I hope Marcello will forgive me for posting this one.
Marcello and I have already agreed YESTERDAY that I should leave
Statalist. David is well intentioned but misinformed. I have never
said I will not stop nor have I declined to stop. I can provide the
copies of the email exchanges, including the one I sent to David,
should anyone be interested.
As many people know, I provided -outreg2- to Stata community in good
faith and good conscience.
http://www.stata.com/statalist/archive/2005-09/msg00428.html
As many people already know, outreg2 has been greatly abused by the
likes of Ben Jann and his German-speaking gangs. We have seen 5 years
of never-ending, shameless smearing campaign implying that somehow
outreg, outreg2, and many smallers programs by many good, honest
people including John Gallup and me, were somehow defective and should
not be used.
I have remained silent until two years ago, when it became clear that
the attacks would only get worse. I have earlier suggested that in the
least they should refrain from blatant thread-jacking, but instead
were attacked for my trouble, and have been indirectly harassed in the
form of download manipulations that has since been conducted. I
personally consider this as an attack on ssc, frivolously downloading
over 100,000 files just to make a statement.
http://www.stata.com/statalist/archive/2008-10/msg00407.html
I have done very little promotion for -outreg2-. I have never claimed
that my program was superior to another program or it should be used
in place of programs by other people. Indeed, from time to time I try
to help and fix any pre-existing programs, even though it is
considerably easier to build them directly into -outreg2-. I have
never said bad thing about programs by other people, including those
by Ben Jann.
This was never the case with Ben Jann and his gangs. They
systematically wen after other people's programs, practically raping
them. Ben Jann would cherry pick ideas and codes from other people's
programs and republish it under his name, claim that it works better
than the original. They do this despite the fact that people asking
for help on ssc has explicitly stated that they want to make it work
with outreg2, etc. We have seen this too many times, including a
recent offense by Maarten Buis, who knew what he was doing.
http://www.stata.com/statalist/archive/2010-01/msg00920.html
When someone borrows something without attribution, a polite thing to
do is to look the other way. I have done this on a number of
occassions, and have done it too many times for Ben Jann. He should
know this, how many times he borrowed and I kept silent. The problem
is that Ben Jann has done it one too many time.
estout is a knockoff of original outreg. esttab is a plagiarized
version of outreg2. This is an unalterable fact, as unflattering as it
may seem. I have previously avoided making this statement, even though
he allowed the fanboys and many people into thinking that his work was
original. The outcome is that I have been repeated accused of being
the knockoff.
As a result, I have been told by people at the Stata Journal that
outreg2 cannot be seriously considered for publication there, given
that it has been badly ripped off by Ben Jann.
Becuase of this incidence, it took me another four years to post
another program, logout, to ssc even though I had it in possession for
a number of years. When I finally did post it, I clearly indicated in
the -TERMS OF USE- that it is unprofessional for anyone to take or
borrow someone's work without giving adequate citation. I even gave
out where to look for examples on how to do this.
http://www.stata.com/statalist/archive/2009-04/msg00903.html
Rather than adding the proper citation, the so-called "stolen" passage
inside estout had been replaced the following summer with a
parapharsed version of paraphrased codes. I retroactively discovered
them by looking at the old copies of estout.
One of reasons I have finally decided to take this stand against this
nonsense of recycled codes is because I realized was uniquely
qualified to do so. The authors of much smaller programs, such as
Glenn Hoetker and his mkcorr, are in no position to protest when Ben
Jann goes out of his way to make their programs obsolete by publishing
deliverate knock off their programs. Glenn gets dismissed, Benn gets
praised, and I suppose it was meant to be submitted to Stata Journal
yet again as another worthy contribution to Stata arcana.
The great wonder is that it took this long for someone to raise a
voice against this great nonsense.
As many people may already suspect, I very much like these little
programs floating around, and I have always endeavored to avoid
killing smaller programs by writing something similar into outreg2.
There is no need for me do something like that, deliberately
overshadowing a hardwork by someone somewhere who was kind enough to
share it with other peole. From time to time, I have been asked if
outreg2 could make a correlation table. My answer to them was that
they should try mkcorr, etc. For this reason I made -dataout-
available. It is a "collaborative" program can be made to work with
other programs. This is one of reasons I like margeff, mfx2, and sum2
very much. They are collaborative, not meant to replace something
else. These programs also do a great job of giving citation to other
programs, clearly indicating that it is meant to be used
collaboratively.
The great mistake of Ben Jann is that he kept on writing "competing"
programs instead of "collaborative" programs.
This is contrary to the philosophy of "open source" programming. You
are not supposed to deliverately knockoff other people. I am not sure
which part of collaborative that anyone fails to understand.
This is one main reason I have publicly urged Kit Baum to stop
publishing the ssc ranking. It give too much incentive for someone to
write unoriginal, knockoff programs at the flimsiest of excuse,
causing great deal trouble later.
I have also publicly urged Nick Cox to help me establish some type of
citation guidelines. Nick says anyone can use his codes but it turns
out there are too many unspoken expectations that made it very
difficult to borrow his codes without tripping oneself, as we have
seen.
As Marcello found out, my N is N+1. Everyone saw it coming. This was
entirely avoidable.
Roy Wada
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
