[Owasp-board] A New Projects Model

Jim Manico jim.manico at owasp.org
Sun Apr 13 22:15:48 UTC 2014


Ø  Dropping one project category is very important as it makes it black and
white in terms of what we're pouring our resources into and what we are
not.



I like it in general, but we still need to address quality.



Ø  Look at ESAPI as an example.  There's a project leader currently, but
has any traction been made lately?  But the concept is fantastic and
probably something that we should be supporting additional development on
and not just sandbagging it



If ESAPI continues as an OWASP project while it is "not production ready"
then this does not accomplish the quality control that I am seeking. I
think we still need additional project labeling.



Something like:



·         OWASP Project - Release Quality

·         OWASP Product - BETA Quality

·         OWASP Product - ALPHA Quality

·         Incubator Project



Or something similar. I want to make sure our labeling is clear as to the
quality and maturity of a project, regardless of what level of OWASP
support the project is at. I feel like the ESAPI guys labeled it RELEASE
quality while it was early Alpha in order to gain adoption. This is
deception of the highest order. That really concerns me and is "bad for
business".



So I totally agree ESAPI deserves more OWASP support, but right now I see
it as a "OWASP Project - BETA Quality" project under the new dual system.



Does my concern make sense?



Aloha,

-          Jim



*From:* Josh Sokol [mailto:josh.sokol at owasp.org]
*Sent:* Sunday, April 13, 2014 10:31 AM
*To:* Jim Manico
*Cc:* Michael Coates; OWASP Foundation Board List
*Subject:* Re: [Owasp-board] A New Projects Model



Dropping one project category is very important as it makes it black and
white in terms of what we're pouring our resources into and what we are
not.  But where the rubber meets the road is in who is actually owning the
project (ie. who decides the features we are working on, the contributors,
the release schedule, etc).  If it's an OWASP project, then our PM should
be providing the organization behind this and the direction should ideally
be driven by the community through some sort of upvote/downvote method.
Look at ESAPI as an example.  There's a project leader currently, but has
any traction been made lately?  But the concept is fantastic and probably
something that we should be supporting additional development on and not
just sandbagging it.  If others agreed, then we should be doing like a
reddit system of features.  What needs to happen to make it ready for the
big show?  Now that we have our priorities, we could arrange for people to
do it.  If someone really wants some functionality, they can sign up to do
it or we can always contract it out.  So, while technically the license
would allow us to do this under the current system, the concept of "project
leaders" makes it difficult.  It also makes it difficult for us to reward
the main contributors because they are likely filling that role of project
leader as well.  I know I've seen a number of different posts on that
subject arguing that it's a conflict of interest to pay people to work on
projects.  This gives us flexibility to do that without any sort of a
conflict of interest.

~josh



On Sun, Apr 13, 2014 at 2:58 PM, Jim Manico <jim.manico at owasp.org> wrote:

I'm confused, how does this differ from what we (are supposed to) have
today other than dropping one project category? What does "OWASP ownership"
provide to us that "open licenses" do not provide?



I like where you are coming from so far, I'm just confused at the details.



-          Jim









*From:* Josh Sokol [mailto:josh.sokol at owasp.org]
*Sent:* Sunday, April 13, 2014 9:54 AM
*To:* Jim Manico
*Cc:* Michael Coates; OWASP Foundation Board List


*Subject:* Re: [Owasp-board] A New Projects Model



Jim,

To answer your question about the review process, I said:

If one of those #2 projects ever decides that they are done running by
themselves and wants #1 status, they can donate their code to OWASP and a
technical council would give it a thumbs up or down as to whether it
belongs in #1 project bucket.  The question ultimately being whether it's a
project that is visionary and benefits the world at large or something
solving some smaller corner case issue.



The key here is that the project no longer becomes the individual's
project, but rather, OWASP's project.  This makes it super easy for us to
replace stagnant leaders, pay people to work on the project, and drive the
project toward our bigger vision.



~josh



On Sun, Apr 13, 2014 at 2:46 PM, Jim Manico <jim.manico at owasp.org> wrote:

Instead of ...



40 hrs in a week x 60 minutes in an hour / 177 projects = <14 mins per
project per week

I was thinking more like.



40 hrs in a week x 60 minutes in an hour / 10 projects = <140 mins
reviewing 10 projects per week with a 17 week cycle. That's a lot more
realistic.

The "OWASP Project" concept is interesting. The big and tough question is,
what project get that status? You are basically saying drop "labs" and only
have "flagship" and "incubator" and nothing in between. We still need a
review process that awards projects with "OWASP Project/Flagship" status. I
like this simplification, by the way.



Aloha,

Jim





*From:* owasp-board-bounces at lists.owasp.org [mailto:
owasp-board-bounces at lists.owasp.org] *On Behalf Of *Michael Coates
*Sent:* Saturday, April 12, 2014 1:27 PM
*To:* Josh Sokol
*Cc:* OWASP Foundation Board List
*Subject:* Re: [Owasp-board] A New Projects Model



Josh

I think there are some interesting ideas here. I like a model that creates
a few very high quality projects that we can rally around and promote. At
the same time u like how people can still experiment under our Owasp
umbrella in the other area (whatever it is called).

You also hit the nail on the head with our current setup. The math and time
just doesn't work. Noone could do it.

Let's keeping thinking through how this might look. I like the bold new
ideas.

Let's celebrate quality.

On Apr 11, 2014 3:22 PM, "Josh Sokol" <josh.sokol at owasp.org> wrote:

Board,

It's clear that our current model for OWASP Projects isn't working.  I
don't think that you can place blame on any one person or entity.  Do the
math for a second.

40 hrs in a week x 60 minutes in an hour / 177 projects = <14 mins per
project per week

Do you really think that any project manager can make headway on 177
projects given less than 14 minutes a week of time to spend on any one
project?  Most of the e-mails that I write take more time than that.  And
that's in addition to everything else that we've assigned to that role.  As
we grow this problem is going to get worse and worse.  We need to address
this issue head on by evolving our projects model.  Now, I can't claim to
have been intimately involved in projects as many of you have in the past
and present.  I also can't claim to have all of the answers.  But I do feel
strongly that the current model will not scale and we've felt the pain for
quite a while.  Hiring Samantha to manage projects was a temporary bandage,
but we need a change in direction to ultimately stop the bleeding.

Far be it from me to present a problem without a solution so I'm going to
give it my best.  It's just an idea at this point and it's not fully
vetted, but hey, that's what a Board is for, right?  To work together,
create a new vision, and hand off to our staff to execute.

*The Proposal:*

We forget about trying to classify projects into buckets based on a level
of maturity.  Instead, projects are effectively one of two things:

1) An OWASP Project: This is a project owned and maintained by OWASP.
Features should be directed by the community via a up/down voting system a
la reddit.  Work on the project can be paid or unpaid.  These should be
projects of massive value not only to OWASP, but to the security community
and even the world.  We should spend 95% of our project time and resources
here because this is where the rubber meets the road.  We should be pimping
the shit out of these.

2) Everything Else: These are cool ideas that OWASP supports.  When they
submit, we give them some basic classification questions to figure out why
a security professional would want to use them.  We create a system to
search and sort these tools.  We help to pimp them when possible, but
direction, vision, etc is driven by the individual project leads.  The only
real requirement here is that the project is open source.  It doesn't even
have to live on our servers.  We're effectively an index for these and do
little more than that.  Maybe we track a last update or level of activity
or something, but nothing more.

All a project needs to do is submit a simple form and be verified as having
the proper licensing/source availability to be listed on our site.  This
should be easily manageable and scalable as we're just managing new
submissions, not existing projects.  If one of those #2 projects ever
decides that they are done running by themselves and wants #1 status, they
can donate their code to OWASP and a technical council would give it a
thumbs up or down as to whether it belongs in #1 project bucket.  The
question ultimately being whether it's a project that is visionary and
benefits the world at large or something solving some smaller corner case
issue.

Under this model we can certainly have stakeholders, but the true owners of
any OWASP project is the community.  The model provides recognition to all,
but allows us to put our eggs in the basket with the tools that matter the
most.  It helps to prioritize resources and allows the community to set the
vision and direction of OWASP.

So, what do you think?  I realize that this is a big time change in
direction, but it's well in line with our mission and builds on our
strengths rather than allowing us to skate along with our weaknesses.  I'm
interested in your feedback.

~josh






_______________________________________________
Owasp-board mailing list
Owasp-board at lists.owasp.org
https://lists.owasp.org/mailman/listinfo/owasp-board
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.owasp.org/pipermail/owasp-board/attachments/20140413/1912f30d/attachment-0001.html>


More information about the Owasp-board mailing list