[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: password protect a directory?



>>>>> "kmself" == kmself  <kmself@ix.netcom.com> writes:

    kmself> http://www.bitkeeper.com/

Thanks for this...

    kmself> BitKeeper is "a scalable configuration management system,
    kmself> supporting globally distributed development, disconnected
    kmself> operation, compressed repositories, change sets, and named
    kmself> lines of development (branches)".  It's written by Larry
    kmself> McVoy, written in part to address the problem of "Linus
    kmself> doesn't scale", and is aimed at being the version control
    kmself> system for Linux kernel development.  Larry's been
    kmself> involved with GNU/Linux for ages, and is the author of
    kmself> "The Sourceware Operating System Proposal", an early
    kmself> formative document regarding options for a free software
    kmself> OS:

    kmself>     http://www.bitmover.com/lm/papers/srcos.html

bitkeeper looks like it has everything I always wanted from CVS 

BUT:

    kmself> It's not DFSG.  It *is* source-available software, but
    kmself> it's not free for commercial use, without some compromises
    kmself> a company might prefer to do without. It *is* free for use
    kmself> on free software, and the source is available.  Licensing
    kmself> has been an issue, this is how Larry thinks he can make
    kmself> money with his company.  We're using BitKeeper at work.
    kmself> It's used by several other companies, including VA Linux.

I would never use it myself. When I use software like this,

- I like to experiment with test data, to see exactly how it
is going to work, what is limitations are, etc.

- I currently use CVS for projects that aren't open-source (eg. my
Thesis, in LaTeX format), that aren't really commercial (I never
intend to make any money from it), or somewhere in-between (ie
software projects I am playing around with, but not significant enough
to release as open-source projects) but I don't particular want to
make public either.

It seems from the license, either I would have to pay the fee for the
commercial version, just so I could legally use it without having all
my code/document changes sent to a public server, at my expense (I pay
for the bandwidth to/from my computer). For details see:

<URL:http://lwn.net/1999/features/BitKeeper.phtml>

Also, it seems to rule out disconnected operation, ie. having a CVS
repository on a laptop computer, for instance. Then again, I wonder -
what does happen in that case? Does it come up with an error "unable
to contact public logging server - aborting", or is it just a warning?

Also I don't buy their argument:

"The goal of the BitKeeper license is to maximize value to all of these
parties. It is trying to make sure that everyone gets as much as they
can get. Contrast this with traditional open source licenses.  These
licenses are completely focussed on one segment: the Open Source
community. That's fine, but it leaves the other people out in the
cold."

I don't think anything that anyone in the open source community is
saying leaves anyone out in the cold. The vendors charge for providing
technical support. That keeps the open source community happy (it
would really would be open-source), the commercial users happy (they
can get technical support), and the vendors happy (they get a profit
for providing commercial quality technical support. Thats all the
parties mentioned in

<URL:http://www.bitkeeper.com/free.html>

All this without gimmicks, such as requirements to use a public
logger.

    kmself> There are several advantages of BitKeeper over traditional
    kmself> version control software.  I'd encourage you to look at
    kmself> the BitKeeper site for details.  In particular, BitKeeper
    kmself> supports the idea of multiple repositories, meaning that
    kmself> different departments within an organization, or different
    kmself> groups, each with specific needs, can work without having
    kmself> to worry about conflicts with the central repository.  I'm
    kmself> probably phrasing this badly, would suggest you look to
    kmself> the website for further data.

Agreed.

-- 
Brian May <bam@debian.org>



Reply to: