There is some ideas in http://code.google.com/p/tryton/wiki/gsoc2009#Ideas
For B2CK, it is not in our priority list, but we will help anybody that wants
to start it.
By the way, I don't see a real benefit of a web client as we have thin client
for Windows (with and without install), MacOS and Linux/Unix.
> OpenERP's Web-
> Client is nice, but it comes with a unfriendly OpenERP Public
> License :)
I don't think it is really nice. It is pretty slow compared to a thin client
(due to his design) and the ergonomy is not so great.
But it is possible to start with eTiny before the release change and adapt it
to Tryton without too much work.
--
Cédric Krier
B2CK SPRL
Rue de Rotterdam, 4
4000 Liège
Belgium
Tel: +32 472 54 46 59
Email: cedric...@b2ck.com
Jabber: cedric...@b2ck.com
Website: http://www.b2ck.com/
imho, a web client is very important.
from a users perspective, you sometimes do not have possibilities to
install additional software (either because you don't have the
permissions, or because it's technically not possible (e.g. thin
clients)). but a browsers can be expected on any machine these days.
from a sysadmin perspective, webfrontends are in many situations to be
prefered over local software on clients.
> I don't think it is really nice. It is pretty slow compared to a thin client
> (due to his design) and the ergonomy is not so great.
it was recently ported away from turbogears2 (which was indeed very
slow) to cherrypy (as of version 5.0.2-0 of the webclient), which is
quite fast now.
> But it is possible to start with eTiny before the release change and adapt it
> to Tryton without too much work.
it's quite bad that they turned the webfrontent proprietary. but using
etiny under the old license is a quite big step backwards as it uses
turbogears. not sure if it makes sense at all to invest time into
working/basing on that for a future tryton/tryton-compatible webclient.
Regards,
Daniel
--
Address: Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
Email: daniel....@panthera-systems.net
Internet: http://people.panthera-systems.net/~daniel-baumann/
Are you really sur that you want to expose your company datas (accounting,
sales, payroll, etc.) on internet ?
Is not everybody have a laptop on which they can connect through a VPN (or not)
to the services of his company ?
> from a sysadmin perspective, webfrontends are in many situations to be
> prefered over local software on clients.
I think sysadmin doesn't care about the ergonomy and usability of
applications.
Of course it is simple to manage only server-side but you can manage easily
thin client with simple script that downloads updated client at startup or use
the exe from a shared folder, etc.
And any way, you will always have client-side applications (at least a
browser) or you use terminal which can use thin client.
A native application will always be better integrated in the desktop than a
pseudo-application in a browser.
> > I don't think it is really nice. It is pretty slow compared to a thin client
> > (due to his design) and the ergonomy is not so great.
>
> it was recently ported away from turbogears2 (which was indeed very
> slow) to cherrypy (as of version 5.0.2-0 of the webclient), which is
> quite fast now.
A web client with template will always be slower than a thin client.
With template, you send a lot of html to dislay information but with a thin
client you send only data (and once the layout).
> > But it is possible to start with eTiny before the release change and adapt it
> > to Tryton without too much work.
>
> it's quite bad that they turned the webfrontent proprietary. but using
> etiny under the old license is a quite big step backwards as it uses
> turbogears.
Why not change also to cherrypy ?
> not sure if it makes sense at all to invest time into
> working/basing on that for a future tryton/tryton-compatible webclient.
I will not.
nobody spoke about internet.
> I think sysadmin doesn't care about the ergonomy and usability of
> applications.
> Of course it is simple to manage only server-side but you can manage easily
> thin client with simple script that downloads updated client at startup or use
> the exe from a shared folder, etc.
no offence, but you appear to not have much sysadmin experience in
bigger environments. speaking about road warriors, it's practically
impossible to 'update' them within reasonable time.
for the rest, you can always find different ways on the paper to
acchieve a similar thing or find an alternative workaround. in practise,
however, there are many people that do not even look at erp/crm and
similar software, if it does not feature at least a webfrontend. and
they are right. ymmv.
> Why not change also to cherrypy ?
sure, who volunteers/has time?
>> not sure if it makes sense at all to invest time into
>> working/basing on that for a future tryton/tryton-compatible webclient.
>
> I will not.
me neither.
>
>> Why not change also to cherrypy ?
>
> sure, who volunteers/has time?
Who has budget?
--
Schönen Gruß - Regards
Hartmut Goebel
Dipl.-Informatiker (univ.), CISSP, CSSLP
Goebel Consult
Spezialist für IT-Sicherheit in komplexen Umgebungen
http://www.goebel-consult.de
Monatliche Kolumne:
<http://www.all-about-security.de/kolumnen/cissp-gefluester/>
Goebel Consult mit Mitglied bei <http://www.7-it.de>
Web is an application built on top of the Internet [1].
> > I think sysadmin doesn't care about the ergonomy and usability of
> > applications.
> > Of course it is simple to manage only server-side but you can manage easily
> > thin client with simple script that downloads updated client at startup or use
> > the exe from a shared folder, etc.
>
> no offence, but you appear to not have much sysadmin experience in
> bigger environments.
What do you call bigger ?
> speaking about road warriors, it's practically
> impossible to 'update' them within reasonable time.
Source of Tryton is about 400K (0.6 sec).
Exe is about 9M, which can be a little slow (12 sec on my ADSL connection),
but it could be possible to use diff patches (like Courgette [2]).
> for the rest, you can always find different ways on the paper to
> acchieve a similar thing or find an alternative workaround.
Web application is a workaround :-)
[1] http://en.wikipedia.org/wiki/World_Wide_Web
[2] http://dev.chromium.org/developers/design-documents/software-updates-courgette
You can even put Tryton on a USB key.
so by definition, i cannot run a web application in an intranet.
it's great to have definitions and apply them without flexibility.
> What do you call bigger ?
>= 50.
>> speaking about road warriors, it's practically
>> impossible to 'update' them within reasonable time.
>
> Source of Tryton is about 400K (0.6 sec).
> Exe is about 9M, which can be a little slow (12 sec on my ADSL connection),
> but it could be possible to use diff patches (like Courgette [2]).
you're missing the point. road warriors are on the road, they do not
have the time/possibility to update their clients.
however, no point in discussing that any further, you're not
understanding what i mean (and i don't have the time right now to
explain better).
> By the way, I don't see a real benefit of a web client as we have thin client
> for Windows (with and without install), MacOS and Linux/Unix.
>
The main benefit I see are PDA, Smartphones or any other device that has
a web browser but is not Linux, Windows or MacOS, or being one of those,
don't allow you to execute current thin client executable.
>
>> OpenERP's Web-
>> Client is nice, but it comes with a unfriendly OpenERP Public
>> License :)
>>
>
> I don't think it is really nice. It is pretty slow compared to a thin client
> (due to his design) and the ergonomy is not so great.
>
> But it is possible to start with eTiny before the release change and adapt it
> to Tryton without too much work.
>
>
Cheers.
--
Carlos Perelló Marín
[P+] SERVICIOS PROFESIONALES
http://www.pemas.es
mailto:car...@pemas.es || mailto:car...@gnome.org
Este mensaje y los ficheros anexos son confidenciales. Los mismos
contienen información reservada de la empresa que no puede ser
difundida. Si usted ha recibido este correo por error, tenga la
amabilidad de eliminarlo de su sistema y avisar al remitente mediante
reenvío a su dirección electrónica; no deberá copiar el mensaje ni
divulgar su contenido a ninguna persona.
Su dirección de correo electrónico junto a sus datos personales forman
parte de un fichero titularidad de PEMAS Servicios Profesionales, S.L.
cuya finalidad es la de mantener el contacto con Ud. De acuerdo con la
Ley Orgánica 15/1999, usted puede ejercitar sus derechos de acceso,
rectificación, cancelación y, en su caso, oposición enviando una
solicitud por escrito, acompañada de una fotocopia de su DNI dirigida a
PEMAS Servicios Profesionales, S.L. C/ Santos Justo y Pastor, 72 - 5,
C.P. 46022 de Valencia (España).
On 17/08/09 20:12 -0700, Oldrev wrote:There is some ideas in http://code.google.com/p/tryton/wiki/gsoc2009#Ideas
>
> hi all:
>
> Do we have any chance to see a Web Client for Tryton?
For B2CK, it is not in our priority list, but we will help anybody that wants
to start it.
By the way, I don't see a real benefit of a web client as we have thin client
for Windows (with and without install), MacOS and Linux/Unix.
I don't think that a general Web client will run well on these devices. I
think you will need a specific mobile web client.
By the way, there is some devices that use GTK as toolkit like N800 [1] which
could perhaps run the thin client.
[1] http://en.wikipedia.org/wiki/Nokia_N800
If it is intranet, download time is really not an issue.
> > What do you call bigger ?
>
> >= 50.
I worked in such env. And it was not really an issue to update client
applications at logon.
> >> speaking about road warriors, it's practically
> >> impossible to 'update' them within reasonable time.
> >
> > Source of Tryton is about 400K (0.6 sec).
> > Exe is about 9M, which can be a little slow (12 sec on my ADSL connection),
> > but it could be possible to use diff patches (like Courgette [2]).
>
> you're missing the point. road warriors are on the road, they do not
> have the time/possibility to update their clients.
You mean they don't have 12 sec every 2 months? (it is the average of new client
release)
And like I said by using patches, it can even be faster.
> I worked in such env. And it was not really an issue to update client
> applications at logon.
Have you work in en environment > 5.000 users? Maintenance is a problem
there, delivering the packages is only a minimal part of it.
> And like I said by using patches, it can even be faster.
Depending on the software distribution system the company is using, ther
may be not "patches".
I will be happy to see Tryton running with 5000 users.
But what is the maintenance problem ?
> > And like I said by using patches, it can even be faster.
>
> Depending on the software distribution system the company is using, ther
> may be not "patches".
But that is a choice of sysadmin.
Any way, thin client (windows and macos) could be binary patched.
On other systems that provides python-gtk environement, downloading the source is
realy not an issue.
> > Have you work in en environment > 5.000 users? Maintenance is a problem
> > there, delivering the packages is only a minimal part of it.
>
> I will be happy to see Tryton running with 5000 users.
> But what is the maintenance problem ?
Deployment of client frontends need an infrastructure and software (which
needs to be bought). This gets problematic not only for road warriors, also
for desktop machines.
Its quite some effort, and for this large companies move away from locally
installed frontends and request for newly introduced software a webfrontend,
to get into the direction of zero-deployment.
Thats a fact & strategic decision in some DAX-30 companies. A web-frontend is
a strategic & competitive advantage for trytoon (even if it may not run in a
DAX-30 company for the moment)
my 2c
Ax
--
Dr.-Ing. Axel K. Braun
Mobile: +49.173.7003.154
VoIP/Skype: axxite
PGP Fingerprint: CB03 964D 1CFA E87B AA63 53F3 1BD6 F53A EB48 EF22
Public Key available at http://www.axxite.com/axel....@gmx.de.asc
This mail was *not scanned* before sending.
It was sended from a virus-free Linux desktop.
It seams as if you never ever worked in *big* companies. Updating a
software package is quite an effort there: managing the change,
packaging the software, testing, roll-out, handling roll-out problems.
It's not about moving some bytes of the network, it's much more,
involved quite a lot of people and costs more than just some bucks.
Ina *big* company, the effort is the same, even if only 10 users use
Tryton. A web client would save a lot of effort and money here.
>>> And like I said by using patches, it can even be faster.
>> Depending on the software distribution system the company is using, ther
>> may be not "patches".
>
> But that is a choice of sysadmin.
The sysadmin has no choice, since the delivery software is already in
place. He has to use it. Changing the delivery software is *really*
expensive: the costs of software evaluation, new licences, re-package
all software, re-test the packages, roll-out new software (mind the
offline users!), handling problems (which are new to all), retrain the
help-center, change management, not to count the costs by
> Any way, thin client (windows and macos) could be binary patched.
> On other systems that provides python-gtk environement, downloading the source is
> realy not an issue.
No offence meant, but it seams as if you are really lacking any
experience in software maintenance and delivery in large environments.
Ok, this make sense only when you have a zero-deployment OS (which I don't
know).
In fact, all of that is a cyclic behavior in IT. 80's was about mainframe,
90's about PC and now we go back to mainframe but on web.
The thin client is a compromize between both architecture, I think:
- no need to update too often
- speed/integration of local application
But again, I'm not against a web client. I just will not start developping one now
(except if someone fund it). There is better thing to do.
debian is. either by using cfengine or pupped, or preseed-* packages.
No, you must do almost the same effort for managing the changes, packaging,
testing, etc. for a web application than on local.
> >>> And like I said by using patches, it can even be faster.
> >> Depending on the software distribution system the company is using, ther
> >> may be not "patches".
> >
> > But that is a choice of sysadmin.
>
> The sysadmin has no choice, since the delivery software is already in
> place. He has to use it. Changing the delivery software is *really*
> expensive: the costs of software evaluation, new licences, re-package
> all software, re-test the packages, roll-out new software (mind the
> offline users!), handling problems (which are new to all), retrain the
> help-center, change management, not to count the costs by
I don't understand.
You can always choose how you will install Tryton.
> > Any way, thin client (windows and macos) could be binary patched.
> > On other systems that provides python-gtk environement, downloading the source is
> > realy not an issue.
>
> No offence meant, but it seams as if you are really lacking any
> experience in software maintenance and delivery in large environments.
Or perhaps the other, I don't have issues on some tasks that others have.
With such solutions, you can manage Tryton update easily.
nope, not on road warriors.
>
>>>>> And like I said by using patches, it can even be faster.
>>>> Depending on the software distribution system the company is using, ther
>>>> may be not "patches".
>>> But that is a choice of sysadmin.
>> The sysadmin has no choice, since the delivery software is already in
>> place. He has to use it. Changing the delivery software is *really*
>> expensive: the costs of software evaluation, new licences, re-package
>> all software, re-test the packages, roll-out new software (mind the
>> offline users!), handling problems (which are new to all), retrain the
>> help-center, change management, not to count the costs by
>
> I don't understand.
> You can always choose how you will install Tryton.
"You" can not! "You" are a sysadmin and you have to follow the processes
defined in your company. If "You" are an upper manger, than "You" can
choose, "You" always want to have a business case for changing. There is
no business case in changing the software delivery systems, thus "You"
do not have a choice.
> Yes but you still have to manage the OS of the road warriors.
Each additional component which has to be managed costs additional money.
what happens if you cannot get distribution updates? system works the
same as before, maybe just a little less secure as before (if affected).
what happens if you cannot get tryton updates after system
administration changed configuration or upgraded tryton? tryton doesn't
work anymore.
forcing up full updates for a client only to get new tryton is not sane.
priorizing tryton updates over distribution updates is possible, in
practise not going to work. teaching users to be clever enough themselfs
neither.
been there, done that.
Sure, but one optimized for that kind of devices would be easier than
port the client to many platforms, however, having a native
port/executable for each architecture would be quite cool.
> By the way, there is some devices that use GTK as toolkit like N800 [1] which
> could perhaps run the thin client.
>
> [1] http://en.wikipedia.org/wiki/Nokia_N800
>
Yeah, I know, I have an N800 device myself, but as usual, I never have
time to hack on it :-(
You have the same with the exe [1] without installation. You can write with ~10
lines of code, the exe that will fetch the new version if there is one.
[1] http://downloads.tryton.org/1.2/tryton-1.2.0.exe
--
Cédric Krier
B2CK SPRL
Rue de Rotterdam, 4
4000 Liège
Belgium
Tel: +32 472 54 46 59
I agree that a web client can be usefull for some cases. But in all
cases a web client must be sponsored or financed in some way. If there
is money for a web client should be created.
Am Montag, den 24.08.2009, 16:20 +0200 schrieb Cédric Krier:
> On 24/08/09 06:53 -0700, RussellB wrote:
> >
> > Hi All
> >
> > Just to let you know that we are currently unable to use Tryton
> > *because* of the lack of a web client, however much I would like to
> > (the new module inheritance method is MUCH cleaner and nicer than
> > OpenERP's!)
> >
> > We need a web client because we are looking to deliver customized
> > software-as-a-service applications for businesses. Most customers
> > looking for SAAS applications are doing so because it saves them money
> > by removing the need for them to manage their own infrastructure (they
> > don't want to install and maintain software), and the applications can
> > be accessed from anywhere.
>
> You have the same with the exe [1] without installation. You can write with ~10
> lines of code, the exe that will fetch the new version if there is one.
>
> [1] http://downloads.tryton.org/1.2/tryton-1.2.0.exe
>
--
Korbinian Preisler
____________________________________
virtual things
Preisler & Spallek GbR
Munich - Aix-la-Chapelle
Windeckstr. 77
81375 Munich - Germany
Tel: +49 (89) 710 481 55
Fax: +49 (89) 710 481 56
>
> We need a web client because we are looking to deliver customized
> software-as-a-service applications for businesses. Most customers
So what about funding the development of a web-client? You seam to have
a business case for that, so you should be able to invest some bucks :-)
--
Schönen Gruß - Regards
Hartmut Goebel
Dipl.-Informatiker (univ.), CISSP, CSSLP
Goebel Consult
Spezialist für IT-Sicherheit in komplexen Umgebungen
http://www.goebel-consult.de
Monatliche Kolumne: http://www.cissp-gefluester.de/
Hi All
I would like to volunteer some of my time to start / continue work on a Web Client for tryton.
I am a very experienced OpenERP developer, but don't like their organizational structure (Publisher & Partners) or the "OpenERP Public License" on the web client!!
I have many years experience of Web Application development (since 1998) so I should be able to do a good job! :)
So a couple of questions:
1. Is there an 'official' place where someone has started a Web Client, that I should contribute to, or should I start a new one. (Set up a new Google Code Project?)
2. I've seen the great work Sharoon has done on the prototype Tryton Web Client using Ext JS. My only concern about using Ext JS is that it is not developed in the open - it is mainly proprietary but released under GPL seemingly just to keep existing users happy. I am concerned that it may go completely proprietary at some point.
So I'm thinking using jQuery's UI Toolkit will be a better approach - they have a mobile version of it too. Any thoughts?
3. In terms of architecture, it seems that the most sensible approach, from a performance and maintainability standpoint, would be to write the majority of the web client in JavaScript, and have the user's browser call trytond's JSON interface directly. Does everyone agree?
Look forward to hearing from you guys :)
Russ
On Tuesday, August 18, 2009 4:12:37 AM UTC+1, Oldrev wrote:
hi all:
Do we have any chance to see a Web Client for Tryton? OpenERP's Web-
Client is nice, but it comes with a unfriendly OpenERP Public
License :)
Haha. Good timing then! :)
I look forward to hearing about the outcome of the sprint, and let me know how I can help once the repo is in place :)
R