Ideas, Ideas!

[Martin Marinschek <ma...@gmail.com>]

Hi everyone,

Google SoC is coming up again - you have been notified already. We
need project ideas - what does everyone want to have done for MyFaces,
Tobago or ADF Faces?

This is the existing list - nothing up for MyFaces so far:

http://wiki.apache.org/general/SummerOfCode2006

regards,

Martin

Re: Ideas, Ideas!

[Martin Marinschek <ma...@gmail.com>]

I think it would make a great component and a great SoC project. Sean,
you want to mentor someone?

regards,

Martin

On 4/20/06, Sean Schofield <se...@gmail.com> wrote:
> Paging is fine for some things but at times it seems excessive.  In
> our system we have a wizard where the user needs to assign a person to
> a document.  When you show the full 2000 users in the system its much
> slower then before we went to JSF.
>
> IMO the user shouldn't see a noticeably slower GUI b/c we've switched
> to JSF.  Paging is a PITA in this case.  I know the name I want, so
> why can't I sort by last name and scroll down to it (and do it fast?)
> Paging sucks in this example.
>
> Of course, this is just MO.
>
> Sean
>
>
> On 4/19/06, Werner Punz <we...@gmx.at> wrote:
> > Sean Schofield schrieb:
> > >> So, +1 for a SoC project on this component; but for sean it may come too late :)
> > >
> > > Yes too late for now but I could use it in the future.  I can think of
> > > many situations where this would be helpful.  IE 6.0 adds to the
> > > already slow process.  It only takes 500 + rows with a few columns of
> > > components and things get noticeably slower then without JSF.
> > >
> > Sean, I am not sure if a dynamic scroller really eases the burden that
> > much, it looks cool, but thats it,  in my opinion a real paginable data
> > model on top of the already existing jsf table is the better option, way
> > less burden on the server and in the end I think the response speed and
> > is way better and the load on the server way less.
> >
> > If you simply only load one page per time or do 5-6 chached pages the
> > results especially with an underlying db which allows proper pagination
> > (most do nowadays) are way better.
> >
> >
>


--

http://www.irian.at

Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German

Professional Support for Apache MyFaces

Re: Ideas, Ideas!

[Werner Punz <we...@gmx.at>]

Sean Schofield schrieb:
> Paging is fine for some things but at times it seems excessive.  In
> our system we have a wizard where the user needs to assign a person to
> a document.  When you show the full 2000 users in the system its much
> slower then before we went to JSF.
> 
Is paging really the issue, couldnt it be simply a deferred rendering,
due to datafetching before rendering?
Pushing the whole table into ajax might be the easier solution that way
you have the rest of the page rendered before the data is.


> IMO the user shouldn't see a noticeably slower GUI b/c we've switched
> to JSF.  Paging is a PITA in this case.  I know the name I want, so
> why can't I sort by last name and scroll down to it (and do it fast?) 
> Paging sucks in this example.
> 
Well I think you only will get this speed with such a control
if you can serve the data entirely from ram,
and yes once you can do that I agree it it faster, but once you hit the
db I even think this is way slower due to the bigger request burden it
puts onto the db.
Anyway having such a control, would be cool nevertheless ;-)

Re: Ideas, Ideas!

[Sean Schofield <se...@gmail.com>]

Paging is fine for some things but at times it seems excessive.  In
our system we have a wizard where the user needs to assign a person to
a document.  When you show the full 2000 users in the system its much
slower then before we went to JSF.

IMO the user shouldn't see a noticeably slower GUI b/c we've switched
to JSF.  Paging is a PITA in this case.  I know the name I want, so
why can't I sort by last name and scroll down to it (and do it fast?) 
Paging sucks in this example.

Of course, this is just MO.

Sean


On 4/19/06, Werner Punz <we...@gmx.at> wrote:
> Sean Schofield schrieb:
> >> So, +1 for a SoC project on this component; but for sean it may come too late :)
> >
> > Yes too late for now but I could use it in the future.  I can think of
> > many situations where this would be helpful.  IE 6.0 adds to the
> > already slow process.  It only takes 500 + rows with a few columns of
> > components and things get noticeably slower then without JSF.
> >
> Sean, I am not sure if a dynamic scroller really eases the burden that
> much, it looks cool, but thats it,  in my opinion a real paginable data
> model on top of the already existing jsf table is the better option, way
> less burden on the server and in the end I think the response speed and
> is way better and the load on the server way less.
>
> If you simply only load one page per time or do 5-6 chached pages the
> results especially with an underlying db which allows proper pagination
> (most do nowadays) are way better.
>
>

Re: Ideas, Ideas!

[Werner Punz <we...@gmx.at>]

Sean Schofield schrieb:
>> So, +1 for a SoC project on this component; but for sean it may come too late :)
> 
> Yes too late for now but I could use it in the future.  I can think of
> many situations where this would be helpful.  IE 6.0 adds to the
> already slow process.  It only takes 500 + rows with a few columns of
> components and things get noticeably slower then without JSF.
> 
Sean, I am not sure if a dynamic scroller really eases the burden that 
much, it looks cool, but thats it,  in my opinion a real paginable data 
model on top of the already existing jsf table is the better option, way 
less burden on the server and in the end I think the response speed and 
is way better and the load on the server way less.

If you simply only load one page per time or do 5-6 chached pages the 
results especially with an underlying db which allows proper pagination 
(most do nowadays) are way better.

Re: Ideas, Ideas!

[Sean Schofield <se...@gmail.com>]

> So, +1 for a SoC project on this component; but for sean it may come too late :)

Yes too late for now but I could use it in the future.  I can think of
many situations where this would be helpful.  IE 6.0 adds to the
already slow process.  It only takes 500 + rows with a few columns of
components and things get noticeably slower then without JSF.

Sean

Re: Ideas, Ideas!

[Gerald Müllan <bi...@gmail.com>]

> Is SoC how we ended up with Gerald?  A good thing btw ;-)

Yeah, right! :)

LiveGrid would be really cool, I often thought about a component like this.

Werner and me wanted to start working on this but because of the fact
that time is rare to all of us, we didn`t. And sure, we will not do it
in the near future..

So, +1 for a SoC project on this component; but for sean it may come too late :)

cheers,

Gerald

On 4/19/06, Sean Schofield <se...@gmail.com> wrote:
> Martin,
>
> Is SoC how we ended up with Gerald?  A good thing btw ;-)
>
> +1 on a live grid.  I think the ability to buffer a large sortable
> table with AJAX would be VERY handy.  We have a need for this right
> now on one of my client projects.
>
> Sean
>
> On 4/19/06, Martin Marinschek <ma...@gmail.com> wrote:
> > Ah, well, it depends on your style. You should in any case pick the
> > student with the best proposal.
> >
> > Last year interestingly the students I got interested in SoC had the
> > best proposals - a lot of the other proposals objectively weren't any
> > good. Very often it's just - hey, I want to work on this.
> >
> > regards,
> >
> > Martin
> >
> > On 4/19/06, Werner Punz <we...@gmx.at> wrote:
> > > Martin Marinschek schrieb:
> > >
> > > > P.S.: The work should be like 3 months - over the summer. You'll
> > > > probably have like 1 1/2 month core working time with the student, and
> > > > then before this there is getting up and running with open source,
> > > > JSF, MyFaces and whatever there is you have to learn until this is
> > > > possible at all; after the core time there is testing, bugfixing,
> > > > documentation, etc.
> > > >
> > > > For mentors, it depends on how much you want to invest. I had to
> > > > invest like two hours a week last summer.
> > > >
> > > Sounds good, although local stutents are preferrable for beer reasons I
> > > guess :-D
> > >
> > >
> >
> >
> > --
> >
> > http://www.irian.at
> >
> > Your JSF powerhouse -
> > JSF Consulting, Development and
> > Courses in English and German
> >
> > Professional Support for Apache MyFaces
> >
>


--
Gerald Müllan
Schelleingasse 2/11
1040 Vienna, Austria
0043 699 11772506
Bierbrauen@gmail.com

Re: Ideas, Ideas!

[Sean Schofield <se...@gmail.com>]

Martin,

Is SoC how we ended up with Gerald?  A good thing btw ;-)

+1 on a live grid.  I think the ability to buffer a large sortable
table with AJAX would be VERY handy.  We have a need for this right
now on one of my client projects.

Sean

On 4/19/06, Martin Marinschek <ma...@gmail.com> wrote:
> Ah, well, it depends on your style. You should in any case pick the
> student with the best proposal.
>
> Last year interestingly the students I got interested in SoC had the
> best proposals - a lot of the other proposals objectively weren't any
> good. Very often it's just - hey, I want to work on this.
>
> regards,
>
> Martin
>
> On 4/19/06, Werner Punz <we...@gmx.at> wrote:
> > Martin Marinschek schrieb:
> >
> > > P.S.: The work should be like 3 months - over the summer. You'll
> > > probably have like 1 1/2 month core working time with the student, and
> > > then before this there is getting up and running with open source,
> > > JSF, MyFaces and whatever there is you have to learn until this is
> > > possible at all; after the core time there is testing, bugfixing,
> > > documentation, etc.
> > >
> > > For mentors, it depends on how much you want to invest. I had to
> > > invest like two hours a week last summer.
> > >
> > Sounds good, although local stutents are preferrable for beer reasons I
> > guess :-D
> >
> >
>
>
> --
>
> http://www.irian.at
>
> Your JSF powerhouse -
> JSF Consulting, Development and
> Courses in English and German
>
> Professional Support for Apache MyFaces
>

Re: Ideas, Ideas!

[Martin Marinschek <ma...@gmail.com>]

Ah, well, it depends on your style. You should in any case pick the
student with the best proposal.

Last year interestingly the students I got interested in SoC had the
best proposals - a lot of the other proposals objectively weren't any
good. Very often it's just - hey, I want to work on this.

regards,

Martin

On 4/19/06, Werner Punz <we...@gmx.at> wrote:
> Martin Marinschek schrieb:
>
> > P.S.: The work should be like 3 months - over the summer. You'll
> > probably have like 1 1/2 month core working time with the student, and
> > then before this there is getting up and running with open source,
> > JSF, MyFaces and whatever there is you have to learn until this is
> > possible at all; after the core time there is testing, bugfixing,
> > documentation, etc.
> >
> > For mentors, it depends on how much you want to invest. I had to
> > invest like two hours a week last summer.
> >
> Sounds good, although local stutents are preferrable for beer reasons I
> guess :-D
>
>


--

http://www.irian.at

Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German

Professional Support for Apache MyFaces

Re: Ideas, Ideas!

[Werner Punz <we...@gmx.at>]

Martin Marinschek schrieb:

> P.S.: The work should be like 3 months - over the summer. You'll
> probably have like 1 1/2 month core working time with the student, and
> then before this there is getting up and running with open source,
> JSF, MyFaces and whatever there is you have to learn until this is
> possible at all; after the core time there is testing, bugfixing,
> documentation, etc.
> 
> For mentors, it depends on how much you want to invest. I had to
> invest like two hours a week last summer.
> 
Sounds good, although local stutents are preferrable for beer reasons I 
guess :-D

Re: Ideas, Ideas!

[Martin Marinschek <ma...@gmail.com>]

Ah, I'm sure we'll find students to do this.

Even if we don't have (known) students willing to do this from the
beginning, we'll get a lot of responses online later on.

I had like 6 requests for each of last years proposals.
Important is that the student proposal is well written, detailed and
clear. Without a good proposal, you don't stand a chance as a student
;)

regards,

Martin

P.S.: The work should be like 3 months - over the summer. You'll
probably have like 1 1/2 month core working time with the student, and
then before this there is getting up and running with open source,
JSF, MyFaces and whatever there is you have to learn until this is
possible at all; after the core time there is testing, bugfixing,
documentation, etc.

For mentors, it depends on how much you want to invest. I had to
invest like two hours a week last summer.

On 4/19/06, Matthias Wessendorf <ma...@apache.org> wrote:
> > "Merging the skinning systems of ADF Faces, Tobago and Tomahawk"
> >
> > "Merging the layouting system of Tobago with Tomahawk and ADF Faces"
>
> any of your students are willing to go ? :-)
>
> -Matthias
>
>


--

http://www.irian.at

Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German

Professional Support for Apache MyFaces

Re: Ideas, Ideas!

[Matthias Wessendorf <ma...@apache.org>]

> "Merging the skinning systems of ADF Faces, Tobago and Tomahawk"
>
> "Merging the layouting system of Tobago with Tomahawk and ADF Faces"

any of your students are willing to go ? :-)

-Matthias

Re: Ideas, Ideas!

[Werner Punz <we...@gmx.at>]

Martin Marinschek schrieb:
> Mario, Werner,
> 
> willing to mentoring this?
> 
Sure...

Re: Ideas, Ideas!

[Martin Marinschek <ma...@gmail.com>]

Mario, Werner,

willing to mentoring this?

want to sign up as mentors?

Would be great projects indeed!

I've also been thinking of

"Merging the skinning systems of ADF Faces, Tobago and Tomahawk"

(where tomahawk has none, of course)

or

"Merging the layouting system of Tobago with Tomahawk and ADF Faces"

regards,

Martin

On 4/19/06, Mario Ivankovits <ma...@ops.co.at> wrote:
> Werner Punz schrieb:
> > A seam like conversation system as tag
> >
> +infinite
> > a seam-less ejb3 opensessioninconversation filter built upon the
> > conversation system
> >
> +infinite
>
> :-D
>
> Ciao,
> Mario
>
>


--

http://www.irian.at

Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German

Professional Support for Apache MyFaces

Re: Ideas, Ideas!

[Mario Ivankovits <ma...@ops.co.at>]

Werner Punz schrieb:
> A seam like conversation system as tag
>   
+infinite
> a seam-less ejb3 opensessioninconversation filter built upon the
> conversation system
>   
+infinite

:-D

Ciao,
Mario

Re: Ideas, Ideas!

[Werner Punz <we...@gmx.at>]

Martin Marinschek schrieb:
> Hi everyone,
> 
> Google SoC is coming up again - you have been notified already. We
> need project ideas - what does everyone want to have done for MyFaces,
> Tobago or ADF Faces?
> 
> This is the existing list - nothing up for MyFaces so far:
> 
> http://wiki.apache.org/general/SummerOfCode2006
> 
JSF Implementation of some of the pending dojo components like floating
frame
etc...

...
Various data models hooking into JPA, Hibernate etc...

...
A seam like conversation system as tag
...
a seam-less ejb3 opensessioninconversation filter built upon the
conversation system

Re: Ideas, Ideas!

[Matthias Wessendorf <ma...@apache.org>]

> How about an implementation of a component equivalent to rico livegrid?
>
> http://openrico.org/rico/livegrid.page

something like that sounds very cool :-)

Re: Ideas, Ideas!

[Mike Kienenberger <mk...@gmail.com>]

On 4/19/06, Martin Marinschek <ma...@gmail.com> wrote:
> Google SoC is coming up again - you have been notified already. We
> need project ideas - what does everyone want to have done for MyFaces,
> Tobago or ADF Faces?

How about an implementation of a component equivalent to rico livegrid?

http://openrico.org/rico/livegrid.page


How about combining newsPaperColumn, extended data table, dataList,
etc.  into one component with multiple renderers?  (dataList may be
too much).


How about combining the navigation menus into one component and
multiple renderers?