commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ceki Gulcu <c...@qos.ch>
Subject Re: [logging] logging vs slf4j
Date Tue, 09 Aug 2011 17:16:04 GMT
On 09.08.2011 11:57, Christian Grobmeier wrote:
>>> Another option is to try to work with Ceki to address some of the
>>> concerns of the commons community with regards to using slf4j.
>>>
>>> * There is a hassle with too many jars for dependencies with slf4j.
>>> * Every time Ceki goes on vacation everything stops.
>>> * Some have a preference for Apache driven projects.
>>> * Figuring out the dependencies that are needed can be difficult.
>>
>> Another option would be to try to convince Ceki to move his project to
>> the ASF?  He is an ASF member, right?  What were his concerns about
>> the ASF that made him start his project elsewhere?
>
> Ceki is an ASF member and even a Logging PMC.
> You can read most about his concerns on his blog, for example:
> http://ceki.blogspot.com/2010/05/forces-and-vulnerabilites-of-apache.html
> http://ceki.blogspot.com/2010/05/committocracy-as-alternative-for.html
>
> He seems to be very opposed to the ASF model; there was much bad
> feelings in the "log4j case" before I started with logging and
> therefore I doubt that Ceki is willing to go back to the ASF. At least
> his blogposts reflect that he is not satisfied with the "Apache Way"
> itself instead of personal trouble, which we might be able to solve.
> Ceki is reading this list, so maybe he wants to elaborate a bit more.

* On the ASF model

In a nutshell, while the ASF is a great organization in many ways, it
is not a meritocracy mainly because merit is not measured at all and
at the project level no responsiblity is accrued on merit beyond
committership. BTW, the BDFL model is not a meritocracy
either. Finding a good model for running organisations is no simple
matter. The Apache model may even be better than some but it bothers
me that the ASF misrepresents itself as a meritocracy.

* On logging

One of the disadvantages of a relatively large and open ecosystem such
as Java is that you get several competing libraries solving common
problems. Logging is a rather devisive case in point.

SLF4J has deficiences with regards to modularity are imho testimony to
the lack of a widely accepted solution to modularity in Java. In any
case, I am unaware of a satisfactory solution. OSGi is of course one
possible approach but OSGi is a complex beast and not that widely
accepted (yet).

As for deficiences in SLF4J API, varargs support [1] and event data
support come to mind. Varargs support is very likely to be added in
SLF4J 2.0. As for event data support mentioned by Ralph, the idea was
initially proposed by Joern Huxhorn. I am not entirely convinced by
it.

However, Ralph is right to observe that had SLF4J been developped at
the ASF and that Ralph, Joern and me were the only SLF4J
developpers, event data support would probably be part of the SLF4J
API. Having said that, since event data support is a little
controversial, it might have been turned down at the ASF as well
(assuming the involvement of more developpers who shared my lack of
enthusiams for event data).

Anyway, Ralph made some very valuable contributiuons to the logback
project and I am sorry to see him stop. Given what I have seen of his
work, he will surely do an excellent job with log4j v2.

[1] http://bugzilla.slf4j.org/show_bug.cgi?id=31

-- 
Ceki

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Mime
View raw message