My primary tasks include sending out NetCAT invitation notices, communicating with applicants and coordinating the selection process. Once a cycle begins, I monitor traffic on the NetCAT mailing list, moderate “hot” discussions, answer or forward questions to NetBeans engineers, and organize surveys for features that are being tested, such as the Editor, Woodstock components, etc.
It is a challenging but interesting
role. I have to make sure that all participants feel valued and
motivated, but sometimes I'm the bearer of bad news. For example,
someone reports a nasty P2 bug but we decide against a fix because
it's dangerous or there is too little time. On the other hand, I
enjoy the fact that 60 people located in various time zones, from
different cultures, work together and share the same goal. It's very
encouraging to feel their passion, watch their lively discussions and
see their contributions. It's an eye-opening experience.
I think NetCAT participants in general
the role of the program very well, and have the right set of
expectations. Still, I sometimes find myself having to emphasize that
NetCAT is only about quality and not
design. On a lighter note, I've had individuals mistake the project for
Interested participants submit their application online and provide us with their personal and professional information, and testing environment—this means what operating systems they can use for testing, which JDKs they have installed, etc. All applicants are evaluated and assigned points. For example, four or more years of experience using the NetBeans IDE translates into 0.6 points; having eight hours per week to volunteer for NetCAT is worth 0.6 points; experience with Issuezilla earns an applicant 0.5 points; and so on. Points can be deducted as well. For example, no Version Control System usage is a -0.1 point loss.
Several NetBeans engineers are involved in the review process also. They select their 10 favorite applicants, which means the applicants get 1 point for each such preference. The points are tallied up and all applicants are compared. The best 60 applicants and 30 substitutes are selected.
Laskowski is a Polish developer
and a NetCAT veteran (he participated in the 5.0 cycle). His areas of
focus are J2EE and Web Services, and his work on NetCAT 6.0 earned
him the award of “Most Helpful Member”. Anuradha Gunasekara is a
developer from Colombo, Sri Lanka. He concentrated on Life Cycle tools
and SOA functionality in the project; he is the runner-up in the
“Most Agile Member” category.
I monitor the activity of all NetCAT participants, and everyone earns CAT points for his or her various tasks—two points for RFEs, four points for bug reports, five to ten points for participating in surveys, etc. The person with the most points is recognized as the “Most Helpful”.
Jacek, for example, was our best bug reporter, submitting an astonishing 82 bugs and 12 RFEs (Request For Enhancement). He participated in 4 surveys and gave excellent reviews of documentation for Enterprise applications and Web Services, to name a few. His P1 issue (#122228) about inconsistency on method removal in the Web Services Designer was accepted as RC1 show stopper and fixed in RC2.
The “Most Agile” category is a new one for NetCAT. Sometimes inactive members are replaced by substitutes who have little chance of being considered the “Most Helpful” grade because they've joined a cycle already in progress. To acknowledge these substitutes, some of whom become as active as the top participants, I introduced this category, which takes the total number of earned CAT points and divides it by the number of days involved in NetCAT. This way, a member with 50 points but less days on the project could technically be considered more active than one who started on track with NetCAT and has 90 points.
Jacek had the most points in the “Most Agile”category as well, but I have singled out Anuradha, the runner-up, for the reasons I just described.
Despite joining NetCAT 6.0 midway through, Anuradha was an excellent participant, sometimes working even faster than our engineers! He is familiar with NetBeans APIs and his speed was a definite benefit for us. He reported 26 bugs and 3 RFEs, took surveys and on two occasions fixed major bugs by himself. But NetCAT isn't only Anuradha's contribution to the NetBeans community; he also has three plugins in the Plugin Portal and succeeded in getting two of them into our Update Center.
NetCAT usually takes about two to four
months, starting with Beta and ending with the RC1 build.
The ideal NetCAT cycle is one with very active participants till the very end. The team members believe in and are invested in the success of the product because of the effort they've contributed. This is also good because they usually promote the IDE to their friends, colleagues, etc. I for one consider NetCAT 6.0 to be the best NetCAT program ever.
NetCAT 6.0 was different in several aspects. First, we opened the final Community Acceptance survey to the whole NetBeans community. Second, we introduced a replacement of inactive participants; perhaps this is why team feedback was good till the end.
This is still to be determined. NetBeans 6.1 will not have a NetCAT cycle because it will be a bug fix release without many new features.
NetCAT is a unique chance for participants to improve their knowledge of NetBeans by learning from others; and of course, friendships are created. People also get special prizes, such as limited edition pens and t-shirts or free vouchers towards the NetBeans Shop. And finally, it's an excellent addition for their CVs!