You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@esme.apache.org by Ethan Jewett <es...@gmail.com> on 2009/07/27 17:02:20 UTC
UI vs. API vs. ESME server (was: Please take a look at)
It sounds like the project has been going down the road of decoupling the UI
from the backend for a while, resulting in a situation where we are pretty
close to having the ability to use a generic UI took like Atlas to build the
canonical front-end for ESME.
I assume that the interaction between the Atlas-built front-end and the ESME
server would be based on either the existing public API or a private API.
I'm not sure if this is the case, but maybe David can clarify.
Would it be of interest to the group to pursue a course of action in which
the ESME UI interacts only with publicly defined API methods (with the
possible exception of authentication)? This might help drive the evolution
of the API. It could also make it easier to have interchangeable UIs, so we
don't end up in UI limbo whenever a major change to the UI is proposed and
so that enterprise implementors can have an easier time deploying
custom(ized) UIs.
I'm going to start working on an ESME wrapper using YQL that will call the
API pretty soon now, so hopefully that will help drive the API development a
bit as well, but I figure the more people developing against the API the
better.
Ethan
On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
<yo...@gmail.com>wrote:
> I agree with David.
> If Altas really is as good as it looks like, I think we should check out
> the option of switching from jQuery to Cappuccino.
>
> I think we need to clarify David's question about using non-open tools in
> Apache-hosted projects first though.
>
> /Anne
>
>
>
> On 21. juli. 2009, at 18.55, David Pollak wrote:
>
> On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <hirsch.dick@gmail.com
>> >wrote:
>>
>> Question is when is the beta going to be released?
>>>
>>>
>> Dunno... but it does raise another issue... what's the Apache policy of
>> using non-open tools in order to build Apache-hosted projects? Atlas is
>> not
>> open.
>>
>>
>>
>>> We might not be able to wait until it is released. The UI (or absence
>>> thereof) is the main probelm at the current time.
>>>
>>
>>
>> Yeah.
>>
>>
>>
>>>
>>> Another issue might be that we are currently focused on JQuery but
>>> 280atlas is Cappuccino-based.
>>>
>>
>>
>> The weight of the jQuery dependency is very light... I'd opt for the cost
>> of
>> switching to Cappuccino if we had a killer GUI builder option.
>>
>>
>>
>>>
>>> D.
>>>
>>> On Tue, Jul 21, 2009 at 12:18 AM, David
>>> Pollak<fe...@gmail.com> wrote:
>>>
>>>> Folks,
>>>>
>>>> Please take a look at http://280atlas.com/
>>>>
>>>> Who has interest in building the ESME GUI with this tool?
>>>>
>>>> Thanks,
>>>>
>>>> David
>>>>
>>>> --
>>>> Lift, the simply functional web framework http://liftweb.net
>>>> Beginning Scala http://www.apress.com/book/view/1430219890
>>>> Follow me: http://twitter.com/dpp
>>>> Git some: http://github.com/dpp
>>>>
>>>>
>>>
>>
>>
>> --
>> Lift, the simply functional web framework http://liftweb.net
>> Beginning Scala http://www.apress.com/book/view/1430219890
>> Follow me: http://twitter.com/dpp
>> Git some: http://github.com/dpp
>>
>
>
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by Anne Kathrine Petterøe <yo...@gmail.com>.
+1 from me too.
I still think it is important that we finish the web UI we are
currently working on, so that we will have one UI "out of the box".
/Anne
On 27. juli. 2009, at 17.34, Mrinal Wadhwa wrote:
> Ethan,
>
> I am very much in favor of that approach ... I had written about
> this in an
> old thread as well ... were I talked about having the controller (in
> the MVC
> sense) of the UI be on the client and not on the server (which is the
> current case). Apart from the benefits you have already
> stated ...the UI
> becomes free to manage its own state and that enables a lot more
> flexibility
> in terms of what the UI can do.
> Also there are very few UI developers out there who know Scala or are
> willing to delve into it, this decoupling will enable any UI
> developer do
> build an ESME UI , as you mentioned this will be needed when
> enterprises try
> to customise ESME form their needs.
>
> Thanks,
> Mrinal
>
>
>
>
>
> On Mon, Jul 27, 2009 at 8:32 PM, Ethan Jewett <es...@gmail.com>
> wrote:
>
>> It sounds like the project has been going down the road of
>> decoupling the
>> UI
>> from the backend for a while, resulting in a situation where we are
>> pretty
>> close to having the ability to use a generic UI took like Atlas to
>> build
>> the
>> canonical front-end for ESME.
>> I assume that the interaction between the Atlas-built front-end and
>> the
>> ESME
>> server would be based on either the existing public API or a
>> private API.
>> I'm not sure if this is the case, but maybe David can clarify.
>>
>> Would it be of interest to the group to pursue a course of action
>> in which
>> the ESME UI interacts only with publicly defined API methods (with
>> the
>> possible exception of authentication)? This might help drive the
>> evolution
>> of the API. It could also make it easier to have interchangeable
>> UIs, so we
>> don't end up in UI limbo whenever a major change to the UI is
>> proposed and
>> so that enterprise implementors can have an easier time deploying
>> custom(ized) UIs.
>>
>> I'm going to start working on an ESME wrapper using YQL that will
>> call the
>> API pretty soon now, so hopefully that will help drive the API
>> development
>> a
>> bit as well, but I figure the more people developing against the
>> API the
>> better.
>>
>> Ethan
>>
>> On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
>> <yo...@gmail.com>wrote:
>>
>>> I agree with David.
>>> If Altas really is as good as it looks like, I think we should
>>> check out
>>> the option of switching from jQuery to Cappuccino.
>>>
>>> I think we need to clarify David's question about using non-open
>>> tools in
>>> Apache-hosted projects first though.
>>>
>>> /Anne
>>>
>>>
>>>
>>> On 21. juli. 2009, at 18.55, David Pollak wrote:
>>>
>>> On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <hirsch.dick@gmail.com
>>>>> wrote:
>>>>
>>>> Question is when is the beta going to be released?
>>>>>
>>>>>
>>>> Dunno... but it does raise another issue... what's the Apache
>>>> policy of
>>>> using non-open tools in order to build Apache-hosted projects?
>>>> Atlas is
>>>> not
>>>> open.
>>>>
>>>>
>>>>
>>>>> We might not be able to wait until it is released. The UI (or
>>>>> absence
>>>>> thereof) is the main probelm at the current time.
>>>>>
>>>>
>>>>
>>>> Yeah.
>>>>
>>>>
>>>>
>>>>>
>>>>> Another issue might be that we are currently focused on JQuery but
>>>>> 280atlas is Cappuccino-based.
>>>>>
>>>>
>>>>
>>>> The weight of the jQuery dependency is very light... I'd opt for
>>>> the
>> cost
>>>> of
>>>> switching to Cappuccino if we had a killer GUI builder option.
>>>>
>>>>
>>>>
>>>>>
>>>>> D.
>>>>>
>>>>> On Tue, Jul 21, 2009 at 12:18 AM, David
>>>>> Pollak<fe...@gmail.com> wrote:
>>>>>
>>>>>> Folks,
>>>>>>
>>>>>> Please take a look at http://280atlas.com/
>>>>>>
>>>>>> Who has interest in building the ESME GUI with this tool?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> David
>>>>>>
>>>>>> --
>>>>>> Lift, the simply functional web framework http://liftweb.net
>>>>>> Beginning Scala http://www.apress.com/book/view/1430219890
>>>>>> Follow me: http://twitter.com/dpp
>>>>>> Git some: http://github.com/dpp
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Lift, the simply functional web framework http://liftweb.net
>>>> Beginning Scala http://www.apress.com/book/view/1430219890
>>>> Follow me: http://twitter.com/dpp
>>>> Git some: http://github.com/dpp
>>>>
>>>
>>>
>>
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by Mrinal Wadhwa <mr...@gmail.com>.
Ethan,
I am very much in favor of that approach ... I had written about this in an
old thread as well ... were I talked about having the controller (in the MVC
sense) of the UI be on the client and not on the server (which is the
current case). Apart from the benefits you have already stated ...the UI
becomes free to manage its own state and that enables a lot more flexibility
in terms of what the UI can do.
Also there are very few UI developers out there who know Scala or are
willing to delve into it, this decoupling will enable any UI developer do
build an ESME UI , as you mentioned this will be needed when enterprises try
to customise ESME form their needs.
Thanks,
Mrinal
On Mon, Jul 27, 2009 at 8:32 PM, Ethan Jewett <es...@gmail.com> wrote:
> It sounds like the project has been going down the road of decoupling the
> UI
> from the backend for a while, resulting in a situation where we are pretty
> close to having the ability to use a generic UI took like Atlas to build
> the
> canonical front-end for ESME.
> I assume that the interaction between the Atlas-built front-end and the
> ESME
> server would be based on either the existing public API or a private API.
> I'm not sure if this is the case, but maybe David can clarify.
>
> Would it be of interest to the group to pursue a course of action in which
> the ESME UI interacts only with publicly defined API methods (with the
> possible exception of authentication)? This might help drive the evolution
> of the API. It could also make it easier to have interchangeable UIs, so we
> don't end up in UI limbo whenever a major change to the UI is proposed and
> so that enterprise implementors can have an easier time deploying
> custom(ized) UIs.
>
> I'm going to start working on an ESME wrapper using YQL that will call the
> API pretty soon now, so hopefully that will help drive the API development
> a
> bit as well, but I figure the more people developing against the API the
> better.
>
> Ethan
>
> On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
> <yo...@gmail.com>wrote:
>
> > I agree with David.
> > If Altas really is as good as it looks like, I think we should check out
> > the option of switching from jQuery to Cappuccino.
> >
> > I think we need to clarify David's question about using non-open tools in
> > Apache-hosted projects first though.
> >
> > /Anne
> >
> >
> >
> > On 21. juli. 2009, at 18.55, David Pollak wrote:
> >
> > On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <hirsch.dick@gmail.com
> >> >wrote:
> >>
> >> Question is when is the beta going to be released?
> >>>
> >>>
> >> Dunno... but it does raise another issue... what's the Apache policy of
> >> using non-open tools in order to build Apache-hosted projects? Atlas is
> >> not
> >> open.
> >>
> >>
> >>
> >>> We might not be able to wait until it is released. The UI (or absence
> >>> thereof) is the main probelm at the current time.
> >>>
> >>
> >>
> >> Yeah.
> >>
> >>
> >>
> >>>
> >>> Another issue might be that we are currently focused on JQuery but
> >>> 280atlas is Cappuccino-based.
> >>>
> >>
> >>
> >> The weight of the jQuery dependency is very light... I'd opt for the
> cost
> >> of
> >> switching to Cappuccino if we had a killer GUI builder option.
> >>
> >>
> >>
> >>>
> >>> D.
> >>>
> >>> On Tue, Jul 21, 2009 at 12:18 AM, David
> >>> Pollak<fe...@gmail.com> wrote:
> >>>
> >>>> Folks,
> >>>>
> >>>> Please take a look at http://280atlas.com/
> >>>>
> >>>> Who has interest in building the ESME GUI with this tool?
> >>>>
> >>>> Thanks,
> >>>>
> >>>> David
> >>>>
> >>>> --
> >>>> Lift, the simply functional web framework http://liftweb.net
> >>>> Beginning Scala http://www.apress.com/book/view/1430219890
> >>>> Follow me: http://twitter.com/dpp
> >>>> Git some: http://github.com/dpp
> >>>>
> >>>>
> >>>
> >>
> >>
> >> --
> >> Lift, the simply functional web framework http://liftweb.net
> >> Beginning Scala http://www.apress.com/book/view/1430219890
> >> Follow me: http://twitter.com/dpp
> >> Git some: http://github.com/dpp
> >>
> >
> >
>
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by Vassil Dichev <vd...@apache.org>.
>> I wasn't aware we already had calls for conversations, actions and pools
>> ...
>
>
> There should be calls for conversations and actions. Pools are new, so
> there may not be calls for those.
The RESTful API calls for adding a pool, adding a user to a pool and
sending a message to a pool were added even before the WebUI. Then
while I improved the UI, the RESTful API got a bit behind, so while I
added ways to list the pools, I haven't yet added a method to list
pools for the RESTful API. It's trivial to implement, so I can add it
during the next days.
I think currently the big differentiator is authentication. It's not
easy to use OpenID from a desktop client so we need the tokens.
Vassil
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by Mrinal Wadhwa <mr...@gmail.com>.
thanks
On Wed, Jul 29, 2009 at 5:23 AM, David Pollak <feeder.of.the.bears@gmail.com
> wrote:
> On Tue, Jul 28, 2009 at 4:51 PM, Mrinal Wadhwa <mrinal.wadhwa@gmail.com
> >wrote:
>
> > David,
> >
> > I wasn't aware we already had calls for conversations, actions and pools
> > ...
>
>
> There should be calls for conversations and actions. Pools are new, so
> there may not be calls for those.
>
>
> >
> > just saw them, are these fully functional, or in progress?
> >
> > Thanks
> > Mrinal
> >
> >
> >
> >
> >
> > On Wed, Jul 29, 2009 at 5:10 AM, David Pollak <
> > feeder.of.the.bears@gmail.com
> > > wrote:
> >
> > > Ethan,
> > >
> > > We've had an Air client since day 1 and have all the APIs to support
> the
> > > AIR
> > > client (other than user authentication). What APIs are missing for
> your
> > > YQL
> > > version?
> > >
> > > Thanks,
> > >
> > > David
> > >
> > > On Mon, Jul 27, 2009 at 8:02 AM, Ethan Jewett <es...@gmail.com>
> > wrote:
> > >
> > > > It sounds like the project has been going down the road of decoupling
> > the
> > > > UI
> > > > from the backend for a while, resulting in a situation where we are
> > > pretty
> > > > close to having the ability to use a generic UI took like Atlas to
> > build
> > > > the
> > > > canonical front-end for ESME.
> > > > I assume that the interaction between the Atlas-built front-end and
> the
> > > > ESME
> > > > server would be based on either the existing public API or a private
> > API.
> > > > I'm not sure if this is the case, but maybe David can clarify.
> > > >
> > > > Would it be of interest to the group to pursue a course of action in
> > > which
> > > > the ESME UI interacts only with publicly defined API methods (with
> the
> > > > possible exception of authentication)? This might help drive the
> > > evolution
> > > > of the API. It could also make it easier to have interchangeable UIs,
> > so
> > > we
> > > > don't end up in UI limbo whenever a major change to the UI is
> proposed
> > > and
> > > > so that enterprise implementors can have an easier time deploying
> > > > custom(ized) UIs.
> > > >
> > > > I'm going to start working on an ESME wrapper using YQL that will
> call
> > > the
> > > > API pretty soon now, so hopefully that will help drive the API
> > > development
> > > > a
> > > > bit as well, but I figure the more people developing against the API
> > the
> > > > better.
> > > >
> > > > Ethan
> > > >
> > > > On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
> > > > <yo...@gmail.com>wrote:
> > > >
> > > > > I agree with David.
> > > > > If Altas really is as good as it looks like, I think we should
> check
> > > out
> > > > > the option of switching from jQuery to Cappuccino.
> > > > >
> > > > > I think we need to clarify David's question about using non-open
> > tools
> > > in
> > > > > Apache-hosted projects first though.
> > > > >
> > > > > /Anne
> > > > >
> > > > >
> > > > >
> > > > > On 21. juli. 2009, at 18.55, David Pollak wrote:
> > > > >
> > > > > On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <
> > > hirsch.dick@gmail.com
> > > > >> >wrote:
> > > > >>
> > > > >> Question is when is the beta going to be released?
> > > > >>>
> > > > >>>
> > > > >> Dunno... but it does raise another issue... what's the Apache
> policy
> > > of
> > > > >> using non-open tools in order to build Apache-hosted projects?
> > Atlas
> > > is
> > > > >> not
> > > > >> open.
> > > > >>
> > > > >>
> > > > >>
> > > > >>> We might not be able to wait until it is released. The UI (or
> > absence
> > > > >>> thereof) is the main probelm at the current time.
> > > > >>>
> > > > >>
> > > > >>
> > > > >> Yeah.
> > > > >>
> > > > >>
> > > > >>
> > > > >>>
> > > > >>> Another issue might be that we are currently focused on JQuery
> but
> > > > >>> 280atlas is Cappuccino-based.
> > > > >>>
> > > > >>
> > > > >>
> > > > >> The weight of the jQuery dependency is very light... I'd opt for
> the
> > > > cost
> > > > >> of
> > > > >> switching to Cappuccino if we had a killer GUI builder option.
> > > > >>
> > > > >>
> > > > >>
> > > > >>>
> > > > >>> D.
> > > > >>>
> > > > >>> On Tue, Jul 21, 2009 at 12:18 AM, David
> > > > >>> Pollak<fe...@gmail.com> wrote:
> > > > >>>
> > > > >>>> Folks,
> > > > >>>>
> > > > >>>> Please take a look at http://280atlas.com/
> > > > >>>>
> > > > >>>> Who has interest in building the ESME GUI with this tool?
> > > > >>>>
> > > > >>>> Thanks,
> > > > >>>>
> > > > >>>> David
> > > > >>>>
> > > > >>>> --
> > > > >>>> Lift, the simply functional web framework http://liftweb.net
> > > > >>>> Beginning Scala http://www.apress.com/book/view/1430219890
> > > > >>>> Follow me: http://twitter.com/dpp
> > > > >>>> Git some: http://github.com/dpp
> > > > >>>>
> > > > >>>>
> > > > >>>
> > > > >>
> > > > >>
> > > > >> --
> > > > >> Lift, the simply functional web framework http://liftweb.net
> > > > >> Beginning Scala http://www.apress.com/book/view/1430219890
> > > > >> Follow me: http://twitter.com/dpp
> > > > >> Git some: http://github.com/dpp
> > > > >>
> > > > >
> > > > >
> > > >
> > >
> > >
> > >
> > > --
> > > Lift, the simply functional web framework http://liftweb.net
> > > Beginning Scala http://www.apress.com/book/view/1430219890
> > > Follow me: http://twitter.com/dpp
> > > Git some: http://github.com/dpp
> > >
> >
>
>
>
> --
> Lift, the simply functional web framework http://liftweb.net
> Beginning Scala http://www.apress.com/book/view/1430219890
> Follow me: http://twitter.com/dpp
> Git some: http://github.com/dpp
>
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by David Pollak <fe...@gmail.com>.
On Tue, Jul 28, 2009 at 4:51 PM, Mrinal Wadhwa <mr...@gmail.com>wrote:
> David,
>
> I wasn't aware we already had calls for conversations, actions and pools
> ...
There should be calls for conversations and actions. Pools are new, so
there may not be calls for those.
>
> just saw them, are these fully functional, or in progress?
>
> Thanks
> Mrinal
>
>
>
>
>
> On Wed, Jul 29, 2009 at 5:10 AM, David Pollak <
> feeder.of.the.bears@gmail.com
> > wrote:
>
> > Ethan,
> >
> > We've had an Air client since day 1 and have all the APIs to support the
> > AIR
> > client (other than user authentication). What APIs are missing for your
> > YQL
> > version?
> >
> > Thanks,
> >
> > David
> >
> > On Mon, Jul 27, 2009 at 8:02 AM, Ethan Jewett <es...@gmail.com>
> wrote:
> >
> > > It sounds like the project has been going down the road of decoupling
> the
> > > UI
> > > from the backend for a while, resulting in a situation where we are
> > pretty
> > > close to having the ability to use a generic UI took like Atlas to
> build
> > > the
> > > canonical front-end for ESME.
> > > I assume that the interaction between the Atlas-built front-end and the
> > > ESME
> > > server would be based on either the existing public API or a private
> API.
> > > I'm not sure if this is the case, but maybe David can clarify.
> > >
> > > Would it be of interest to the group to pursue a course of action in
> > which
> > > the ESME UI interacts only with publicly defined API methods (with the
> > > possible exception of authentication)? This might help drive the
> > evolution
> > > of the API. It could also make it easier to have interchangeable UIs,
> so
> > we
> > > don't end up in UI limbo whenever a major change to the UI is proposed
> > and
> > > so that enterprise implementors can have an easier time deploying
> > > custom(ized) UIs.
> > >
> > > I'm going to start working on an ESME wrapper using YQL that will call
> > the
> > > API pretty soon now, so hopefully that will help drive the API
> > development
> > > a
> > > bit as well, but I figure the more people developing against the API
> the
> > > better.
> > >
> > > Ethan
> > >
> > > On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
> > > <yo...@gmail.com>wrote:
> > >
> > > > I agree with David.
> > > > If Altas really is as good as it looks like, I think we should check
> > out
> > > > the option of switching from jQuery to Cappuccino.
> > > >
> > > > I think we need to clarify David's question about using non-open
> tools
> > in
> > > > Apache-hosted projects first though.
> > > >
> > > > /Anne
> > > >
> > > >
> > > >
> > > > On 21. juli. 2009, at 18.55, David Pollak wrote:
> > > >
> > > > On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <
> > hirsch.dick@gmail.com
> > > >> >wrote:
> > > >>
> > > >> Question is when is the beta going to be released?
> > > >>>
> > > >>>
> > > >> Dunno... but it does raise another issue... what's the Apache policy
> > of
> > > >> using non-open tools in order to build Apache-hosted projects?
> Atlas
> > is
> > > >> not
> > > >> open.
> > > >>
> > > >>
> > > >>
> > > >>> We might not be able to wait until it is released. The UI (or
> absence
> > > >>> thereof) is the main probelm at the current time.
> > > >>>
> > > >>
> > > >>
> > > >> Yeah.
> > > >>
> > > >>
> > > >>
> > > >>>
> > > >>> Another issue might be that we are currently focused on JQuery but
> > > >>> 280atlas is Cappuccino-based.
> > > >>>
> > > >>
> > > >>
> > > >> The weight of the jQuery dependency is very light... I'd opt for the
> > > cost
> > > >> of
> > > >> switching to Cappuccino if we had a killer GUI builder option.
> > > >>
> > > >>
> > > >>
> > > >>>
> > > >>> D.
> > > >>>
> > > >>> On Tue, Jul 21, 2009 at 12:18 AM, David
> > > >>> Pollak<fe...@gmail.com> wrote:
> > > >>>
> > > >>>> Folks,
> > > >>>>
> > > >>>> Please take a look at http://280atlas.com/
> > > >>>>
> > > >>>> Who has interest in building the ESME GUI with this tool?
> > > >>>>
> > > >>>> Thanks,
> > > >>>>
> > > >>>> David
> > > >>>>
> > > >>>> --
> > > >>>> Lift, the simply functional web framework http://liftweb.net
> > > >>>> Beginning Scala http://www.apress.com/book/view/1430219890
> > > >>>> Follow me: http://twitter.com/dpp
> > > >>>> Git some: http://github.com/dpp
> > > >>>>
> > > >>>>
> > > >>>
> > > >>
> > > >>
> > > >> --
> > > >> Lift, the simply functional web framework http://liftweb.net
> > > >> Beginning Scala http://www.apress.com/book/view/1430219890
> > > >> Follow me: http://twitter.com/dpp
> > > >> Git some: http://github.com/dpp
> > > >>
> > > >
> > > >
> > >
> >
> >
> >
> > --
> > Lift, the simply functional web framework http://liftweb.net
> > Beginning Scala http://www.apress.com/book/view/1430219890
> > Follow me: http://twitter.com/dpp
> > Git some: http://github.com/dpp
> >
>
--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Git some: http://github.com/dpp
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by Mrinal Wadhwa <mr...@gmail.com>.
David,
I wasn't aware we already had calls for conversations, actions and pools ...
just saw them, are these fully functional, or in progress?
Thanks
Mrinal
On Wed, Jul 29, 2009 at 5:10 AM, David Pollak <feeder.of.the.bears@gmail.com
> wrote:
> Ethan,
>
> We've had an Air client since day 1 and have all the APIs to support the
> AIR
> client (other than user authentication). What APIs are missing for your
> YQL
> version?
>
> Thanks,
>
> David
>
> On Mon, Jul 27, 2009 at 8:02 AM, Ethan Jewett <es...@gmail.com> wrote:
>
> > It sounds like the project has been going down the road of decoupling the
> > UI
> > from the backend for a while, resulting in a situation where we are
> pretty
> > close to having the ability to use a generic UI took like Atlas to build
> > the
> > canonical front-end for ESME.
> > I assume that the interaction between the Atlas-built front-end and the
> > ESME
> > server would be based on either the existing public API or a private API.
> > I'm not sure if this is the case, but maybe David can clarify.
> >
> > Would it be of interest to the group to pursue a course of action in
> which
> > the ESME UI interacts only with publicly defined API methods (with the
> > possible exception of authentication)? This might help drive the
> evolution
> > of the API. It could also make it easier to have interchangeable UIs, so
> we
> > don't end up in UI limbo whenever a major change to the UI is proposed
> and
> > so that enterprise implementors can have an easier time deploying
> > custom(ized) UIs.
> >
> > I'm going to start working on an ESME wrapper using YQL that will call
> the
> > API pretty soon now, so hopefully that will help drive the API
> development
> > a
> > bit as well, but I figure the more people developing against the API the
> > better.
> >
> > Ethan
> >
> > On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
> > <yo...@gmail.com>wrote:
> >
> > > I agree with David.
> > > If Altas really is as good as it looks like, I think we should check
> out
> > > the option of switching from jQuery to Cappuccino.
> > >
> > > I think we need to clarify David's question about using non-open tools
> in
> > > Apache-hosted projects first though.
> > >
> > > /Anne
> > >
> > >
> > >
> > > On 21. juli. 2009, at 18.55, David Pollak wrote:
> > >
> > > On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <
> hirsch.dick@gmail.com
> > >> >wrote:
> > >>
> > >> Question is when is the beta going to be released?
> > >>>
> > >>>
> > >> Dunno... but it does raise another issue... what's the Apache policy
> of
> > >> using non-open tools in order to build Apache-hosted projects? Atlas
> is
> > >> not
> > >> open.
> > >>
> > >>
> > >>
> > >>> We might not be able to wait until it is released. The UI (or absence
> > >>> thereof) is the main probelm at the current time.
> > >>>
> > >>
> > >>
> > >> Yeah.
> > >>
> > >>
> > >>
> > >>>
> > >>> Another issue might be that we are currently focused on JQuery but
> > >>> 280atlas is Cappuccino-based.
> > >>>
> > >>
> > >>
> > >> The weight of the jQuery dependency is very light... I'd opt for the
> > cost
> > >> of
> > >> switching to Cappuccino if we had a killer GUI builder option.
> > >>
> > >>
> > >>
> > >>>
> > >>> D.
> > >>>
> > >>> On Tue, Jul 21, 2009 at 12:18 AM, David
> > >>> Pollak<fe...@gmail.com> wrote:
> > >>>
> > >>>> Folks,
> > >>>>
> > >>>> Please take a look at http://280atlas.com/
> > >>>>
> > >>>> Who has interest in building the ESME GUI with this tool?
> > >>>>
> > >>>> Thanks,
> > >>>>
> > >>>> David
> > >>>>
> > >>>> --
> > >>>> Lift, the simply functional web framework http://liftweb.net
> > >>>> Beginning Scala http://www.apress.com/book/view/1430219890
> > >>>> Follow me: http://twitter.com/dpp
> > >>>> Git some: http://github.com/dpp
> > >>>>
> > >>>>
> > >>>
> > >>
> > >>
> > >> --
> > >> Lift, the simply functional web framework http://liftweb.net
> > >> Beginning Scala http://www.apress.com/book/view/1430219890
> > >> Follow me: http://twitter.com/dpp
> > >> Git some: http://github.com/dpp
> > >>
> > >
> > >
> >
>
>
>
> --
> Lift, the simply functional web framework http://liftweb.net
> Beginning Scala http://www.apress.com/book/view/1430219890
> Follow me: http://twitter.com/dpp
> Git some: http://github.com/dpp
>
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by Ethan Jewett <es...@gmail.com>.
Hey David,
I don't think anything is missing for the YQL client, but I have the feeling
that as more people start developing on top of the API directly we'll start
to see ways to improve it, be that to make it more RESTful as was discussed
previously, or something else we haven't thought of yet. I'm just hopeful
that use will drive development :-)
Ethan
On Tue, Jul 28, 2009 at 7:40 PM, David Pollak <feeder.of.the.bears@gmail.com
> wrote:
> Ethan,
>
> We've had an Air client since day 1 and have all the APIs to support the
> AIR
> client (other than user authentication). What APIs are missing for your
> YQL
> version?
>
> Thanks,
>
> David
>
> On Mon, Jul 27, 2009 at 8:02 AM, Ethan Jewett <es...@gmail.com> wrote:
>
> > It sounds like the project has been going down the road of decoupling the
> > UI
> > from the backend for a while, resulting in a situation where we are
> pretty
> > close to having the ability to use a generic UI took like Atlas to build
> > the
> > canonical front-end for ESME.
> > I assume that the interaction between the Atlas-built front-end and the
> > ESME
> > server would be based on either the existing public API or a private API.
> > I'm not sure if this is the case, but maybe David can clarify.
> >
> > Would it be of interest to the group to pursue a course of action in
> which
> > the ESME UI interacts only with publicly defined API methods (with the
> > possible exception of authentication)? This might help drive the
> evolution
> > of the API. It could also make it easier to have interchangeable UIs, so
> we
> > don't end up in UI limbo whenever a major change to the UI is proposed
> and
> > so that enterprise implementors can have an easier time deploying
> > custom(ized) UIs.
> >
> > I'm going to start working on an ESME wrapper using YQL that will call
> the
> > API pretty soon now, so hopefully that will help drive the API
> development
> > a
> > bit as well, but I figure the more people developing against the API the
> > better.
> >
> > Ethan
> >
> > On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
> > <yo...@gmail.com>wrote:
> >
> > > I agree with David.
> > > If Altas really is as good as it looks like, I think we should check
> out
> > > the option of switching from jQuery to Cappuccino.
> > >
> > > I think we need to clarify David's question about using non-open tools
> in
> > > Apache-hosted projects first though.
> > >
> > > /Anne
> > >
> > >
> > >
> > > On 21. juli. 2009, at 18.55, David Pollak wrote:
> > >
> > > On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <
> hirsch.dick@gmail.com
> > >> >wrote:
> > >>
> > >> Question is when is the beta going to be released?
> > >>>
> > >>>
> > >> Dunno... but it does raise another issue... what's the Apache policy
> of
> > >> using non-open tools in order to build Apache-hosted projects? Atlas
> is
> > >> not
> > >> open.
> > >>
> > >>
> > >>
> > >>> We might not be able to wait until it is released. The UI (or absence
> > >>> thereof) is the main probelm at the current time.
> > >>>
> > >>
> > >>
> > >> Yeah.
> > >>
> > >>
> > >>
> > >>>
> > >>> Another issue might be that we are currently focused on JQuery but
> > >>> 280atlas is Cappuccino-based.
> > >>>
> > >>
> > >>
> > >> The weight of the jQuery dependency is very light... I'd opt for the
> > cost
> > >> of
> > >> switching to Cappuccino if we had a killer GUI builder option.
> > >>
> > >>
> > >>
> > >>>
> > >>> D.
> > >>>
> > >>> On Tue, Jul 21, 2009 at 12:18 AM, David
> > >>> Pollak<fe...@gmail.com> wrote:
> > >>>
> > >>>> Folks,
> > >>>>
> > >>>> Please take a look at http://280atlas.com/
> > >>>>
> > >>>> Who has interest in building the ESME GUI with this tool?
> > >>>>
> > >>>> Thanks,
> > >>>>
> > >>>> David
> > >>>>
> > >>>> --
> > >>>> Lift, the simply functional web framework http://liftweb.net
> > >>>> Beginning Scala http://www.apress.com/book/view/1430219890
> > >>>> Follow me: http://twitter.com/dpp
> > >>>> Git some: http://github.com/dpp
> > >>>>
> > >>>>
> > >>>
> > >>
> > >>
> > >> --
> > >> Lift, the simply functional web framework http://liftweb.net
> > >> Beginning Scala http://www.apress.com/book/view/1430219890
> > >> Follow me: http://twitter.com/dpp
> > >> Git some: http://github.com/dpp
> > >>
> > >
> > >
> >
>
>
>
> --
> Lift, the simply functional web framework http://liftweb.net
> Beginning Scala http://www.apress.com/book/view/1430219890
> Follow me: http://twitter.com/dpp
> Git some: http://github.com/dpp
>
Re: UI vs. API vs. ESME server (was: Please take a look at)
Posted by David Pollak <fe...@gmail.com>.
Ethan,
We've had an Air client since day 1 and have all the APIs to support the AIR
client (other than user authentication). What APIs are missing for your YQL
version?
Thanks,
David
On Mon, Jul 27, 2009 at 8:02 AM, Ethan Jewett <es...@gmail.com> wrote:
> It sounds like the project has been going down the road of decoupling the
> UI
> from the backend for a while, resulting in a situation where we are pretty
> close to having the ability to use a generic UI took like Atlas to build
> the
> canonical front-end for ESME.
> I assume that the interaction between the Atlas-built front-end and the
> ESME
> server would be based on either the existing public API or a private API.
> I'm not sure if this is the case, but maybe David can clarify.
>
> Would it be of interest to the group to pursue a course of action in which
> the ESME UI interacts only with publicly defined API methods (with the
> possible exception of authentication)? This might help drive the evolution
> of the API. It could also make it easier to have interchangeable UIs, so we
> don't end up in UI limbo whenever a major change to the UI is proposed and
> so that enterprise implementors can have an easier time deploying
> custom(ized) UIs.
>
> I'm going to start working on an ESME wrapper using YQL that will call the
> API pretty soon now, so hopefully that will help drive the API development
> a
> bit as well, but I figure the more people developing against the API the
> better.
>
> Ethan
>
> On Mon, Jul 27, 2009 at 5:19 AM, Anne Kathrine Petterøe
> <yo...@gmail.com>wrote:
>
> > I agree with David.
> > If Altas really is as good as it looks like, I think we should check out
> > the option of switching from jQuery to Cappuccino.
> >
> > I think we need to clarify David's question about using non-open tools in
> > Apache-hosted projects first though.
> >
> > /Anne
> >
> >
> >
> > On 21. juli. 2009, at 18.55, David Pollak wrote:
> >
> > On Tue, Jul 21, 2009 at 1:11 PM, Richard Hirsch <hirsch.dick@gmail.com
> >> >wrote:
> >>
> >> Question is when is the beta going to be released?
> >>>
> >>>
> >> Dunno... but it does raise another issue... what's the Apache policy of
> >> using non-open tools in order to build Apache-hosted projects? Atlas is
> >> not
> >> open.
> >>
> >>
> >>
> >>> We might not be able to wait until it is released. The UI (or absence
> >>> thereof) is the main probelm at the current time.
> >>>
> >>
> >>
> >> Yeah.
> >>
> >>
> >>
> >>>
> >>> Another issue might be that we are currently focused on JQuery but
> >>> 280atlas is Cappuccino-based.
> >>>
> >>
> >>
> >> The weight of the jQuery dependency is very light... I'd opt for the
> cost
> >> of
> >> switching to Cappuccino if we had a killer GUI builder option.
> >>
> >>
> >>
> >>>
> >>> D.
> >>>
> >>> On Tue, Jul 21, 2009 at 12:18 AM, David
> >>> Pollak<fe...@gmail.com> wrote:
> >>>
> >>>> Folks,
> >>>>
> >>>> Please take a look at http://280atlas.com/
> >>>>
> >>>> Who has interest in building the ESME GUI with this tool?
> >>>>
> >>>> Thanks,
> >>>>
> >>>> David
> >>>>
> >>>> --
> >>>> Lift, the simply functional web framework http://liftweb.net
> >>>> Beginning Scala http://www.apress.com/book/view/1430219890
> >>>> Follow me: http://twitter.com/dpp
> >>>> Git some: http://github.com/dpp
> >>>>
> >>>>
> >>>
> >>
> >>
> >> --
> >> Lift, the simply functional web framework http://liftweb.net
> >> Beginning Scala http://www.apress.com/book/view/1430219890
> >> Follow me: http://twitter.com/dpp
> >> Git some: http://github.com/dpp
> >>
> >
> >
>
--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Git some: http://github.com/dpp