gmane.comp.web.zope.plone.devel

Re: It's time to talk Plone 5

Philip Bauer — 2012-05-25T09:53:18

Am 25.05.2012 um 07:07 schrieb Dylan Jay:

At Starzel.de we are using plone.app.contentypes for all our new projects and are very happy with our dexterity-only websites. It's really nice using behaviors to extend the default types provided by p.a.ct. I think migration-steps from ATContentTypes to p.a.ct might be possible but are a lot of work and will break once existing types are customized with schemaextender. I would not dream about using p.a.ct in sites that have a lot of existing content.

I wrote an email to the co-developers of plone.app.contentypes a while ago in which I made the assumption that since Deco will provide the "next content-type" p.a.ct will never become part of the plone-core and will continue being a tool for developers. Therefore we should go ahead with moving the types from pure xml to interface-driven types and the views to browser-views. This would give developers much more freedom to work with these types effectively.

I still don't think we could make p.a.ct more than an option you can choose when installing or an extension. The pros simply don't outweight the cons.

On the other hand my assumption could have been wrong since many don't trust that Deco will be production-ready in time for Plone 5 (whenever that will be). To prevent this from happening we put aside enough money from the earnings of the Plone-Konferenz in Munich to sponsor two Sprints devoted to Deco but haven't had the time yet to kick that off. If anyone is interested in such a sprint please contact me.

Philip

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Jean-Michel FRANCOIS — 2012-05-25T08:00:30

I remember this from bristol conferences:

"Evolution is better than revolution"

I guess we should add diazo and dexterity to core as addons but not
activated and let integrator do the rest.

And it should be the same for every addons (deco, plone.app.event, ...)

Remember plone.app.caching, plone.app.theming, I have used theses addons
from their first release, make some fixes and just see it in action.

I would like to use deco as released addon before seeing it in the plone
core.

Regards / Cordialement,
JeanMichel FRANCOIS
Find me on Twitter <http://twitter.com/toutpt> / Retrouvez moi sur
Twitter<http://twitter.com/toutpt_nantes>




2012/5/25 Armin Stroß-Radschinski <developer-itUlVD6B3SM< at >public.gmane.org>

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers

Re: It's time to talk Plone 5

Armin Stroß-Radschinski — 2012-05-25T05:58:01

+1
Am 25.05.2012 um 01:38 schrieb Eric Steele:


There are big and happy Plone 3 based installations out there. With  
Plone 5 we do not need to make people with Plone 4 based applications  
happy at once, because no project manager will switch to Plone 5 when  
there is no need. Staying at 4 until migration works and starting with  
5 where it makes sense is always very clever. Version buzz is a  
(vegetative) brain problem. Ignoring Dexterity with working AT can  
bring you to 5 very easy (even if it is useless). We need Plone 5 to  
be leading edge but with an early and serious knowledge where the  
migration traps are! This can be the ground for good communication!  
Avoid the PR disaster Typo3 has now with Phoenix.

We should encourage some singular people / projects with important add  
ons to take the pain and create working migrations to the new  
scenario. Everybody should help to get this done fast and with serious  
feedback and documentation. Then the results should help to estimate  
the amount of work (even if it is not promising) and have the results  
influence the further process very early. This may be far more  
effective than having all the developers working in parallel on their  
own stuff and ending up all of them with crap after too much time.

Even a good scenario test with coexisting and internationalized  
Dexterity and AT-Content will make sense as well.

Please dear native english speaking developers: there are  
internationalized sites out there! During always working and talking  
in your mothertongue the sense for the bigger scope seems to be lost  
sometimes! This is a Plone USP! Always beeing Unicode and i18n aware  
is a must and we need to keep this important competence permanently  
ahead of other systems. If all the stuff gets stuck when we try to  
internationalize it, we end up with a big problem!

Armin

P.S.
When I had to go from 2 to 2.1+ I was limited soo long due to the lack  
of serious migration documentation understandable for semi- 
programmers. E.g. I was trapped with CMFPhoto... It was not a real  
problem at all finally. But emotional pain can hurt!


--
Armin Carl Stroß-Radschinski, Dipl. Designer
acsr industrialdesign, Landgrafenstraße 32, 53842 Troisdorf, Germany

Telefon +49 (0) 22 41 / 94 69 94, FAX +49 (0) 22 41 / 94 69 96
eMail a.stross-radschinski-itUlVD6B3SM< at >public.gmane.org - http://www.acsr.de
UST. ID Nr: DE154092803 (EU VAT ID)





------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Dylan Jay — 2012-05-25T05:07:32

On 25/05/2012, at 3:48 PM, T. Kim Nguyen wrote:


yeah sorry, stop energy is not what I intended either. More that I  
think deco is the sugar coating that will help the transition to  
dexterity go down easier. So more go go energy for Deco, than stop  
energy :) I know deco seems too far off by some but as far as I can  
see most of the code for Deco is there it's just in the process of  
being reintegrated with p.a.toolbar so it's usable again, and that  
work is already underway by Rok. I'm no expert but I'd say we're  
closer to Deco than we are to p.a.ct. Rok?




------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

T. Kim Nguyen — 2012-05-25T05:48:11

Since I seem to have added fuel to the fire here, I would like to clarify that I am NOT trying to add stop-energy: I would, however, like everyone to consider whether or not each added feature improves our ease-of-adoption and ease-of-continuing-to-use-Plone stories.  

Kim

On May 24, 2012, at 8:00 PM, Dylan Jay wrote:



------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Dylan Jay — 2012-05-25T01:00:37


If you are going to have AT and p.a.ct types side by side you will  
have two Events, two news items etc. Thats pretty confusing.

Why not a release where Dexterity is in the core with the schema  
editor to support any additional types integrators want to create but  
the standard types are still AT? Then plugin authors don't have to  
support both dexterity without deco and dexterity with deco. In fact  
lots of plugins go away with deco so even less work would need to  
happen overall.


I fear converting all the standard AT objects to dexterity is not  
going to be a low-cost-of-upgrade release.




------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Jon Stahl — 2012-05-25T00:54:50

A couple of friendly amendments to Eric's observations:


... and, obviously, the benefits of Dexterity (e.g. TTW schema
editing, filesystem roundtripping, behaviors) won't be available to
that content.

IMHO, we probably do want to organize/document a concerted effort to
get add-on developers to upgrade/refactor/create Dexterity-versions of
their important affected products. The earlier we start that, and the
better directions we provide, the sooner it will happen.


$0.02,
jon

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Eric Steele — 2012-05-24T23:38:53

Archetypes is not vanishing. It's just not what Plone will use by default.
Any existing content created with it and schemaextender will still work, but
will require that those packages be installed first.  

Given the above, I firmly believe that substantial layout changes are the
biggest migration headache. See: Plone 3.  

Given that, I'd like us to have a low-cost-of-upgrade release including the
already in core Dexterity and Diazo that those unwilling to go through a larger
upgrade can happily sit on for some time. See: Plone 2.5.

Do not get me wrong, I absolutely want Deco in Plone. But from a project management standpoint, I think a tight – in both scope and timeline – release makes sense here. We're reaching a point where there are a number of backwards-incompatibile code cleanups in the works that would dramatically simplify things for integrators/developers. Deco and Dexterity are in core, FWT-approved, in use, and with a focused effort, we can polish them to 100% awesome. While it's nowhere near as bad as it was at the end of the 3.x cycle, momentum is slowing. Our limited pool of front-end expertise is solidly booked.  

What I want is Awesome now instead of  waiting for Potentially Earth-Shattering in 2 years.  

Where's my beer?------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers

Re: It's time to talk Plone 5

Dylan Jay — 2012-05-24T22:25:54


I think you miss the point. From an outside perspective an upgrade to
dexterity with p.a.ct is not really an upgrade. It will look and feel
the same as before except lots of plugins such as those that rely on
schema extender will no longer work. You will get the ability to
create content types but you can also get that now by installing
dexterity as an add on.
Think of python3. Why has it taken so long to be supported? Because it
gives you no immediate end user benefit even though it's a "good"
change.
I'm installing plone 4,2 in production but turning p.a.collections off
because even though it's nicer it doesn't have table functionality yet
and it's not compatible with easy newsletter so will be confusing.
I can see that creating p.a.ct is not that hard so why not? But I
think the effort that goes into polishing it and creating upgrades
will be a lot.
Also consider p a ct will create exactly the situation you are trying
to solve. Why would I upgrade my add ons to be compatible with p a ct
when deco is right around the corner?

Dexterity with deco on the other hand offers a big obvious benifit. So
as to avoid the pain of creating a release that everyone will hold off
using I'd ask everyone to consider skipping p a ct and sprint on deco
instead.



------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Sean Upton — 2012-05-24T17:27:09

Because we need to phase Archetypes out of Plone's learning curve --
it creates TMTOWDTI problem for add-on developers.  There should be
one-right-way to create content types, and it makes it arguably easier
if the core content types are built using the same technology as
add-ons.  Only supporting AT for integrating older add-ons is a
reasonable way to get here.

Sean

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: Deprecated Unit Test Methods in Plone core

Héctor Velarde — 2012-05-24T12:10:45

thank you, Jon; that makes me feel better :-)

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers

Re: It's time to talk Plone 5

Yuri — 2012-05-24T08:48:27

Il 24/05/2012 10:13, Gauthier Bastien ha scritto:

I was thinking of copy the module from buildout-cache to a "custom" 
directory, then edit it and see changes.

I think autorestart is plone.reload + automatically reloading. Can the 
automatically reloading take care of the custom directory, so it 
replaces the main eg with the customized one? :)

Is this process reliable, then?




------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Gauthier Bastien — 2012-05-24T08:13:02

Python can everything ;-)

http://pypi.python.org/pypi/collective.autorestart/0.2.0

Have a nice day,

Gauthier Bastien
IMIO
Zoning industriel, 34 - 5190 Mornimont
Tél. 071 780 979
www.communesplone.org


Le 24/05/12 10:10, Yuri a écrit :

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Yuri — 2012-05-24T08:10:03

Il 24/05/2012 03:14, T Kim Nguyen ha scritto:

Good questions!

a - what does Dexterity offer that will make it easier for designers to 
adopt Plone over other choices they have?

I think the zope approach is not comparable with others, so it does not 
offer anything "easier", it is always a new thing to learn with a high 
learning curve. It is a good framework for good developers. Just see the 
number and the quality of new projects based on dexterity/z3c.form.

b - CMSUI and Diazo are relevant in that respect, and Deco I could see, 
however.

Yes for development over TTW. This is the "strong" point of plone and 
zope for a newcomers and integrators.

What we miss is a mid-way between TTW (where you can do something but 
not all, and you can't "save" your changes to reuse them) and file 
system development. I would call it TTW on file system, where you can 
just copy a file/module and work on it, with no server restart or going 
to /< at >< at >reload. Can't just python check if a module has an updated version 
on fs and load it the next request came in?

We did this in Perl for TT templates (adding a "customization" path to 
the list of path where it looks for templates, just like you do for 
binaries lookup using the $PATH envinronment variable.

Maybe python can't, who knows?

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Dylan Jay — 2012-05-24T01:53:00


TTW views are really only needed if Deco isn't there so in a way Deco  
will replace TTW Views.  p.a.ct is perhaps only needed if Deco lite   
isn't there.
Dexterity being in the core is good as it allows the user to create  
new TTW types, but it's doesn't let you edit existing types does it?
In reality what I think would happen with p.a.ct is that most  
integrators will skip it as it will be an upgrade nightmare, and wait  
for Deco which gives the users a much nicer UI for creating custom  
views than TTW views.



------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Nathan Van Gheem — 2012-05-24T01:27:59

Sorry for the thread spam. Also, FWIW, we're going to try and work on
toolbar during the sprints here in state college. The hope is to make a
release by the end of it.

On Wed, May 23, 2012 at 9:20 PM, Nathan Van Gheem <vangheem-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>wrote:

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers

Re: It's time to talk Plone 5

Nathan Van Gheem — 2012-05-24T01:20:50

I'm not sure I understand what your point in bring up dexterity is. I think
almost everyone thinks it's a good idea to move it into core. If the TTW
story progresses more, TTW views will be done for dexterity content types.

On Wed, May 23, 2012 at 9:18 PM, Nathan Van Gheem <vangheem-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>wrote:

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers

Re: It's time to talk Plone 5

Nathan Van Gheem — 2012-05-24T01:18:21

The obvious thing being that dexterity offers the TTW content type editor.

On Wed, May 23, 2012 at 9:14 PM, T Kim Nguyen <nguyen-EYpa+rurPZaHXe+LvDLADg< at >public.gmane.org> wrote:

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers

Re: It's time to talk Plone 5

T Kim Nguyen — 2012-05-24T01:14:55

As a bit of an outsider who is concerned with our story vis other CMSs I have to ask: what does Dexterity offer that will make it easier for designers to adopt Plone over other choices they have? CMSUI and Diazo are relevant in that respect, and Deco I could see, however.

    Kim

On May 23, 2012, at 6:56 PM, Dylan Jay <djay-n0pU0XVUApFWk0Htik3J/w< at >public.gmane.org> wrote:

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers

Re: Deprecated Unit Test Methods in Plone core

Jon Stahl — 2012-05-23T23:50:27

On Wed, May 23, 2012 at 3:18 PM, Héctor Velarde
<hector.velarde-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

ATCT should/is planned to (AFAIK) become an optional add-on in Plone
5, not be completely "killed."  So, I think this was very worthwhile
effort. :-)

:jon

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Re: It's time to talk Plone 5

Dylan Jay — 2012-05-23T22:56:55

I'd say let's try and get deco light done but be prepared to push it off to
a 5.1 release if it's not ready.

deco light is dependent on toolbar begin solid...

Writing and supporting migration code from AT to dexterity and then soon
after to deco would be a lot of effort that is wasted. Lots of plugins are
going to break that rely on AT. If integrators had to redevelop them for
p.a.contenttypes and then again to work with tiles... I can see a lot of
frustration.
I can see the reasoning that its a pragmatic step that p.a.conyentypes is
easy to understand and with the slow rate of development it is more
realistic.
But we have limited resources and limited good will. If we are asking
developers and users to make two incompatible upgrades in quick succession
we may lose people, similar to what happened when we moved to browser
views.

I personally think plone needs to jump ahead of the competition. I think
deco allows us to do this and we should be putting all our effort into
this.

Rok is putting a lot of effort into deco at the moment. He's already got
the toolbar stable (minus overlays) and I believe is now working on mankind
deco work again with the toolbar. If we can at least get deco lite stable
enough to work as a plugin I think it will seem obvious to us we don't need
an interim step. Lets talk to Rok about what needs to be done and redirect
p.a.ct into deco.

On Wed, May 23, 2012 at 11:18 AM, Timo Stollenwerk <lists-YMfZcoNdYps< at >public.gmane.org> wrote:

_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
Plone-developers mailing list
Plone-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/plone-developers