On 7 Jun 2011, at 11:10, Henry Story wrote:
> On 7 Jun 2011, at 09:03, Reto Bachmann-Gmuer wrote:
>> 
>> To be fair I should also mention that you did a great job spreading the word
>> about clerezza and presenting clerezza on various occasions.

yes, but I was not just spreading the word about Clerezza because of what it 
is, but because of what it could be. All my commits were geared to putting an 
initial viral application of ZZ - a prototype Social Web application - in place 
to be able to get people interested in contributing to the project so that it 
could move along faster. 

It is a real pity you did not come to Berlin, because you would have seen how 
dynamic and coordinated the Open Social people are and how far Diasopora has 
progressed since I met you to help implement WebId in Clerezza last summer. 
Diaspora have the advantage of working with Ruby-on-Rails which is very stable 
and comes with a lot of bells and whistles.  But they did not know anything 
about linked data! And that is our big advantage. 

The Foaf-WebId community was very well represented. We had a few workshop with 
20-30 people attending (large for this conf). 

But there was a lot less I could demo in Clerezza than I had hoped to be able 
to. Still I would not have been able to implement even that much had we been 
going all formal since the beginning of when I joined. I understood that the 
main aim of ZZ project at Apache was to get new users, and so to that effect I 
developed WebID and the social aspect without trying to close every issue.  
There were 3 elements to getting a viral part:

  - WebID authentication
  - friend support
  - pinging new friends (to get the system to grow)

  - clearly it also needs rule based access control

Each of these depend somewhat on the other, and there are 1000s of ways to 
implement them. What is important is to implement something to demo, then one 
can refactor later.

> [snip]
> First we had your commits breaking existing functionality. Your argument was
> that the existing security infrastructure is fundamentally wrong and the
> broken functionality will be provided in future by some new means.

There were 2 issues that broke some things. 

In the first case to do with the Subject refactoring a month or so ago, you 
immediately insisted that I had to revert all code I had done. Then I fixed it 
within a few hours. Bugs happen.

The second breaking was related to your latest -1. The only functionality that 
was broken was that of current WebID users, so not a major breakage in any way 
(affected people: 2). One could even think of it as discovering how existing 
authentication systems are broken. Anyway I removed the code, though it is 
difficult to fix interesting things like that in small patches as you wish.

>> But for what
>> the development process is concerned I really don't see reasons to be happy
>> about the "free reviewing" but rather to be annoyed of you blocking progress
>> and having a quite poor ratio between actually closed issues and restraining
>> others (by committing breaking code and vetoing changes)
> 
> Most of any attempts I have made at closing an issue have been blocked by 
> you. True my
> code is not perfect, but neither is yours. And that is a little bit why I am 
> giving you
> a bit of heat here.
> 
> Your criticism of me seems to be that I have contributed a lot, but that you 
> have refused
> to accept any of the code.

This reminds me for example of the WebProxy code. You could have closed my 
issue, then
opened a new one to refactor WebProxy to the way you did. No problem really. 
But instead
you don't allow me to close an issue, refactor the contribution I made, and 
then say I don't
participate. 

And then when I told you that your refactoring broke the code I had built on 
the old WebProxy, 
you had the arrogance to  tell me that the issue I had had not been closed, 
this was not a problem 
for you. All of this just as I was trying to get ready to improve a bunch of 
code for the Social Web conference, in order to be able to make a bigger impact 
here. 


> I will spend today to work through your code and review it for real. Then 
> I'll come
> to a decision on the +1/-1 issue, and I will argue them very carefully.

I will still do that, because I was not just doing tit for tat slow down 
recently when
bringing up these issues. I was also trying to make sure that architecturally 
things are
clean so that we can build apps efficiently on our base. And there is a code 
smell here.
(but I may be wrong, so let me look at this a lot more carefully today)

> 
> Henry
> 
>> 
>> Reto
> 
> 

Social Web Architect
http://bblfish.net/

Reply via email to