Re: sasl version mismatch: 2.1.15 vs 2.1.17

[Igor Brezac <igor@ipass.net>]

On Wed, 10 Dec 2003, John Beamon wrote:

>
> Igor Brezac wrote:
> > On Tue, 9 Dec 2003, Andreas wrote:
> >
> >
> >>On Tue, Dec 09, 2003 at 01:16:13PM -0800, Howard Chu wrote:
> >>
> >>>As an example, BDB 4.1.17 changed its db_open API and added an additional
> >>>parameter in the middle of the parameter list. The API is completely
> >>>incompatible with what came before it, but the minor version is still the
> >>>same. The version number is > 4.0 so we should be happy with it, from a
> >
> >
> > DB format changes complicate matters even further.
> >
> >
> >>This is an example of a vendor (Sleepycat) who doesn't care about incompatible
> >>changes between minor release. Bad cat :)
> >>
> >>With this type of vendor one cannot rely on the soname (if it contains only
> >>the major version) at all. That's why I asked if cyrus-sasl had such behaviour
> >>in the past.
> >>
> >
> >
> > What vendor?  There is no such thing.  I am not sure where you are going
> > with this rant.  The software you are talking about is contributed to the
> > community for free.  Sure it is not perfect, but it is very good and
> > getting better.  You can make it better.  All all these projects welcome
> > quality contributions.
> >
>
> This is a common reply to complaints about this feature or that quality
> issue in open source software, but the truth of the matter is that a few
> dozen people are capable of fixing this and a few thousand people use
> the product.  I am personally highly dependent on the quality and
> consistency of the open source softwares I use, and I am completely,
> 100% incapable of coding an improvement or a bugfix.  There is no
> reasonable evaluation of my job description that implies I should be
> able to fix one of these problems, but the vendor (Sleepycat) should be
> able to keep up a consistent and stable product.  BerkeleyDB is a fine
> product and widely used, so my issue's not really with them.  My issue
> is with the mailing list mindset that prompts replies of "if you don't
> like it, fix it yourself".  "Bad cat :)" does not constitute a "rant".

I disagree.  Open source developers are not accountable to anyone, so you
may want to review your high dependency on the software.  It is just that
they choose to write high quality stuff and they respond to problems
rather quickly.  I never said ""if you don't like it, fix it yourself";
you do not have to be a programmer in order to make quality contributions,
how about testing, writing documentation, etc...

As Andreas pointed out, we are beating a dead horse, and it is my fault
for bringing this up.  I felt that the criticism was not constructive in
this instance and I chimed in.

-- 
Igor