On Wednesday 09 June 2010 15:40:47 Wolfram Schlich wrote:
> * Kern Sibbald <[email protected]> [2010-06-09 15:18]:
> > On Wednesday 09 June 2010 14:20:32 Wolfram Schlich wrote:
> > > Dear Kern, dear Bacula developers,
> > >
> > > do you plan to change this part of the implementation
> > > of Bacula to greatly improve the speed of copy jobs
> > > for tape-to-tape copies?
> >
> > Yes, I am working on this now (a very low priority project for the
> > moment) but it will probably only be in the Enterprise version and not in
> > the community version.
>
> Ah, interesting.
> I thought the enterprise version would not have more/better
> features than the community version?!
> AFAIR it was said that the enterprise version would just
> be more stable/mature (and has probably less features than
> the community version) and backed by service agreements.

That is what I had previously said, I will explain the change and why below.

> So from that point of view, your statement is kind of
> worrying to me. 

No need to worry.

> Does it also mean the source of the enterprise version is closed?

No, it is and will remain Open Source (probably GPLv2) as is the case for the 
community version.

>
> What about funded development for that feature, so that
> the community version would also benefit from it?

Anything that is contributed by the community will be available to the 
community. 

Concerning the Enterprise version -- what has happened is the following:
In early May, I went to a RedHat partner's conference in Valencia, Spain, and 
several things happened that surprised me.  First, I learned that Bacula will 
be in RHEL 6, and Amanda will be removed.  Having Bacula in RedHat has been 
my dream over the last 8+ years, but when I started thinking about it, I 
realized that with it in RHEL, RedHat will be supporting Bacula and hence in 
serious competition with Bacula Systems.   After discussions with RedHat, it 
may turn out that Bacula is not going to be actually in RHEL, but rather in 
EPEL (if I have the mnemonic right), which means that Bacula will be 
available to RedHat users but not directly supported by RedHat.

The second thing that happened is many partners have been complaining that 
they cannot sell the Bacula Enterprise for the reasons you mention above -- 
the community version has everything the Enterprise version has so why would 
anyone want to "pay" for the Enterprise version?   Fortunately some people 
will pay for support, but not enough for Bacula Systems to continue doing 
business the way it does today.

The third thing that happened is that I spent about two weeks fixing bugs that 
occur in Bacula 5.0.2 mostly in installations running upward of 200 jobs a 
night.  So I spent two weeks of my time working for corporations for free 
when I would rather be developing code.  I am not complaining about those 
companies -- they are doing nothing wrong, but on the other hand, we cannot 
pay bills by spending two weeks fixing two bugs for big users.  I don't mind 
spending a lot of time fixing bugs for the "community", but I don't enjoy 
fixing difficult bugs for people who can afford to pay but do not.  Again, I 
want to make it clear that I am not criticising anyone -- this is the way 
Open Source works, but at the same time, since I have and am contributing for 
free, I should be able to decide the use of my own time.

Finally, if you look at the New Features section of the 5.0.x manual, you will 
begin to realize how much Bacula Systems has invested in Bacula over the past 
year to 18 months.  This has cost Bacula Systems a lot of money and brought 
no return.

So, I am faced with a problem: either Bacula Systems stops contributing and 
the Bacula project sprials down, or we change the way we do business and 
differentiate the community and Enterprise versions a bit more than today.  
Since I have devoted 10+ years on Bacula, I wouldn't like to see it spiral 
down.  My solution is the following:

1. Future features that are developed by Bacula Systems (includes contracted 
work) may on a case by case basis either be put into the community version or 
appear only in the Enterprise version.

2. All features contributed by the community will go into both versions.

3. Both Bacula Enterprise and the community version will remain fully Open 
Source.

4. We will release the Enterprise source code only to customers.

5. Certain developers will have access to both the Enterprise and the 
community versions.  Not all those developers are Bacula Systems employees.

6. I have wanted to migrate the bugs database for quite some time to a new 
bugs system, which we will be doing in the next week or two.  This new bug 
tracking system will be RT and will be housed on Bacula Systems computers.  
It will feature three different bug levels: Community, Basic, and Enterprise.  
The Community bugs will be much like it is today open to everyone, you get 
what you pay for, either the bug is fixed by the community or when we find 
the time (not so easy) we will do our best to fix bugs.  The Basic and 
Enterprise bug tracking will be paying and you will get faster response and a 
lot more effort in fixing bugs and getting patches.

7. Bacula Systems will be offering several lower cost subscriptions for small 
enterprises so that it will be easy enough to get the Enterprise version if 
you really need or want it.

Bottom line: not a whole lot is going to change concerning the community.  
Bacula will remain Open Source, and all community contributions, and a good 
number (but not all) of contributions from Bacula Systems will be freely 
available.

Contributions from Bacula Systems have been enormous over the last 18 months 
will in the future in part go into the community version and to probably a 
larger extent go into the Enterprise version.  This will be decided on a case 
by case basis. The Enterprise version will remain Open Source, but will be 
only available to special developers and to customers with a support 
contract.

To the extent that the community fixes bugs and submits new features, there 
will be little or no difference between the community and the Enterprise 
versions. Unfortunately, I don't think this will be the case because over the 
last 8+ years, the largest part of the code has been written by me or Bacula 
Systems.  Many of the features that are being or will be developed by Bacula 
Systems that are candidates for being only in the Enterprise version are 
features that most "community" users would not really find that useful.

Best regards,

Kern



------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
Bacula-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-devel

Reply via email to