bogus bogotuning

[Jason A. Smith]

On Wed, 2004-01-28 at 15:12, Gyepi SAM wrote:
> Open source licenses allow you to modify, improve, experiment and learn from the code.

I know what open-source is about.  Are you being condescending now?

> It does not prevent anyone from writing code the way they see fit and it is not
> a cudgel with which to beat others into writing code the you want.

I am not trying to beat anyone up.  In my opinion this arbitrary limit
reminds me too much of the stupidity of MS software that doesn't give
the user the freedom to do what they want to do.

> We certainly welcome discussion about any part of bogofilter, but people are
> more likely to pay attention to your requests when they are polite.

I have tried to politely explain my opinion to the list but everyone
except Joerg is being stubborn and closed-minded.  They won't even
consider the possibility that changing the fatal error to a strong
warning would protect them from complaints and still let people
experiment a little.  I only see a slight positive result and no
negative consequences.  What possible negatives could there be from
giving the user the freedom to do what they want at their own risk? 
They can already write their own config file with completely bogus
parameters and then complain to the list.  Why not let them experiment a
little with bogotune and with different corpus files of various sizes?

As Joerg said, "There are probably a hundred reasons to use it with a
too small corpus, and you can't know them and shouldn't try to
anticipate them."  Give the user the freedom to experiment a little
without messing around editing source files and compiling.  What if they
already have 1800 spam email saved and don't want to wait a few more
weeks for more, or fiddle with the sources?  Do you really think it is
the right thing to do to slap their hand, treating them like a stupid
little child just because they are 10% shy of your arbitrary minimum?

> If you believe that bogotune could use improvement, by all means improve it and if
> you think that your changes might be useful to others, feel free to contribute
> them to the project.

There is no point in submitting a patch if it will be rejected, right?

~Jason