axis-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dennis Sosnoski <...@sosnoski.com>
Subject Re: question on consuming service with lopsided security
Date Sun, 07 Nov 2010 01:44:02 GMT
On 11/06/2010 04:06 PM, Sanjiva Weerawarana wrote:
> On Sat, Nov 6, 2010 at 3:10 AM, Dennis Sosnoski <dms@sosnoski.com
> <mailto:dms@sosnoski.com>> wrote:
>
>     > does better. Do you seriously thnk Axis2 doesn't have many scenarios
>     > where its not hugely better than CXF?
>
>     I'd very much like to know what these scenarios are.
>
>
> I'm afraid I'm not going to waste my time to help you figure those out :).


If there are areas where Axis2 is hugely better than CXF (and presumably
Metro, too) I'd very much like to know what they are. I work with client
companies on choosing technologies, as well as providing information to
developers through my articles, and I truly want to provide the best
advice I can on technology selection.

I'm trying to cover all aspects of web services in my devWorks column.
At the rate of at most one article per month that's a slow process, and
the last year+ has been covering the security area. It happens that in
that particular area Axis2 has some issues, especially around
performance. I'm trying to wrap up the security area for now, and hope
to move on to other areas soon, so if you can point out what you see as
Axis2's special strengths I'd definitely be interested in covering these
in the column.


> You've demonstrated repeatedly that you have the power and apparently
> the will to write articles that are intentionally damaging to Axis2.

I'm not sure exactly what you're suggesting here, unless it's
"intentionally damaging to Axis2" to publish performance results which
show Rampart's AXIOM-based security handling operating much slower than
other alternatives. But I didn't start out doing performance comparisons
to show that Axis2 was slow, that's just the way the results came out.
When I found the problems, I did my best to figure out why they were
occurring and to help point the Rampart and AXIOM developers in the
right direction to resolve them. See
http://marc.info/?l=rampart-dev&m=126017138404209&w=2 (which I sent well
in advance of the publication of the article, and which didn't even get
a response) and http://marc.info/?l=rampart-dev&m=126489321412411&w=2
(after the article was published, which led to considerable discussion
on both the Rampart and Axis2 lists and work to improve the performance
issue).

I do realize that WSO2 has built its whole operation around the
commercialization of the Axis2 stack and related technologies, and as
founder and chairman of WSO2 you may rightly feel that negative
publicity for Axis2 hurts your company. But I don't think that gives you
the right to effectively accuse me of some sort of personal vendetta
against the project.

  - Dennis

Mime
View raw message