BSDDB copyright and licensing restrictions while in use viaPython

[Gordon McMillan]

Oleg Broytmann wrote:
> 
>    Dear Fred, don't get me wrong - I am not flaming on you. Before, I use
> Berkeley DB a little (doing Netscape's history files manipulations), but
> never touched MetaKit. After your message I downloaded and installed
> Metakit on Linux and Solaris - just to test it. I found it a little behind
> of BSDDB on quality. ./configure don't work quite right, generated Makefile
> has errors (README says about them :). Tarball is packed strange (all text
> files are executable).
>    I think MetaKit is not such a perfect solution, but may be I just wrong.
> And what's wrong with using Berkeley DB 2.0+ ?

The only complaint was about the license, (although I think 
the use of the name "DB" is presumptuous).
 
If what you need is bsddb, then bsddb is the perfect solution. 
In a test of bsddb functions, Metakit will be much slower in 
adding, roughly the same speed in retreiving, much smaller 
and use less system resources. In a test of Metakit functions 
(select, join, nested views) bsddb will be a non-starter.

> On Tue, 15 Feb 2000, Fredrik Lundh wrote:
> > Oleg Broytmann <phd at phd.russ.ru> wrote:
> > > > metakit: fast, efficient, small, highly portable,
> > > > great python interface, and free.
> > >
> > >    Hm, portable? Well, exerpt from README:
> > >
> > >    I cannot consider it "portable". Oh, yes, I have exactly 2.8.1, even
> > > worse, on Solaris 2.5.1.
> > 
> > considered using a recent version of gcc?
> > after all, the upgrade is free.
> 
>    No, it is not. Opensource software is not so cheap. It is free - like in
> freedom - but not cheap. I need to pay with my time, and no - my time is
> not cheap :)
>    BTW, I guess your time is even less cheaper, so I want to thank you for
> taking part in the discussion. Credits to you, now I have MetaKit
> installed, and will try to test it some more.
> 
> > > No other program reveals bugs in 2.8.1.
> > 
> > cool. but if that was true, why did they
> > release 2.95.2?
> 
> But being a paranoid sysadmin, I prefer to install rather old - but
> well-tested - software. 2.7.2.3 and 2.8.1 do good for me (except for
> MetaKit) - why install new gcc?  It is problem with little poor 
> boy named
> MetaKit, not gcc, I think.

In 1996 I had a client (briefly) who used the same argument to 
resist ANSI C. Being comfortable with bronze axes is one 
thing; complaining about iron another.

But the complaint was about the optimizer. Python warns 
about some optimizers - if you haven't encountered any 
problems, it's pure luck.

- Gordon