Re: reiser4 inclusion?

[Christian Stroetmann OntoLab <stroetmann@ontolab.com>]

Dear Alli;
> On Sat, Apr 25, 2009 at 10:15 PM, Christian Stroetmann OntoLab
> <stroetmann@ontolab.com> wrote:
>  
>> Dear Alli,
>>    
>>> Well, as you noted at the end of your reply, it is polite to reply to
>>> answers. Maybe you can be the one to start.
>>>
>>>       
>> What am I doing since months?
>>     
> You didn't reply to the questions for the first time and some things
> are still not clear to me.
Please, that this is a thread. So I won't repeat again and again what 
was said before in this mailing-list.
>  The questions were simple
I gave answers.
>  and yet you for
> some reason seem to hide something.
>   
???
>>> On Sat, Apr 25, 2009 at 6:26 PM, Christian Stroetmann OntoLab
>>> <stroetmann@ontolab.com> wrote:
>>>
>>>      
>>>> Hello Alli,:
>>>>
>>>>        
>>>>> Hello,
>>>>> I don't know the background of OntoLabs
>>>>>
>>>>>           
>>>> OntoLab. There can only be one. ;)
>>>>
>>>>        
>>>>>  and even when I googled and
>>>>> read something in teh archives, I haven't learned much. I would like
>>>>> to ask a few questions.
>>>>> 1. @Christian: have you done any actual development
>>>>>
>>>>>           
>>>> I did a further development, which is based on R4 but transforms it to
>>>> something different/new. Besides some other issue this new storage 
>>>> system
>>>> was given a new name, a process which is common in the open source 
>>>> area
>>>> (examples can be given on request).
>>>>
>>>>         
>>> It is common to give it a completly different name in case there are
>>> some legal issues, for example Gaim -> Pidgin (AIM legal problem).
>>>       
>> Okay
>>    
>>>  It
>>> is common to give a name back-referencing the original project you are
>>> forking, for example Go-OO (enhanced OpenOffice.org which tries to
>>> cooperate with the original project).
>>>       
>> Okay
>>    
>>>  But I'm sure that you can give
>>> examples of forks that chose distinct names to tell the two projects
>>> apart (Mambo - Joomla).
>>>       
>> Is it a typo: Can give or can't give?
>>     
> Can give.
>  
>>>  I would not say that one or another custom is
>>> more common.
>>>
>>>       
>> I don't understand this remark. All your possibilies you enumerated 
>> does say
>> the same: It is common for a reason to give something a new name. And 
>> my for
>> simplicity of discussion not named real examples are a database named 
>> X as
>> the foundation and the same database with transaction enhancement but 
>> named
>> Y. Than the same database X and another version of this database with 
>> other
>> goodies now named Z. A one liner on the websites of Y and Z, which is
>> referencing database X, and ready. A real world example (I can look 
>> after
>> the real names in my archive, if you really need to know them).
>>     
> Usually you choose a familiar name to give credit to the original
> project and a distinct name to distinguish yourself from the original
> project, because you're heading in a different direction.
That's your opinion, which I can't follow.
>  You took the
> very same code,
Yes, that's right.  And I explained in this thread why.
>  did nothing,
No, that's wrong. The development of software consists of different 
phases and differnet works. I think I did much and explained why there 
was no patch or code delievered until today.
>  but chose a completely different name
> because of an event concerning only the main developer, not the
> project itself.
No, that's wrong and I can't understand why you bring it again. I 
answered this subject in my last e-mail (its written some lines below 
"50 to 50").
>  Strange. But I don't really care that much for the
> name, even though I think it signifies the real problem.
>   
Yes, and again: I explained it in the e-mail before (its written some 
lines below "50 to 50"). But I also offered in the past the R4 naming 
compromise. So, now I do ask: Where are the steps in my direction?
>>>>>  or have you just
>>>>> renamed the filesystem, created a webpage with feature highlight and
>>>>> gave no credit to original developers except for NOT explicitly
>>>>> renaming the files you are mirroring? In the text there is no mention
>>>>> of Reiser4 even though it in fact is Reiser4. Why?
>>>>>
>>>>>
>>>>>           
>>>> See above. And it was discussed in mailing-lists to rename it due to a
>>>> special issue around a person, which let all main sponsors to jump 
>>>> of the
>>>> bandwagon. In this respect the new name was also meant to restore 
>>>> again
>>>> the
>>>> acceptance for the foundational filesystem.
>>>>
>>>>         
>>> I think that the "special issue" is not exactly what caused somebody
>>> to leave;
>>>       
>> The reaction to leave were a more than clear answer.
>>    
>>>  I still use reiser4 simply because it is a very good
>>> filesystem.
>>>       
>> Yes, okay. But the opinions in this case were 50 to 50. You say, that 
>> you
>> belong to the group of persons, who are interested in the filesystem. I
>> belong to the other group, and for us the context and the persons 
>> around a
>> project are important concerns. Only a short reminder: The "special 
>> issue"
>> was not stealing some candies.
>>     
> Yes, and it had nothing to do with the filesystem. Just a short reminder.
>   
But with the development, and that's enough. A would like at this point 
also to repeat: You have this opinion, and I have the other position. 
This leads again to my question: Where do get a compromise?
>>>  Also, you didn't say what development other then renaming
>>> and setting up a webpage and stuff have you done.
>>>       
>> Sorry, but this is wrong. I do understand that my project is highly 
>> complex,
>> but experts understand directly the genius solution, eg. universities in
>> Germany have arranged the education in Computer Sciences around 
>> nearly all
>> of the mentioned themes/items on the OntoLinux webpage. I think it 
>> will take
>> 10 to 15 years to see for persons not deeply involved in these areas of
>> Informatics how elegant and progressive my development really is.
>>     
> I don't care all that much for your project (sorry), when I browsed
> through the webpage (software/hardware) I saw links to other projects
> (T2, LFS, Gentoo, other Linux distributions, ..., hardware platforms,
> ...) but I didn't get the main idea, what is it all about. Is it some
> unification project?
Yes, it is an integrating distribution.
>  Is it a distribution? I guess this is beyond this
> mailing list.
Okay
>  BUT - I asked about the filesystem. I don't need to
> understand the whole project to understand if you have done some
> development on the filesystem or not. So have you? And what
> development have you done?
>   
I answered it in the e-mails before and clearified it in this e-mail.
>>>  Sure, you can say it
>>> because you didn't want someone to steal from you. But THAT is
>>> definitely not a common opensource process.
>>>       
>> Yes, because it is not a common solution.
>>     
> It is a filesystem.
No, it is more an integrating system with a Semantic Storage.
>  There are many other filesystems. many of them are
> targeted for specific purposes.
Yes
>  many of them do what reiser4 was doing
>   
No, that's wrong. There is only R4. The other storage systems, which are 
comparable, are based on a database (that means a propritary 
filestructure) or are not really functioning.
> and none of the opensource behaves like that.
That's no argument. See it as something new.
>  See btrfs wiki - it is
> an open project with ideas publicly available online so that everyone
> can steal from them. Why can't you do this?
>   
No one steals from eg. btrfs. Someone takes from a project items and the 
source project is referenced. But this only happens if a project is 
known to a wider audience. And it's much of work to explain it, to 
discuss it with other persons, and as I explained experts do understand 
my solution.
>>>  And there is no original
>>> code I could download from you.
>>>
>>>       
>> Yes, but only until it is better known and accepted. After these the bad
>> guys can't steal anymore.
>>     
> How can it be better known and accepted when it is not available?
>   
See it as the common process of publicating a whitepaper. There is now 
the general/theoretical concept publicated.
> Also, I take this as a "no", because really, what good it is to the
> reiser4 project when you don't give the changes back?
Please, read above
>  BTW aren't you
> obligated by the GPL licence? I'm not really sure about this but AFAIK
> that is how it works.
>   
Yes, and guess what I have clearified it on the OntoLinux website. Also, 
this isn't a point of discussion due to the clearness of the subject. 
Btw.: The GPL says not that I have to make my source public. It says, if 
I publicate it, then I have to give others the rights following the GPL.
>>>>> 2. @Christian: What did you have in mind when you set the page up?
>>>>>
>>>>>
>>>>>           
>>>> See above. And at that time there was no project page for R4. As I
>>>> mentioned
>>>> also this happened before the source code was archieved at kernel.org.
>>>>
>>>>         
>>> Before reiser4 patches were available from kernel.org they were
>>> available from namesys.
>>>       
>> Yes, as long as the company namesys was running.
>>    
>>>  Then there was time when it was available from
>>> both sources and then namesys went down.
>>>       
>> No, AFAIK that is wrong.
>>     
> I was using reiser4 for quite a long time and believe me, I would have
> stopped if it wasn't available online. But there is no way of proving
> this is true or that it is not.
>   
At that time I did what I did the webpage of namesys was gone and there 
was no code on kernel.org. Otherwise I wouldn't have done the mirroring.
>>>  So you set this up at the
>>> time when namesys webpage (and company) was still functional? This
>>> makes me wonder even more why would you set up your own page when
>>> there was a company developing the filesystem.
>>>
>>>       
>> Please see above.
>>    
>>>>> What features were you going to work on if Edward cooperated with you
>>>>>
>>>>>
>>>>>           
>>>> It's over. I only take part in the discussion, because I would like 
>>>> other
>>>> people to know my motivations.
>>>>
>>>>         
>>> Well, that might be why I put in the past tense. So please answer me,
>>> what features were you going to work on?
>>>
>>>       
>> General functionality and common development, the hardlink problem 
>> (which is
>> after Al Viro one of the NO MAINLINE FOR REISER4 problems), a potential
>> solution for the cyclic dependance problem, some ideas for the plug-in
>> architecture problem (which is also a NO MAINLINE FOR REISER4 
>> problem), the
>> described Semantic (World Wide) Web transformation, and much more.
>>     
> Some of those were bug fixes. I believe you had to deal with them to
> use the filesystem for your own project.
Yes
>  Did you send the patches back
> to this ML?
No, because for my work I don't need patches. I only need a 
bootstrapping system. That means it doesn't matter if I use software, 
which is 2 years old or brandnew.
>  I really don't know because I signed to the ML only a few
> months ago, that's why I'm asking.
>   
I don't want to offend you if I say that I got this impression some 
e-mails before.
>>>>> (as you say he refused to; I don't know the history so please excuse
>>>>> me everybody)?
>>>>>
>>>>>           
>>>> In another thread I publicated the e-mail, which was sent by me to the
>>>> maintainer. I got no answer until today. Than Jason asked for the 
>>>> actual
>>>> maintainer and mentioned that he also send an e-mail. Later we 
>>>> found out
>>>> that a maintainer for R4 really exists.
>>>>
>>>>        
>>>>>  For instance there are three things I'm missing in
>>>>> reiser4: defragmentation tool, resize tool, xattrs. I haven't learned
>>>>> from the text what YOU wanted to work on.
>>>>>
>>>>>
>>>>>           
>>>> Yes, I described it only so far, that the open source community has no
>>>> problems with patents by the large companies. The details of my 
>>>> work are
>>>> described between the lines. I'm sorry for not giving all details or a
>>>> better description, but there are some really bad guys in th IT-circus
>>>> running around.
>>>>
>>>>         
>>> I really don't know about any other project that would use this
>>> process and it looks cheesy.
>>>       
>> Sorry, that is a personal problem. See also above.
>>    
>>>  And I don't know about patents,
>>>       
>> Sorry, but you don't have insight into this technological area. If I 
>> tell
>> you that the situation is like I say, than you should trust me.
>>     
> Most people don't accept this as an argument and neither do I.
Okay, I do know this common situation.
>  If your
> reasons are real, then you are can tell me. Even say something you
> think I wouldn't understand. I'm not dumb, I can read and use google.
>   
Okay, but at this point I have to say that you need more than 15 years 
of learning, researching, and googling to get the needed knowledge for 
understanding and following discussion. Besides this, you should find 
some really good arguments, why I should write endless e-mails. What you 
want is the following: You want to drive a car, but before you start the 
engine Albert Einstein should explain why the atoms of the chassis hold 
together. You are not dumb, I'm sure you understand maybe everything, 
but we haven't the time. So you can trust me or in another field like 
the Linux kernel other persons like I do.
>>>  for
>>> example the btrfs filesystem is implementing a lot of features used in
>>> commercial filesystems and they don't seem to have any problems.
>>>       
>> Oh, please, there are many reasons for this, eg. old patents, no 
>> patents,
>> open source, bought patents by the opensource community, and so on. 
>> To your
>> btrfs example: I'm sure that Oracle has thoroughly looked at potential
>> patent problems.
>>     
> I don't get your point. If you have problems because patents do not
> exist yet and you've had a freat idea, you can publish it or patent
> it, simply in some way make it known that it is your idea. No problem
> in telling us in this case. If there are patents that you would like
> to use then you have to deal with it, but it is still nothing to
> prevent you from telling us. So what is the problem?
The patent things were said in conjunction with your example filesystem 
btrfs.
> Again, you didn't answer when I asked you to be specific.
I answered this in the e-mail before.
>  Instead you are suggesting I
> wouldn't understand. 
No
> Maybe I wouldn't, but first you have to give me a
> chance.
>   
Yes, and I repeat: You have to know at least everything to the bare bone 
that is listed in the software and hadware list on the OntoLinux website.
>>>  So
>>> could you please be more specific, either in explaining the features
>>> or the reasons why you don't want to/can't publush them?
>>>
>>>       
>> See above. I do know who is taking everything from other projects and is
>> using it for its business without referencing the sources (no names 
>> will be
>> given here).
>>     
> You can send me a personal email, I'm curious. Still you are using
> code that was largely connected with the person of Hans Reiser
No, there were other persons, companies, and institutes also in the game.
>  and it
> is rude and impolite not to give him credit for this. Maybe everything
> you're doing is legal, but it doesn't mean it is not rude.
>   
???
>>>>> 3. Everybody else: Could you please point me somewhere I can read
>>>>> about this whole history?
>>>>>
>>>>>           
>>>> The history of R4 or this special thread?
>>>>
>>>>         
>>> The history of reiser4, you, OntoLab. I just want a link to where this
>>> problem started.
>>>
>>>       
>> The history, todo-lists, manuals, helping documents, and the other 
>> items of
>> R4 are scattered in the internet. I wanted to collect them on the 
>> proposed
>> project page.
>> My opinions are given in this mailing-list.
>> OntoLab is the name of the laboratory. The project is named 
>> OntoLinux. The
>> storage system can be found in the Components section. Also, you must 
>> be at
>> least familiar to the bare bone with everything linked in the 
>> Software and
>> Hardware lists.
>>     
> OK, thanks.
>  
>>>>>  I believe that Edward and others have good
>>>>> reasons to act the way they do (and I think they have the full right
>>>>> to), but I would really like to read on this.
>>>>>
>>>>>
>>>>>           
>>>> No, I have the opinion that R4 is a GPL'ed code and that the open 
>>>> source
>>>> development community, especially around Linux, has established some
>>>> other
>>>> kinds of "rules" to work together. One of this rule is to answer
>>>> questions
>>>> or to say her/his opinion directly to someone else. But this is in 
>>>> fact
>>>> one
>>>> of the points in discussion too.
>>>>
>>>>         
>>> Sure. But from what I've seen it looks like they did in the past and
>>> don't consider it necessary now, because repeating the same thing over
>>> again would be a waste of time. That's why I want to read the first
>>> mails about this.
>>>
>>>       
>> Have fun searching, finding, reading, and understanding. It took me some
>> portions of precious time.
>>    
>>>>> Hope I didn't offended anybody and BTW it would actually help to have
>>>>> a stable hosting for wiki/trac or something that like system and a
>>>>> centralized place for the tools, it's manuals, some howtos, roadmap,
>>>>> todo and this kind of stuff. kernel.org is not exactly the best place
>>>>> for this AFAIK. Just my two cents.
>>>>>
>>>>>
>>>>>           
>>>> Yes, that is exactly the point. And there are so many other unsolved
>>>> items.
>>>> Last but not least I do repeat a question a nice person asked me: How
>>>> want
>>>> one developer alone manage the code? (It doesn't matter if there 
>>>> are in
>>>> reality 5 or 10 developers.)
>>>>
>>>>         
>>> I was talking about setting a project page.
>>>       
>> Please, read the mailing-list first.
>>    
>>>  You were setting up a
>>> project page in a context of a completely different company (probably
>>> owned by you) with a different name, with no useful information on
>>> future development (just a load of "it will be great" text). There is
>>> a difference between the two.
>>>
>>>       
>> Please, read the mailing-list first. All different versions were 
>> proposed.
>> And I really do apologize for repeating, but: We are running in circles
>> since 2 or 3 years. :-D
>>    
>>> Have a nice day,
>>> al-Quaknaa
>>>
>>>       
>> Have fun
>> Christian *<:o)
>>
>>     
>
> al-Quaknaa
> -- 
> To unsubscribe from this list: send the line "unsubscribe 
> reiserfs-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>
>   
Sincerely
Christian *<:o)