Nmap Development mailing list archives

Re: Nmap SoC 2007 Wrap Up


From: jah <jah () zadkiel plus com>
Date: Fri, 05 Oct 2007 15:48:00 +0100

Well done all.  A great tool keeps getting better.

Fyodor wrote:
Hi Everyone.  The Google Summer of Code is now over and things are
calming down a bit.  So I finally have time to send a report on our
results.  For comparison, here are our previous summer wrap ups:

2005 Results: http://slashdot.org/comments.pl?sid=183143&cid=15133184
2006 Results: http://seclists.org/nmap-dev/2007/q1/0235.html

First, let me thank Google for accepting Nmap into this program for 3
years in a row!  It really has made a huge difference in Nmap
development.  It is hard to imagine any Nmap users reading those
previous reports (and this one) and not finding at least a few changes
which have helped them personally.

This year we tried a few things different.  For one, we took on fewer
students so that we would have more time to spend on each project.
Instead of 10 students as in '05 and '06, we had 6.  This worked out
well in helping me keep up with all the changes.  The project SVN
server is now public (unlike last year), which helped a great deal.  I
didn't feel like such a bottleneck in manually tracking and applying
patches.

I started the summer by making an /nmap-exp/soc07 Nmap branch for SoC
development.  That was probably a mistake, as it lead to less testing
because many people still used /nmap and it was also a major hassle to
merge stuff back.  So late in the program we moved back to /nmap and
let people use their own special experimental changes, them merge
directly to /nmap when ready.  Another problem with our first
approach, as we've seen this week, is that some people still don't
realize that we've moved back to /nmap :).

Perhaps this is the most exciting news for this summer: Two of last
years' SoC students became SoC Mentors this year!  Diman Todorov, who
helped create NSE last year mentored two NSE-related projects this
year.  And Adriano Monteiro, who worked in '05 and '06 as an Nmap SoC
student to create the new UMIT Nmap front end, was sponsored by Google
as a separate SoC project with 7 students of his own!  I'm also
delighted that both Diman and Adriano are visiting the SF Bay Area
right now for a Google SoC summit.  I met Adriano for the first time
in person yesterday, and will likely meet Diman for the first time in
person today!

In terms of success rates, we are improving.  7 out of the 10 Nmap
students succeeded in '05 and 8 out of 10 in '06.  For '07, we had 5
successes out of 6.  The increasing success rates may be due to better
selection and mentoring, but a big part of the increased success rate
is that we have been inviting the best students back for the next
year.

If we are invited back next year, our biggest change will probably be
more intense recruiting.  We have received fewer and fewer applicants
each year, and I think part of the reason is that the number of
projects participating in SoC is ballooning.  So applicants have
hundreds of projects to choose from rather than dozens.  Therefore we
will have to work harder to attract the best candidates, and I hope
members of the Nmap community will help by recommending the program to
talented college students (or by applying if you are such a student!)

Now for the important part of this email: What those five successful
Summer students accomplished!  You can read more about any of these
changes in the official Changelog at
http://insecure.org/nmap/changelog.html .

Stoiko Ivanov accomplished a whole lot of NSE work this summer.  In
addition to many smaller features and bug fixes, he added the NSE
library (nselib) of useful functions, added garbage collection
support, and added the --script-args system for passing arguments to
scripts.  He also made major improvements to the NSE documentation at
http://insecure.org/nmap/nse/ .  Diman deserves part of the credit for
this due to his excellent mentorship!  And he even took numerous
matters into his own hands, such as testing and integrating Marek's
raw IP packet NSE support patch.

David Fifield had a busy summer, accomplishing an incredible amount of
Nmap work.  His contributions include substantial UMIT work,
integrating huge numbers of your OS detection fingerprint submissions,
adding the --servicedb and --versiondb command-line options,
dramatically reducing build dependencies, the huge massping migration,
and more.  Whew!

Meanwhile, Kris Katterjohn kept very busy completing a large number of
small tasks rather than one huge project.  He helped integrate UMIT
into the Nmap build system, wrote several valuable NSE scripts, added
the Snprintf() and Vsnprintf() wrappers which are safer than the
normal lowercased calls, upgraded libpcre and libpcap, improved ICMP
protocol unreachable response handling, added several features to the
Nmap XML output format, and more.

Doug Hoyte is our only three-time SoC student, and unfortunately (for
us) he will probably graduate and become ineligible next year.  But
maybe he can come back as a mentor!  He accomplished a lot this
summer, including writing NSE scripts and integrating tons of your
version detection signature submissions.

Eddie Bell was our other repeat student, and he did a good job on a
wide variety of tasks.  He wrote five NSE scripts, improved UMIT's
results searching, fixed numerous bugs bugs, upgraded Winpcap, and
optimized the Nmap ./configure system.  Also, we added his nifty
--reason feature which fell through the cracks and didn't make it in
last year.

While Adriano was leading a separate UMIT GSoC project, he and his
team accomplished some great things.  After 3 summers of development,
UMIT is finally ready for integration into Nmap and it is included in
the latest SOC release.  I believe that 6 of Adriano's 7 SoC students
succeeded.

Obviously we also had many great contributions from non-SoC developers
over the summer, as demonstrated by the Changelog.

Note that all of the changes discussed here have already been
integrated into Nmap.  We are up to version 4.22SOC6, and quickly
approaching a big stable release.

Well, that wraps up GSoC 2007 for the Nmap project.  We're looking
forward to Summer 2008, but certainly won't be sitting on our hands
until then!  I'm particularly pleased that many of the SoC students
have continued contributing even though the summer has ended.

Cheers,
Fyodor

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://SecLists.Org

--
This email has been verified as Virus free
Virus Protection and more available at http://www.plus.net

  

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://SecLists.Org


Current thread: