Subject: Re: [argouml-dev] commitment for 0.28 features - msg#00024

List: db.axion.devel

Date: Prev Next Index Thread: Prev Next Index

I agree absolutely with your motivations here Thomas.

Its just that some of our issues are great long trails of discussion
and difficult to follow. When they get raised again there is a great
deal of repeating of previous conversations.

When we have a wiki it will be a lot easier to iteratively work out a
design and dependencies.

BTW - making undo dependant on eUML isn't strictly true. We don't want
to make the core argouml product on that subsystem.

I think it is more that eUML and undo are dependant on RCP. eUML uses
RCP but we currently have no roadmap of bringing RCP usage into core
ArgoUML.

Not having developer resource actually is an excuse to some extent for
not having tackled things like undo. At least it has been my excuse.

I got involved in ArgoUML because I had the skillset as a swing
developer. I had to learn the codebase of ArgoUML and GEF but I could
be productive almost immediately on the project with small things and
gradually dig deeper.

Now I find that I have to go and learn a new framework to move to RCP.
I'd love to have the time to do that but it doesn't fit too well with
having a busy day job and I'm sure I'll go through a learning curve
through which I'll need some mentoring myself.

I'm very pleased to see the likes of Brian added to our skillset and
help us move in the right direction. I won't bother adding
congratulations to that thread, I'll just add here that you're very
much welcome on board Brian.

I'm happy to work under the direction of the argoecplise team members
on how we best move forward until my own skills are more fully rounded
regarding RCP development.

I have no great feelings of protectionism for any of my code,
tigris-GEF or even my role in the project. What is important is the
future of ArgoUML. I think that requires as much effort as possible
into a move to RCP but I am frustrated by my own inability to help
that happen at the moment.

Bob.

On 05/09/2008, Thomas N. <thn-d@xxxxxx> wrote:
> Hi Bob,
>
> oh I see I gave the wrong link to the Undo issue, it is:
> http://argouml.tigris.org/issues/show_bug.cgi?id=1834
>
> I know of the dependency of undo from euml. But if we really want to wait
> with undo until we move to euml, then the above issue has to be made
> dependent from an appropriate euml issue. I already proposed to do that step
> without introducing UML2 features, but I have not enough knowledge if it is
> possible to do that and thus get from the GSOC result to a stable euml
> ArgoUML. Anyway, I would say: the massive voting for Undo urges us to really
> start working on that!
>
> The problem with ArgoUML is, that we make releases without considering enough
> the users wishes. And votes are all we have to measure them. Having not
> enough developer resources is not an excuse for doing other things than
> requested, otherwise we should declare ArgoUML an internal tool for some
> hobbyists that by accident is freely downloadable.
>
> Thomas
>
> -------- Original-Nachricht --------
> > Datum: Thu, 4 Sep 2008 17:40:57 +0200
> > Von: "Bob Tarling" <bob.tarling@xxxxxxxxx>
> > An: dev@xxxxxxxxxxxxxxxxxx
> > Betreff: Re: [argouml-dev] commitment for 0.28 features
>
> > This needs some balance between how serious we are in moving toward
> > eclipse RCP.
> >
> > Why continue with our own Undo framework when that would be provided
> > by RCP. I'd hate to do lots of needless work that will just be thrown
> > away.
> >
> > Even if we did continue with our undo implementation it would be
> > dependant on eUML being complete I think.
> >
> > I'd like to make sure we do concentrate on seq diagrams but that does
> > not have to what for a full release. It could be just a module
> > release.
> >
> > Personally I think it is just a little too early to discuss these
> > plans in detail now. I'd like to wait till we have a dev wiki so that
> > we can start to build pages to agree designs for future. We can then
> > discuss what of those designs we would like to see soonest taking into
> > account both user votes and technical constraints.
> >
> > The wiki will hopefully make that process a lot easier than bouncing
> > emails around.
> >
> > Bob.
> >
> > On 04/09/2008, Thomas N. <thn-d@xxxxxx> wrote:
> > > Hi,
> > >
> > > let's agree on the plan for 0.28 as early as possible, because 0.26
> > lacks of many important things.
> > >
> > > First, as we always encourage users to vote for issues, let's get
> > serious here and put the issues with the most votes on highest priority!
> > These
> > are: Undo, distributed models, and movable labels of edges in diagrams.
> > Furthermore, we have a better sequence diagram implementation that is in
> > need of
> > a migration path from the current seq diagram impl.
> > >
> > > So, I propose to plan the next release to be released in a reasonable
> > short time (in great contrast to 0.26), with the following features:
> > >
> > > - Implement Undo, see
> > http://argouml.tigris.org/issues/show_bug.cgi?id=3497
> > > - Distributed models, see
> > http://argouml.tigris.org/issues/show_bug.cgi?id=3497
> > > - Positioning of text labels attached to fig lines, see
> > http://argouml.tigris.org/issues/show_bug.cgi?id=1048
> > > - New sequence diagram implementation, e.g. see
> > http://argouml.tigris.org/issues/show_bug.cgi?id=5030
> > >
> > > Of course, anyone is free to add more into the unstable 0.27.* releases,
> > but these four items should form a set of more organized subprojects, with
> > assigned team members and a better public communication (e.g. dev wiki).
> > >
> > > What do you think? You might argue about the exact set of features, but
> > will this approach bring ArgoUML to the next level?
> > >
> > > Regards,
> > > Thomas
> > > --
> > > Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen!
> > > Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx
> > > For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx
> > >
> > >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx
> > For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx
>
> --
> Psssst! Schon das coole Video vom GMX MultiMessenger gesehen?
> Der Eine für Alle: http://www.gmx.net/de/go/messenger03
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx
> For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx
>
>

Was this page helpful?

Thread at a glance:

Previous Message by Date: click to view message preview

Re: [argouml-dev] commitment for 0.28 features

Hi Bob, oh I see I gave the wrong link to the Undo issue, it is: http://argouml.tigris.org/issues/show_bug.cgi?id=1834 I know of the dependency of undo from euml. But if we really want to wait with undo until we move to euml, then the above issue has to be made dependent from an appropriate euml issue. I already proposed to do that step without introducing UML2 features, but I have not enough knowledge if it is possible to do that and thus get from the GSOC result to a stable euml ArgoUML. Anyway, I would say: the massive voting for Undo urges us to really start working on that! The problem with ArgoUML is, that we make releases without considering enough the users wishes. And votes are all we have to measure them. Having not enough developer resources is not an excuse for doing other things than requested, otherwise we should declare ArgoUML an internal tool for some hobbyists that by accident is freely downloadable. Thomas -------- Original-Nachricht -------- > Datum: Thu, 4 Sep 2008 17:40:57 +0200 > Von: "Bob Tarling" <bob.tarling@xxxxxxxxx> > An: dev@xxxxxxxxxxxxxxxxxx > Betreff: Re: [argouml-dev] commitment for 0.28 features > This needs some balance between how serious we are in moving toward > eclipse RCP. > > Why continue with our own Undo framework when that would be provided > by RCP. I'd hate to do lots of needless work that will just be thrown > away. > > Even if we did continue with our undo implementation it would be > dependant on eUML being complete I think. > > I'd like to make sure we do concentrate on seq diagrams but that does > not have to what for a full release. It could be just a module > release. > > Personally I think it is just a little too early to discuss these > plans in detail now. I'd like to wait till we have a dev wiki so that > we can start to build pages to agree designs for future. We can then > discuss what of those designs we would like to see soonest taking into > account both user votes and technical constraints. > > The wiki will hopefully make that process a lot easier than bouncing > emails around. > > Bob. > > On 04/09/2008, Thomas N. <thn-d@xxxxxx> wrote: > > Hi, > > > > let's agree on the plan for 0.28 as early as possible, because 0.26 > lacks of many important things. > > > > First, as we always encourage users to vote for issues, let's get > serious here and put the issues with the most votes on highest priority! These > are: Undo, distributed models, and movable labels of edges in diagrams. > Furthermore, we have a better sequence diagram implementation that is in need > of > a migration path from the current seq diagram impl. > > > > So, I propose to plan the next release to be released in a reasonable > short time (in great contrast to 0.26), with the following features: > > > > - Implement Undo, see > http://argouml.tigris.org/issues/show_bug.cgi?id=3497 > > - Distributed models, see > http://argouml.tigris.org/issues/show_bug.cgi?id=3497 > > - Positioning of text labels attached to fig lines, see > http://argouml.tigris.org/issues/show_bug.cgi?id=1048 > > - New sequence diagram implementation, e.g. see > http://argouml.tigris.org/issues/show_bug.cgi?id=5030 > > > > Of course, anyone is free to add more into the unstable 0.27.* releases, > but these four items should form a set of more organized subprojects, with > assigned team members and a better public communication (e.g. dev wiki). > > > > What do you think? You might argue about the exact set of features, but > will this approach bring ArgoUML to the next level? > > > > Regards, > > Thomas > > -- > > Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! > > Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx > > For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx > For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx -- Psssst! Schon das coole Video vom GMX MultiMessenger gesehen? Der Eine für Alle: http://www.gmx.net/de/go/messenger03

Next Message by Date: click to view message preview

Re: [argouml-dev] New developer - Brian Hudson

The prior e-mail was referring to me being granted committer status on ArgoEclipse. This e-mail is for ArgoUML.Thanks all for the warm welcome, looking forward to helping out.Brian On Fri, Sep 5, 2008 at 2:51 AM, Thomas N. <thn-d@xxxxxx> wrote: Welcome Brian! LOL, I thought I got crazy because of the strong feeling of having a deja vu, but luckily (for me) the mailing list archive clearly shows that you were already announced as committer in July 18. What's the reason for this, Tom? :-) Anyway, welcome again, Thomas -------- Original-Nachricht -------- > Datum: Thu, 4 Sep 2008 16:05:28 -0400 > Von: "Tom Morris" <tfmorris@xxxxxxxxx> > An: "ArgoUML Dev" <dev@xxxxxxxxxxxxxxxxxx> > CC: "Brian Hudson" <brian.r.hudson@xxxxxxxxx>, "Bogdan Ciprian Pistol" <bogdanciprian@xxxxxxxxx> > Betreff: [argouml-dev] New developer - Brian Hudson > Please join us in congratulating Brian Hudson on his new role as an > ArgoUML committer. Brian was a 2008 Google Summer of Code student for > ArgoEclipse, mentored by Bogdan Pistol, and has had full commit > privileges for the ArgoEclipse project since the middle of the summer. > He did a great job there and Linus has approved the recommendation by > Bogdan and I that he be granted commit privileges for ArgoUML as well. > This will allow him to directly commit refactorings and other > improvements to ArgoUML which support the modularization for reuse in > ArgoEclipse (after 0.26 is release, of course). > > Congratulations Brian! We look forward to your continuing contributions. > > Tom & Bogdan > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx > For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx -- GMX Kostenlose Spiele: Einfach online spielen und Spaß haben mit Pastry Passion! http://games.entertainment.gmx.net/de/entertainment/games/free/puzzle/6169196 --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx

Previous Message by Thread: click to view message preview

Next Message by Thread: click to view message preview

Re: [argouml-dev] commitment for 0.28 features

Hi all, On Fri, Sep 5, 2008 at 12:02 PM, Bob Tarling <bob.tarling@xxxxxxxxx> wrote: > I think it is more that eUML and undo are dependant on RCP. eUML uses > RCP but we currently have no roadmap of bringing RCP usage into core > ArgoUML. eUML does not depend on RCP. eUML uses Eclipse UML2 and Eclipse EMF which consists of a few jars that are self contained, there are no other references, in fact ArgoUML uses the eUML subsystem as is without any additional dependencies. More info about the eUML dependencies here http://euml.wikispaces.com/ IMO, Undo should not depend on just the eUML implementation, but it is required that the model provides undo facilities to the core ArgoUML, and MDR doesn't do this (as I recall from a past discussion). In my opinion, an agnostic undo interface that can work with eUML, MDR (even if MDR doesn't do this, stub implementations) or another future model subsystem could be a feasible solution. But the eUML implementation isn't complete yet, and this is another issue. Bogdan

Loading Comments...