On Dec 21, 2009, at 6:28 PM, Adam Heath wrote:

> David E Jones wrote:
>> On Dec 21, 2009, at 2:04 AM, Adam Heath wrote:
>> 
>>> David E Jones wrote:
>>>> On Dec 21, 2009, at 1:49 AM, Adam Heath wrote:
>>>> 
>>>>> David E Jones wrote:
>>>>>> On Dec 21, 2009, at 1:35 AM, Adam Heath wrote:
>>>>>> 
>>>>>>> [email protected] wrote:
>>>>>>>> Author: hansbak
>>>>>>>> Date: Mon Dec 21 07:31:58 2009
>>>>>>>> New Revision: 892712
>>>>>>>> 
>>>>>>>> URL: http://svn.apache.org/viewvc?rev=892712&view=rev
>>>>>>>> Log:
>>>>>>>> Upgrade Axis1 to Axis2. Ofbiz now supports complex parameters in 
>>>>>>>> webservices including WSDL generation. see OFBIZ-3363 for more info. A 
>>>>>>>> contribution of Antwebsystems employee Chatree
>>>>>>>> 
>>>>>>>> Added:
>>>>>>>> ofbiz/trunk/framework/service/lib/XmlSchema-1.4.3.jar   (with props)
>>>>>>>> ofbiz/trunk/framework/service/lib/axiom-api-1.2.8.jar   (with props)
>>>>>>>> ofbiz/trunk/framework/service/lib/axiom-impl-1.2.8.jar   (with props)
>>>>>>>> ofbiz/trunk/framework/service/lib/axis2-kernel-1.5.1.jar   (with props)
>>>>>>>> ofbiz/trunk/framework/service/lib/axis2-transport-http-1.5.1.jar   
>>>>>>>> (with props)
>>>>>>>> ofbiz/trunk/framework/service/lib/axis2-transport-local-1.5.1.jar   
>>>>>>>> (with props)
>>>>>>>> ofbiz/trunk/framework/service/lib/commons-httpclient-3.1.jar   (with 
>>>>>>>> props)
>>>>>>>> ofbiz/trunk/framework/service/lib/neethi-2.0.4.jar   (with props)
>>>>>>>> ofbiz/trunk/framework/service/src/org/ofbiz/service/test/ServiceSOAPTests.java
>>>>>>>>    (with props)
>>>>>>>> Modified:
>>>>>>>> ofbiz/trunk/LICENSE
>>>>>>>> ofbiz/trunk/framework/common/servicedef/services_test.xml
>>>>>>>> ofbiz/trunk/framework/common/src/org/ofbiz/common/CommonServices.java
>>>>>>>> ofbiz/trunk/framework/service/src/org/ofbiz/service/ModelParam.java
>>>>>>>> ofbiz/trunk/framework/service/src/org/ofbiz/service/ModelService.java
>>>>>>>> ofbiz/trunk/framework/service/src/org/ofbiz/service/engine/SOAPClientEngine.java
>>>>>>>> ofbiz/trunk/framework/service/testdef/servicetests.xml
>>>>>>>> ofbiz/trunk/framework/webapp/src/org/ofbiz/webapp/event/SOAPEventHandler.java
>>>>>>> What about .classpath?
>>>>>> What about it? We can't assume everyone uses Eclipse...
>>>>> What does using eclipse have to do with updating that file?  I don't
>>>>> use eclipse, but I still update .classpath.  I also don't use ant.bat,
>>>>> yet I just updated that.
>>>>> 
>>>>> If you know that you have to keep .classpath updated to make eclipse
>>>>> work, and you knowingly check in new libraries, or remove old ones,
>>>>> then why make eclipse broken?  It's really not that hard to change.
>>>>> 
>>>>> Honestly, I don't understand why you would think not keeping it
>>>>> uptodate is a good thing.
>>>>> 
>>>>> ... perplexed ...
>>>> Where did I write "not keeping it uptodate is a good thing?" And if I 
>>>> didn't write it, how can you assert I was thinking it?
>>> How else could what you wrote be interpreted?  Because you responded,
>>> it seemed to imply that you disagreed that it wasn't important to keep
>>> it uptodate.
>>> 
>>> If you actually meant something else, then you should have said so.
>>> 
>>> My 3 words meant that it should have been updated.  What did you
>>> actually mean?
>>> 
>>> I still can't see how your response could be interpreted in any other way.
>>> 
>>> And I'm rather perplexed at having to explain my 3 simple words in
>>> such detail, esp. to you, David.
>> 
>> What does it have to do with your 3 words?
>> 
>> What did I mean? I meant:
>> 
>> What about .classpath? If Hans doesn't use Eclipse, how can you expect him to
>> keep that Eclipse-specific file updated, especially since it has been
>> historically maintained by Eclipse users. Some things, like the
> LICENSE and
>> NOTICE files that Jacques pointed out, really are responsibilities
> of committers.
>> I wouldn't consider keeping the Eclipse .classpath file one of those
> responsibilities.
> 
> We are all in this together.  The more people who keep an eye on the
> various parts of the project, and the more people who keep things
> working, the better.
> 
> There are some things that are *very* easy to check.  .classpath is
> most definately one of those.  Keeping it in sync is so very very very
> simple, why shouldn't it be kept in sync when library changes occur?
> 
> At what level should due diligence be thrown out the window?
> 
> When the default memory requirements for ofbiz change, should only the
> shell scripts be changed, only the .bat files, or both?  What if new
> system properties need to be set?  Should that only be done in one set
> of files?
> 
> Or, for another example, what about keeping the embedded ant working,
> vs. an external, system installed ant?
> 
> I completely understand how some people may not use some small part of
> ofbiz; there are so many parts, it can be hard to keep them all in
> working order.  However, some of those small parts are at least easy
> to work with, and simple enough to see obviously broken behaviour,
> that updating them when other parts change is easy.
> 
> I really still can't believe that you are suggesting that keeping this
> very simple to understand text file uptodate is not a worthwhile goal
> for everyone.

Don't be ridiculous, I never wrote not implied any such thing. You really 
should check your assumptions and re-read what I did actually write.

And here I go, shouting into the wind again... I may be an idiot, but I'll try 
not to prove myself insane by writing more (even if I think that definition of 
insanity is BS).

-David


> 
>> Is that clear enough?
> 
> Clear, yes.  However, it still doesn't make any sense to me.
> 
>> Shall I go on? Why are you trying to imply that Hans should do this, as if 
>> he's
>> made some sort of mistake and it's his responsibility to correct it?
> Why didn't
>> you just correct it yourself if you care so much about it? Who says the
>> responsibility belongs to one more than another, and if it doesn't
> why are you
>> trying to get someone else to do something by implying it is their
> responsibility
>> instead of working with them in a more friendly and respectful way?
> 
> I didn't correct it myself, because more eyes/hands makes for a better
> project.  The more people we have looking over the system, the better
> the system will be.
> 
> Of course I could have correct this issue.  I could have even done it
> silently.  But then no one would learn, and no one would become better.
> 
> Besides, the best practices for contributors(1) says: first, do no harm.
> 
> I can understand when mistakes get made.  We are all human, none of us
> are perfect.  However, when a mistake is eventually made, if it is
> never pointed out, how can you expect the person who made the mistake
> to get better?  I don't point out issues to make people feel bad; I
> point them out to make them better.
> 
> I've said this before, I don't consider the background of a person
> when I read a commit message.  I just comment on that particular commit.
> 
>> Okay, I think that's about all I can figure out to say on this little 
>> topic...
>> if you need more clarification and expansion I'll try to put
> together a more
>> general philosophical background. On the other hand, I've written a
> few blog
>> entries that are related to this and that I think explain it rather
> well.
> 
> Which blog entries have you written?  Links would be nice.
> 
> 1:
> http://cwiki.apache.org/confluence/display/OFBADMIN/OFBiz+Contributors+Best+Practices
> 
> 

Reply via email to