers, the survey is mixed at best, but doesn't clearly speak for NetBeans 6.7. If my post is too harsh, I am sorry. But this kind of marketing is more MS style, and I really don't like it.

[nleck]

Come on guys. There a numerous people complaining about the project scanning issues.



Here is one movie (http://www.netbeans.org/nonav/issues/showattachment.cgi/83834/200906180201.ogg) showing the performance regression between 6.5 and 6.7



Just from today's nbusers emails


>

> I am trying disparately to be able to, once again, use a project which has had an unfixable "The project XXXXX has a broken platform reference. You have to fix the broken reference and invoke the action again." error.  While trying to use 6.7rc3, starting from scratch, I encountered a lock up of the ide while exiting. I've created the issue at

>

> <http://www.netbeans.org/issues/show_bug.cgi?id=167623>

>

> which illustrates the fact that the IDE, seemingly all over the place, is calling into the scanning/indexing code with the AWT-EDT and being blocked by bugs and implementation details of the prolonged locking associated with the cache implementation.

>

> Please vote for this one too so that we can, hopefully get enough documented issues to help the dev team understand the impact of this issue.

>

> Gregg Wonderly

>


and another


>

> Has anyone else noticed this on the recent builds:  I'm in the IDE and have a few source tabs in the editor with one active and the methods/fields showing in the Navigator.  I do something in the IDE outside the editor window (e.g. do a search or something) and then subsequently double-click on a method in the navigator that is supposed to take me to that method in the currently active editor window.  Only it doesn't.  Instead, the editor tab  seems to "wake up" and refresh the navigator window.   If I double-click on the method again, things work again.

>

> I can't reproduce this on demand - either it's occasional or is a sequence I haven't pinned down yet.  Because of that, I hesitate to open an issue.

>

> Thnx,

> tom

>

[rohitbrai]

+1

Yup!

I hope we fix the scanning and error badging issues before the release.

I saw the error badging bug marked as fixed, haven't got a chance to check it out yet. But that was definitely a show stopper and real productivity killer.

Netbeans lags, drastically in performance behind Eclipse and there is no comparison with IDEA. While I think being from Sun, Netbeans should be performance benchmark for all Java applications and not just IDEs.  Does anyone use Open Type/Open file features with large projects?

Though I won't consider performance a show stopper, considering the loads of new features we are releasing, we'll soon certainly be put to shame on that.

Doesn't everyone in the community feel that? 

Come on guys, stop lying to yourself and face the real thing. We need to improve, If the Netbeans team is too busy releasing features, may be we can pick some volunteers from the community, train them about the platform and the source and let them work on performance improvement. and these can in turn train others and actually build up netbeans as a community project than a pseudo open source project? Eclipse has become so strong owing to involving community in the entire development process and not only for testing. 

I know it is Open Source and we have the source and all that stuff, but it is not everyone's cup of tea to on there own go through the huge code base such as netbeans and moreover, there isn't much ROI in it. 

Simpler solution, will be switching to Eclipse, which I am inclined to more than ever, considering the Netbean's UML project going the scrap way, Visual Web being scrapped, we might soon loose the edge we have in tooling over the other IDEs. 

Yes I say WE, cause Netbeans loosing won't only be Netbean's team loss, but also loss of the community and each and every user who have so much invested in it in terms of knowledge and experience.

I agree with Jiri, that we could have raised the issues earlier, even I am at fault for putting in this mail so late, probably I should have raised it in the survey itself, but  if the user base at large has issues, I think it will be decent enough to respect opinion and release a throughly improved build rather than releasing a product for which the "Patch 1" is already defined. That really reminds me of MS Service Packs :D

The NetCAT and CA are not testers vs netbeans team game as we are making it to be here in this and few other mail chains, it is us uniting together with a common goal of "near perfect release". I hope everyone agrees to that.

Regards,

Rohit

The diamond cannot be polished without friction, nor man perfected without trials
http://mytechrantings.blogspot.com

On Thu, Jun 25, 2009 at 5:01 AM, nleck <nleck@...> wrote:

Come on guys. There a numerous people complaining about the project scanning issues.



Here is one movie (http://www.netbeans.org/nonav/issues/showattachment.cgi/83834/200906180201.ogg) showing the performance regression between 6.5 and 6.7



Just from today's nbusers emails


>

> I am trying disparately to be able to, once again, use a project which has had an unfixable "The project XXXXX has a broken platform reference. You have to fix the broken reference and invoke the action again." error.  While trying to use 6.7rc3, starting from scratch, I encountered a lock up of the ide while exiting. I've created the issue at

>

> <http://www.netbeans.org/issues/show_bug.cgi?id=167623>

>

> which illustrates the fact that the IDE, seemingly all over the place, is calling into the scanning/indexing code with the AWT-EDT and being blocked by bugs and implementation details of the prolonged locking associated with the cache implementation.

>

> Please vote for this one too so that we can, hopefully get enough documented issues to help the dev team understand the impact of this issue.

>

> Gregg Wonderly

>


and another


>

> Has anyone else noticed this on the recent builds:  I'm in the IDE and have a few source tabs in the editor with one active and the methods/fields showing in the Navigator.  I do something in the IDE outside the editor window (e.g. do a search or something) and then subsequently double-click on a method in the navigator that is supposed to take me to that method in the currently active editor window.  Only it doesn't.  Instead, the editor tab  seems to "wake up" and refresh the navigator window.   If I double-click on the method again, things work again.

>

> I can't reproduce this on demand - either it's occasional or is a sequence I haven't pinned down yet.  Because of that, I hesitate to open an issue.

>

> Thnx,

> tom

>