You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by Rainer Döbele <do...@take1.de> on 2008/06/23 23:32:48 UTC
RE: [PROPOSAL] Empire-db
Martijn Dashorst wrote:
>> ==== Inexperience with Open Source: ====
>> Empire-db has been Open Source from its start in 2001, but it has only
>> been publicly available since January 2008.
>
>This is odd to say the least, and from the empire-db.org [1] website I
>find this quote:
>
>"In summer 2007 ESTEAM Software decided that the solution was
>now mature enough to be released as an Open Source project under
>the name Empire-db."
>
>So it was actually open source since Jan 2008? It doesn't matter much, but
I do
>find it odd :)
>
What we mean is that from the start in 2001 we provided the source code to
anyone interested and we would also accept advice and improvements from
anyone. However there was little documentation and no sample code
available so we felt that it would be hard to get other developers
interested. In Summer 2007 we decided that it was now time to go public,
create a website and a logo, write some good sample applications and improve
javadoc to make it appealing to a bigger audience. This was also when we
decided to go for the name Empire-db. Before it was a 'component with no
name'.
It took us till January 2008 to get everything together that you can now see
under www.empire-db.org.
Hope this makes sense to you now.
>> All committers have long
>> experience in using Open Source projects, but none has served as a
>> committer on other Apache projects. We do, however, not expect any
>> difficulty in adapting the Apace development process and follow the
>> meritocracy rules.
>
>This is something the Incubator exists for.
>
>> ==== Homogenous developers: ====
>> All core developers have initially worked for the same employer, with
>> one now working for a different employer at a different location. It is
>> one of our primary goals to become a more heterogeneous community.
>
>This does make the barrier to entry a bit high for non-involved folks.
There is
>little evidence of open development available for empire-db. Only the
>sf.net forum
>is public and consists of just one active thread in the last 6 months.
>
>Also the development of an open source project should happen out in
>the open. The CVS statistics of sourceforge.net [2] show little to no
>activity regarding this project:
>
>Date (UTC) Anon Read Transactions Devel Read Transactions Write
Transactions
>Jun 2008 * 1 28 7
>May 2008 0 0 0
>Apr 2008 0 0 0
>Mar 2008 4 0 0
>Feb 2008 0 8 7
>Jan 2008 0 0 0
>Total 5 36 14
>
>What is the current development style of the committers? Is there no
>activity in CVS because the project is "done"? Or because development
>occurs off-line?
We have only moved to source-forge in January 2008 with release 2.0.0.
For the current community's needs Empire-db is already quite mature hence
there has not been a lot of demand for new features or bug-fixes recently.
This may change with a larger community which is what we're looking for.
We are aware of the fact that the lack of size and diversity of our
community is our biggest handicap.
However we hope that we are able to increase the community by giving good
reasons why Empire-db is a good choice for DBMS independent,
high-performance, full-control database access. Especially our concepts for
metadata management and compile-time safety is something we think people
really should consider.
So to answer your question: The project is NOT "done" it's just awaiting new
challenges and new participants.
>
>Martijn Dashorst
>
>[1] http://www.empire-db.org/empiredb/faq.htm#question1
>[2]
http://sourceforge.net/project/stats/detail.php?group_id=214540&ugn=empire-d
b&type=cvs&mode=12months
Best regards,
Rainer
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
AW: [PROPOSAL] Empire-db
Posted by Ma...@t-systems.com.
I'm agreeing with Rainer. I often debate with my colleagues of the achitecture unit about the benefits of Empire-db. Mostly they agree with me, but they also mention the lack of some features and integration with other frameworks. E.g. they're asking for Spring support (transaction management), caching possibilities or "Java-Object-Model" generators from existing database schema.
So, I think there are challenges awaiting the project. And only a powerful community around the Empire-db project can handle these challenges under the patronage of ASF.
Best regards,
Manuel Gamerdinger
-----Ursprüngliche Nachricht-----
Von: Rainer Döbele [mailto:doebele@esteam.de]
Gesendet: Donnerstag, 26. Juni 2008 22:20
An: general@incubator.apache.org
Betreff: Re: [PROPOSAL] Empire-db
Referring to: http://wiki.apache.org/incubator/Empire-dbProposal
Martijn Dashorst wrote:
> I find this alarming: when there are no new challenges awaiting the
> project, why join Apache? The code is stable and mature, you can just
> leave it at sf.net. There doesn't seem to be a plan other than "let's
> join Apache and hope the community grows automatically". If the
> current core developers don't have challenges for the project, how
> would the community grow?
I feel a little misunderstood here. Mature does not mean finished. We can think of a lot of improvements that we just did't have the demand for yet. And we cannot implement new features just for the sake of it. For example there are currently only four DBMS supported by Empire-db: Oracle, SQL-Server, MySQL and HSQLDB. We neither have the demand nor the expertise to support other systems, but other people might have. This is a good entry point for other developers to join in the development process. There is also demand for more in-depth work in the drivers e.g. to provide better support for stored procedures or DBMS specific features.
But most of all, we see Empire-db as more than just a DB layer. The true benefit comes from utilizing its metadata services in the presentation layer. With our Stuts2 extensions we show one possible solution how this can be achieved - limited to be used in conjunction with Struts2 of course. A similar thing can possibly be done for JSF, Wicket or others. And it's certainly also feasible for rich client user interfaces. This is not a trivial task and requires a sound understanding of the corresponding presentation layer framework. We can only provide the vision, but it takes some experts and a strong community to make it reality.
BTW: we are currently working on a new release that will allow the definition of metadata for non relational data objects like data from flat files or Web-Services in a similar manner in order to provide a uniform client level access for data and metadata. Any help with this is deeply appreciated.
> I'm not convinced yet that incubating at Apache is the best thing for
> empire-db. I think that seeking out new ideas, users and (future)
> committers for the project is the highest priority. Of course this can
> be done whilest incubating, but having incubator plastered all over
> the project won't attract more users or developers.
>
> Some mild marketing such as posting an article on javalobby
> (java.dzone.com) or theserverside.com to get some publicity will help
> the project more than becoming an Apache project.
Posting articles on javalobby or theserverside.com is certainly a good idea. But isn't this another thing that a community is good for? I will certainly support anyone who is planning on doing so.
Overall I am not convinced that leaving he project on sf.net and writing a view articles will really bring the project forward. IMO the ASF is the right platform. But of course I respect your opinion.
Regards,
Rainer
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [PROPOSAL] Empire-db
Posted by Rainer Döbele <do...@esteam.de>.
Referring to: http://wiki.apache.org/incubator/Empire-dbProposal
Martijn Dashorst wrote:
> I find this alarming: when there are no new challenges awaiting the
> project, why join Apache? The code is stable and mature, you can just
> leave it at sf.net. There doesn't seem to be a plan other than "let's
> join Apache and hope the community grows automatically". If the
> current core developers don't have challenges for the project, how
> would the community grow?
I feel a little misunderstood here. Mature does not mean finished. We can think of a lot of improvements that we just did't have the demand for yet. And we cannot implement new features just for the sake of it. For example there are currently only four DBMS supported by Empire-db: Oracle, SQL-Server, MySQL and HSQLDB. We neither have the demand nor the expertise to support other systems, but other people might have. This is a good entry point for other developers to join in the development process. There is also demand for more in-depth work in the drivers e.g. to provide better support for stored procedures or DBMS specific features.
But most of all, we see Empire-db as more than just a DB layer. The true benefit comes from utilizing its metadata services in the presentation layer. With our Stuts2 extensions we show one possible solution how this can be achieved - limited to be used in conjunction with Struts2 of course. A similar thing can possibly be done for JSF, Wicket or others. And it's certainly also feasible for rich client user interfaces. This is not a trivial task and requires a sound understanding of the corresponding presentation layer framework. We can only provide the vision, but it takes some experts and a strong community to make it reality.
BTW: we are currently working on a new release that will allow the definition of metadata for non relational data objects like data from flat files or Web-Services in a similar manner in order to provide a uniform client level access for data and metadata. Any help with this is deeply appreciated.
> I'm not convinced yet that incubating at Apache is the best thing for
> empire-db. I think that seeking out new ideas, users and (future)
> committers for the project is the highest priority. Of course this can
> be done whilest incubating, but having incubator plastered all over
> the project won't attract more users or developers.
>
> Some mild marketing such as posting an article on javalobby
> (java.dzone.com) or theserverside.com to get some publicity will help
> the project more than becoming an Apache project.
Posting articles on javalobby or theserverside.com is certainly a good idea. But isn't this another thing that a community is good for? I will certainly support anyone who is planning on doing so.
Overall I am not convinced that leaving he project on sf.net and writing a view articles will really bring the project forward. IMO the ASF is the right platform. But of course I respect your opinion.
Regards,
Rainer
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [PROPOSAL] Empire-db
Posted by Martijn Dashorst <ma...@gmail.com>.
On Mon, Jun 23, 2008 at 11:32 PM, Rainer Döbele <do...@take1.de> wrote:
> For the current community's needs Empire-db is already quite mature hence
> there has not been a lot of demand for new features or bug-fixes recently.
> This may change with a larger community which is what we're looking for.
This reminds me of a saying posed here @apache: "bad code generates
good community" (or something similar). When a project is mature, and
doesn't have a list of issues and especially new features that need to
be solved it is hard to attract new developers. The market for ORM
solutions is already crowded, so you'll probably need to convince
folks of the benefits of the project.
> So to answer your question: The project is NOT "done" it's just awaiting new
> challenges and new participants.
I find this alarming: when there are no new challenges awaiting the
project, why join Apache? The code is stable and mature, you can just
leave it at sf.net. There doesn't seem to be a plan other than "let's
join Apache and hope the community grows automatically". If the
current core developers don't have challenges for the project, how
would the community grow?
I've done some googling, and the only references to empire-db are from
the incubator list and the site itself. Is the project known outside
these 2 groups?
I'm not convinced yet that incubating at Apache is the best thing for
empire-db. I think that seeking out new ideas, users and (future)
committers for the project is the highest priority. Of course this can
be done whilest incubating, but having incubator plastered all over
the project won't attract more users or developers.
Some mild marketing such as posting an article on javalobby
(java.dzone.com) or theserverside.com to get some publicity will help
the project more than becoming an Apache project.
Martijn
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org