[os-infrastructure] The infrastructure

[Joe Abbate]

Hi Daryl,

Daryl Monge wrote:
> On Jun 12, 2008, at 4:10 PM, Joe Abbate wrote:
>
>> * Current thinking is that a vibrant and active community will want 
>> to be free from commercial priorities
> Huh?  Who is thinking that?  Everyone has commercial priorities.  They 
> just might be different than yours.  Even a University wants solid 
> research to justify grant funding.

Those are the words of Roy Hahn, UK IUA Chair and the main instigator of 
the Ingres Janitors project.  If everyone has commercial priorities, are 
you saying no companies are using Fedora, CentOS, Debian and other 
non-commercial Linuxes?

>
>> * Product could fork (bifurcate)
> This simply is not going to happen.  Any more than MySQL has been 
> vulnerable to a "fork" over the last several years.  Plus, we still 
> own the IP.

"Own" the IP in my considered opinion is somewhat meaningless if you 
have released the source under GPL.  Anybody  determined enough could 
create a fork.  All they have to do is feed back the changes.  If they 
haven't done so (in both the MySQL and Ingres cases) is because it 
requires a significant amount of work, i.e., it's not like forking FTjam 
from Jam.  But Roy's premise is that there could be a "vibrant and 
active community".

>
>> * The community version is expected to drift ahead of the commercial 
>> version
>>  - ?possibly acquire a distinct name
> Why?  Where has that happened?  MySQL? Postgres? Apache? OpenLDAP? 
>  Sendmail? Postfix? Samba? X11? OpenSSL? Kerberos? (Microsoft not 
> withstanding) The closest proper example might be the Firefox forks 
> like Camino.  
> (Great example too.  Firefox 3 is awesome.  It is staying ahead of its 
> forks in features and reliability.  And the folks are not 
> "competitors" anyway)

There have been forks in open source.  Apache was a fork of NCSA httpd.  
Xorg was forked from XFree86 due to licensing disputes.

>> The above has *major* implications.  CentOS (the example given by 
>> Roy) "is not maintained or supported by Red Hat" (Wikipedia) and 
>> their home page doesn't even recognize Red Hat (referring to it only 
>> as "a prominent North American Enterprise Linux vendor").  Do we or 
>> more importantly does the Ingres board want an Ingres fork?  If so, 
>> will they be willing to fund or staff such an effort?
> Are you arguing that we should not be open source?  Are you saying 
> that a contribution process will cause a fork, but the existence of 
> the source by itself will not?

I was only reflecting on what Roy Hahn, an Ingres customer since 1984, 
was stating in his presentation.

Joe