[Insight-users] Kd-tree implementation in ITK is buggy and unreliable

[Ali -]

Luca,

The answer is clear, I am here as a 'user' focusing on a subject that has almost nothing to do with kd-tree search algorithm. Also, I do use other open source libraries, and, given each library having its own bugs, if I want to volunteer to save the world, my own project will never get done.

In this specific case, the problem has been reported previously together with examples of reconstructing the bug.


-Ali

> 
> Ali,
>   why don't you volunteer and try to fix the bug(s) you found? (or  
> help in that direction by providing a thorough report of what happens  
> in your case?)
> I'm not being provocative, I really mean it.
> 
> Luca
> 
> 
> 
> On Apr 22, 2008, at 5:51 PM, Ali - wrote:
> 
>>
>> Hi Luis,
>>
>> I was surprised how you found out the anger between the lines as I  
>> did my best to filter it out and put it politically right. It's  
>> good to see there are some high-tolerance in some open source  
>> communities these days. In many cases, a micro-critic can kick you  
>> out of the list quickly -- it shouldn't be that way, but what's the  
>> correlation between open source and democracy anyway? Many  
>> developers feel like a king once they release a something.
>>
>> Normally, when there is a consistent bug in your code which uses  
>> other well established libraries, the last place that you want to  
>> look at the somewhere 'out' of your code where you assume that the  
>> two factors of time and number of developers-users has hygienically  
>> debugged the code. Due to the first factor, this anticipation  
>> increases when that feature has been around for a long time.
>>
>> Hope to see the the clean code soon.
>>
>> -Ali
>>
>>>
>>>
>>> Hi Ali,
>>>
>>>
>>>              Thanks for your angry email !
>>>
>>>
>>> There is nothing better than an energetic rage-saturated
>>> message for waking us up on a Tuesday morning !
>>>
>>>
>>> (Today is Tuesday, isn't it ?)
>>>
>>>
>>> To start with, I had to get more coffee to complete the
>>> calculation that half a decade is actually five years...
>>>
>>> You were worrying me at first, but now I have to admit that
>>> reporting bug's lifespan in units of decades is definitely
>>> an effective scare tactic to call attention to the problem.
>>> That was a very clever choice of words!.
>>>
>>>
>>> You see, at that lifespan scale we will have to stop calling
>>> them bugs, and refer to them as cicadas.   :-)
>>>
>>> Ok... ( geek joke == bad joke ).
>>>
>>>
>>>
>>> ...Anyways...
>>>
>>>
>>> I share your philosophical conundrum:
>>>
>>>
>>>        "Why is that this bug has not been fixed ?"
>>>
>>>
>>> It is indeed a deep philosophical question worthy of reflexion
>>> and consideration.
>>>
>>>
>>> After meditating over this difficult issue, and consulting
>>> several experts on socio-economical matters, a couple of
>>> foreign policy observers, two Buddhist priests and a Lama
>>> I have arrived to the unavoidable answer:
>>>
>>>
>>>                        Destiny !
>>>
>>>
>>> The solution of this bug has been retained by the Gods in
>>> anticipation for the events that will trigger the prophetic
>>> sequence of actions that will result in the modification
>>> of the source code used in the Kd-Tree.
>>>
>>>
>>> The event that we all have been prophetically waiting for,
>>> is without a doubt:
>>>
>>>                       Your email !
>>>
>>>
>>> Angry users are the fuel that propels the progress of software
>>> engineering. Without angry users, we will all be using Microsoft
>>> products. Without angry-user-attitude Richard Stallman would
>>> have not been pissed-off with that Xerox printer and its
>>> closed-source drivers, and would have never started the
>>> Free Software movement.
>>> [http://www.oreilly.com/openbook/freedom/ch02.html]
>>>
>>>
>>>
>>>               In this Tuesday morning,
>>>               you have showed us the Way !
>>>
>>>
>>> Angry and vocal users are the strength of Democracy and Freedom.
>>> Whe users say:  "Enough is enough!", then developers will listen.
>>> and if developers don't listen, users are still empowered to fix
>>> the problems on their own. After all, that's the ultimate power
>>> of Open Source.
>>>
>>>
>>> We must celebrate today your epic expression of dissatisfaction,
>>> and respond to it by letting the prophecy unchain the actions
>>> that will modify the source code of the Kd-tree.
>>>
>>>
>>>
>>> We need more users with your angry attitude,
>>> because otherwise:
>>>
>>>
>>>              "When users go silent,
>>>               software goes bad."
>>>
>>>
>>>
>>>
>>>
>>>        Regards,
>>>
>>>
>>>            Luis
>>>
>>>
>>>
>>> --------------
>>> Ali - wrote:
>>>> Hi,
>>>>
>>>> A quick search in the mailing list shows that, during the past  
>>>> half a decade, it has been reported many times that the kd-tree  
>>>> classes in ITK are buggy. Writing some library based on ITK, it  
>>>> took me a few days to find out that the bug in my library is  
>>>> actually introduced by the kd-tree implementation in ITK which  
>>>> FINDS THE WRONG NEAREST NEIGHBOUR.
>>>>
>>>> I have no idea, against many warnings from the users, why the bug  
>>>> has not been resolved yet. In the case it is difficult to address  
>>>> the bug, one option is to wrap many other existing  
>>>> implementations, personally I switched to libkdtree++.
>>>> _______________________________________________
>>>> Insight-users mailing list
>>>> Insight-users at itk.org
>>>> http://www.itk.org/mailman/listinfo/insight-users
>>>>
>>>
>>>
>>>
>>
>> _________________________________________________________________
>> Search and win with BigSnapSearch.com
>> http://www.bigsnapsearch.com
>> _______________________________________________
>> Insight-users mailing list
>> Insight-users at itk.org
>> http://www.itk.org/mailman/listinfo/insight-users
> 

_________________________________________________________________
Be a superhero and win! Play the Iron Man Mashup Game 
http://www.ironmanmashup.co.uk