openflow-dev

[Openflow-dev] Document for migrating OpenFlow from version 0.7 to version 1.0

[Daniele <d.t...@ic...>]

Dear Openflowers,

thanks to the sponsoring of Zflow Corporation (http://www.Zflow.com) in the 
documentation section of the openflow site you'll find a how-to document  to 
migrate OpenFlow 0.7 applications to OpenFlow 1.0 applications. 

http://www.openflow.it/Documentation/documentation/Migrating_OF07_OF10

Cheers,
	Daniele =)


-- 
Daniele Tarini - Research & Development - Icube S.r.l.
Address:        Via Ridolfi 15 - 56124 Pisa (PI), Italy
E-mail: d.t...@ic...     Web:    http://www.icube.it
Phone:  (+39) 050 97 02 07      Fax:    (+39) 050 31 36 588

Re: [Openflow-dev] openflow modifications and questions

[Daniele <d.t...@ic...>]

Hello Adrian,

you have big plans for OpenFlow, don't you? We are happy about that and we'll 
be glad to double check your patches and include them in the OpenFlow 
release. Let's see the details you were talking about.

All the "ironing" seems pretty reasonable. Tidy code is readable code. We are 
a bit concerned about the compatibility with the current OpenFlow 1.0 
instances. The following points could be dangerous for compatibility:

>  - use ClassCase for class names
>  - rename the openflow class to WorkflowEngine with meta_type
>    'Openflow Workflow Engine' or similar - remember 'workflow' is one
>    word in English
>  - refactor interface methods - eg. startWorkitem() -> instance.start()

Could it be arranged so that both interfaces work? We can then keep both 
interfaces for a few releases to give people time to switch from the old one 
to the new one.

The large refactorings left are definitely good. We were thinking about using 
BTreeFolder folders already; actually Shane Hathaway got out the 
BTreeFolder2, and we were focusing on using that one. Here's a link:
http://hathaway.freezope.org/Software/BTreeFolder2

Thanks for your support to the OpenFlow project: we greatly appreciate it.

Cheers,
	Daniele =)


PS: about the transition ids:
>  - generate transition ids from the endpoint activities (there is no
>    need for human-generated transition ids)
... actually if you don't specify an id, the transition id will already 
default to: <FromActivity>_<ToActivity>
=)




On Tuesday 04 February 2003 05:24, Adrian van den Dries wrote:
> Hi Openflowers,
>
> At my work, we need to deploy a working workflow engine, but I need to
> make some changes (get the timeouts working), but before I do I'd like
> to iron out some wrinkles that irk me.  Here are some of them:
>
>  - choose either of DTMLFile or PageTemplateFile and then
>  - move all the .pt or .dtml into one flat directory
>  - use ClassCase for class names
>  - rename the openflow class to WorkflowEngine with meta_type
>    'Openflow Workflow Engine' or similar - remember 'workflow' is one
>    word in English
>  - clarify what the package name is (OpenFlow, openflow_engine, or
>    openflow-engine as in CVS - perhaps move the package to Openflow
>    (my preference) to under the openflow-engine module in CVS?)
>  - fix spelling errors ('registar' should be 'registrar' etc.)
>  - remove Italian and tidy English (unactivate -> deactivate etc.)
>  - generate transition ids from the endpoint activities (there is no
>    need for human-generated transition ids)
>  - use symbolic permission names from a permissions module
>  - clarify terminology (actor/client/customer... etc.)
>  - specify exceptions for bare except: clauses
>
> And then move onto some larger refactoring, like:
>
>  - refactor large conditions/expressions into separate methods
>  - refactor interface methods - eg. startWorkitem() -> instance.start()
>  - move to BTrees over Folders, as workflows could potentially have
>    large numbers of objects, and other such optimisations
>
> Are these too out of scope?  Are they too hard?  Are they too petty?
> If the icube guys are happy with the above and would like to make some
> decisions (names, etc) would you be happy for me to submit patches for
> the above?  Obviously changing class names (and the package name) will
> break existing instances, but is perhaps worth it for the sake of
> clarity and consistency.
>
> a.

-- 
Daniele Tarini - Research & Developement - Icube S.r.l.
Address:        Via Ridolfi 15 - 56124 Pisa (PI), Italy
E-mail: d.t...@ic...     Web:    http://www.icube.it
Phone:  (+39) 050 97 02 07      Fax:    (+39) 050 31 36 588

[Openflow-dev] openflow modifications and questions

[Adrian v. d. D. <ad...@fl...>]

Hi Openflowers,

At my work, we need to deploy a working workflow engine, but I need to
make some changes (get the timeouts working), but before I do I'd like
to iron out some wrinkles that irk me.  Here are some of them:

 - choose either of DTMLFile or PageTemplateFile and then
 - move all the .pt or .dtml into one flat directory
 - use ClassCase for class names
 - rename the openflow class to WorkflowEngine with meta_type
   'Openflow Workflow Engine' or similar - remember 'workflow' is one
   word in English
 - clarify what the package name is (OpenFlow, openflow_engine, or
   openflow-engine as in CVS - perhaps move the package to Openflow
   (my preference) to under the openflow-engine module in CVS?)
 - fix spelling errors ('registar' should be 'registrar' etc.)
 - remove Italian and tidy English (unactivate -> deactivate etc.)
 - generate transition ids from the endpoint activities (there is no
   need for human-generated transition ids)
 - use symbolic permission names from a permissions module
 - clarify terminology (actor/client/customer... etc.)
 - specify exceptions for bare except: clauses

And then move onto some larger refactoring, like:

 - refactor large conditions/expressions into separate methods
 - refactor interface methods - eg. startWorkitem() -> instance.start()
 - move to BTrees over Folders, as workflows could potentially have
   large numbers of objects, and other such optimisations

Are these too out of scope?  Are they too hard?  Are they too petty?
If the icube guys are happy with the above and would like to make some
decisions (names, etc) would you be happy for me to submit patches for
the above?  Obviously changing class names (and the package name) will
break existing instances, but is perhaps worth it for the sake of
clarity and consistency.

a.

-- 
 Adrian van den Dries                           ad...@fl...
 Development team                               www.dev.flow.com.au
 FLOW Communications Pty. Ltd.                  www.flow.com.au

[Openflow-dev] OpenFlow 1.0.5

[Daniele <d.t...@ic...>]

Dear OpenFlowers,

I just released OpenFlow version 1.0.5. You'll find it at:
http://www.openflow.it/Download/openflow.1.0.5.tgz

Changes from version 1.0.3 are:

- minor bug fixes

- CMF is not needed anymore (fixed bug in importing custom TAL Expressions 
module instead of CMF one).


Hoping you're having fun with OpenFlow,
  Daniele =)


-- 
Daniele Tarini - Research & Developement - Icube S.r.l.
Address:        Via Ridolfi 15 - 56124 Pisa (PI), Italy
E-mail: d.t...@ic...     Web:    http://www.icube.it
Phone:  (+39) 050 97 02 07      Fax:    (+39) 050 31 36 588

RE: [Openflow-dev] How are you using OpenFlow ?

[Jeff M. <mar...@ch...>]

Hi Paolo,

Besides our sponsored work with Icube on the OpenFlow project -

We work with SMB and SME companies to improve their competitiveness by
increasing productivity, reducing cost, and by introducing technology that
allows an increase in the scale of the operation.
We offer critical business process mapping and comprehensive workflow
management products to maximize use of the existing technology
infrastructure while decreasing the drain on the bottom line.

We are 100% dedicated to providing open source only solutions with the
underlying goal of reducing cost while increasing flexibility. Our goal is
to become an integral part of the open source workflow community. In doing
so, we can help ensure that all the necessary workflow features and
capabilities are available as a single cohesive product.

Our focus is on three main categories of products.
.	Specific workflow business productivity solutions for business services
.	Generic workflow business productivity solutions for Managed Service
Providers
.	Workflow solutions for Federal, State, and local municipalities

Our current project is with an Internet Services Provider. We are working
with the President & CEO, and key constituents to map their critical
business processes. Not unlike similar organizations, this company has five
main departments: Sales, Support, Operations, Finance, and Legal with the
typical customer interface through the Sales and Support teams. The critical
business processes are the ones relating to the Sales and Support teams.
Thus, our main focus is on Customer Relationship Management activities.
The environment in which we are working is composed of a hosted NT
Coldfusion server providing a repository portal for employee access. Their
environment includes Coldfusion and an MS SQL database.

We are proposing a sales-driven work flow management engine consisting of
OpenFlow coupled to their system via WEB-based services, direct database
access, and possibly access to their Email servers. Our focus is on CRM and
other high payback activities.

As far as our wish list for "feature request" on the future development of
OpenFlow (in no particular order) -

1.	Time based capabilities for activities to support CRM.
2.	Example workflow production ready products.
3.	Documentation including published books in English.
4.	Openflow friendly Integrated Development Environment (IDE) for commercial
development as proposed by Eron Lloyd:
http://www.zope.org/Members/Earthtone/ZGUI/ZGUI_dev-paper02
5.	Successful Web Services Execution as it relates to WF-XML, Soap, & UDDI


Jeffrey S. Martin
President & CEO
Zflow Business Productivity Solutions
6 Hillside Drive
Brookline, NH 03033

603.672.3888
je...@Zf...
http://www.Zflow.com




-----Original Message-----
From: ope...@li...
[mailto:ope...@li...]On Behalf Of Paolo
Bizzarri
Sent: Saturday, January 25, 2003 2:41 PM
To: ope...@li...
Subject: [Openflow-dev] How are you using OpenFlow ?


Hi everyone,

I would like to ask all the users of OpenFlow, if they are interested in
describing what they are doing right now with OpenFlow.

Can you write a page or two, describing what you are using OpenFlow for ?

We would like to publish a set of case studies on the web site of OpenFlow.

Many thanks to everyone who will contribute.

Best regards.
--
Paolo Bizzarri - Presidente - Icube S.r.l.
Sede:   Via Ridolfi 15 - 56124 Pisa (PI), Italia
E-mail: p.b...@ic...		WWW: www.icube.it
Tel: 	(+39) 050 97 02 07		Fax: (+39) 050 31 36 588


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Openflow-dev mailing list
Ope...@li...
https://lists.sourceforge.net/lists/listinfo/openflow-dev

[Openflow-dev] Re: [Openflow-pa] Openflow & Python

[Daniele <d.t...@ic...>]

OpenFlow 0.7:
X = openflow.getInstance(instance_id)

OpenFlow 1.0:
X = getattr(openflow, instance_id)


Cheers,
	Daniele =)


On Friday 31 January 2003 16:21, Eric wrote:
> Oops,
>
> Additional question, what IS the normal way to do it then in 1.0?
>
> Cheers!

-- 
Daniele Tarini - Research & Developement - Icube S.r.l.
Address:        Via Ridolfi 15 - 56124 Pisa (PI), Italy
E-mail: d.t...@ic...     Web:    http://www.icube.it
Phone:  (+39) 050 97 02 07      Fax:    (+39) 050 31 36 588

[Openflow-dev] Re: [Openflow-pa] Openflow & Python

[Daniele <d.t...@ic...>]

Dear Eric,

I saw the mail you sent to the OpenFlow-PA list. That one list is not used 
much anymore since it's intended specificallty for Public Administration (the 
ending pa in the list name).

I am replying to the mail posting it where usually that kind of question are 
posted: the Openflow-dev mailing list. 

The answer to your question is very easy: the getInstance API is not there 
anymore. We are considering putting it back since you are not the only one 
wondering where such method went. If you really miss it create a 
"getInstance" python script in the workflow object contents and put:

## parameters: instance_id
return getattr(container, instance_id)

and you'll get the same effect.

The API usage tips where written for OpenFlow 0.7. They are still there since 
a lot of people are still using OpenFlow 0.7. I know, I have to do some 
cleaning of the available stuff so that they won't induce that kind of errors 
anymore.

Not much of the API has changed, anyway. You'll probably have to double check 
with the source code, but the Tips are still pretty much useful even for 
OpenFlow 1.0, since the OpenFlow logic hasn't changed from 0.7 to 1.0.


Daniele =)


On Friday 31 January 2003 11:04, Eric wrote:
> Folks,
>
> I have been fiddling with Openflow for the past week and I do like it,
> however I am not all too familiar with Zope (not to say not familiar at
> all) and I am running into a some issues.
> For example, the APIUsage tips say I should use
> getInstance(instance_id).manage_addProperty(...), however when I try that,
> it doesn't work.
>
> The Python script used: (very basic indeed)
> # Example code:
>
> # Import a standard function, and get the HTML request and response
> objects. from Products.PythonScripts.standard import html_quote
> request = container.REQUEST
> RESPONSE =  request.RESPONSE
> customer="customer_id_%s_" % customer_id
> instance_id=context.ProjectFlows.addInstance(flow_name, customer,
> "comments", "title", activation=1)
>
> test = context.ProjectFlows.getInstance()
> test.manage_addProperty("customer_id",customer_id,"int")
> print "<h2>Received customer_id: %s and flow_name: %s<br>Created %s</h2>" %
> (customer_id, flow_name,instance_id)
> return printed
>
> An instance is created, so I figure I got the namespace right. However the
> getInstance doesn't work, nor did I find it in any of the .py files in the
> Openflow directory in Products.
>
> Any suggestions?
>
> Cheers!

-- 
Daniele Tarini - Research & Developement - Icube S.r.l.
Address:        Via Ridolfi 15 - 56124 Pisa (PI), Italy
E-mail: d.t...@ic...     Web:    http://www.icube.it
Phone:  (+39) 050 97 02 07      Fax:    (+39) 050 31 36 588

Re: [Openflow-dev] Timeouts

[Adrian v. d. D. <ad...@fl...>]

On January 29, Paolo Bizzarri wrote:
> The resulting flow is the following:
> 
>        S  
>       and
>     A     T1
>            T2
>        xor
>        T3
>        E

Thanks very much for this.  I thought it might be something like this,
but the combination of AND-split and XOR-join, and all the petri net
stuff I've been reading was making my brain go mush.

> This is not the possibly best solution, as OpenFlow currently lacks a good 
> support for time-driven event (here is another item for our todo list).

Don't bother; this is a problem for Zope.  Zope runs the event loop,
so it should provide a method for scheduling actions to be performed
in the future.  We're working on a simple product that allows this.

a.

-- 
 Adrian van den Dries                           ad...@fl...
 Development team                               www.dev.flow.com.au
 FLOW Communications Pty. Ltd.                  www.flow.com.au

[Openflow-dev] OpenFlow CVS is now OK

[Daniele <d.t...@ic...>]

Dear OpenFlowers,

the OpenFlow cvs is now updated and aligned to the latest OpenFlow release. 
Needless to say, CVS will be used for developing OpenFlow and will always 
have the latest developed version (released or not).

Cheers,
	Daniele =)

Re: [Openflow-dev] Timeouts

[Paolo B. <p.b...@ic...>]

Hi Adrian,

basically, you have to model the timeout as a separate activity.

Suppose you want to build a timeout on the activity A. Your workflow is 
something like:

S -> A -> E.

You have to:

-  add an "and" split among S and A;

- add two activities, T1 and T2, on the other branch of the and split;

- add a "xor" and among A, T2 and E.

- add an activity T3.

The resulting flow is the following:

       S  
      and
    A     T1
           T2
       xor
       T3
       E

T1 is an application activity, that adds an entry in the cron daemon, with the 
required timeout. 

T2 is a manual activity, which is invoked using something like the ZClient by 
the Cron, on the end of the timeout.

T3 is the activity that, if the timeout has not expired, provide to remove the 
entry from the Cron.

This is not the possibly best solution, as OpenFlow currently lacks a good 
support for time-driven event (here is another item for our todo list).

Best regards.

Paolo

On Wednesday 29 January 2003 04:09, Adrian van den Dries wrote:
> How do you model timeouts on an activity?
>
> a.

-- 
Paolo Bizzarri - President - Icube S.r.l.
Address:   	Via Ridolfi 15 - 56124 Pisa (PI), Italy
E-mail:	p.b...@ic...	Web:	http://www.icube.it
Phone:	(+39) 050 97 02 07	Fax: 	(+39) 050 31 36 588
		

[Openflow-dev] Timeouts

[Adrian v. d. D. <ad...@fl...>]

How do you model timeouts on an activity?

a.

-- 
 Adrian van den Dries                           ad...@fl...
 Development team                               www.dev.flow.com.au
 FLOW Communications Pty. Ltd.                  www.flow.com.au

Re:[Openflow-dev] How are you using OpenFlow?

[Michael D. <mic...@ea...>]

We provide business and technology alignment services, primarily in the
Small, Medium Business (SMB) market.  What that means in simple terms is we
review the operational business processes and help our clients automate
their operations to improve workflow, continuity and profitability.

At the base of the services we provide we are "technology agnostic" meaning
we will provide the best of breed solution that fits the customer
environment.  In the SMB market we often find existing systems are out of
date, lack integration or scalability, and have little support.  When that
is the case, OpenFlow represents a clean slate that allows our clients to
reevaluate their business processes and find an approach that meets their
needs without requiring them to bend to fit a predetermined operational
plan.

Our current project is for a private investigations company.  Our project
will allow their customers to add and track new cases, their investigators
to receive the cases and make annotations as the casework progresses, and to
accumulate the results into a final report that can be delivered to the
client electronically.  All this work will be able to be done remotely
across a VPN and the workflow will have an integration with QuickBooks
through a Python COM implementation.  Because all of the workflow side is
all web based, any user with access to a web browser can use the system.  We
are using Sendmail for mail services and Samba for Windows file service
across the VPN.  The hardware is all Intel-based.  We have an application
server, a document server (OpenBSD running Samba and Sendmail), a Windows
server for COM services and a sandbox server for development staging.

The base platform is OpenBSD and of course Zope on the application server.
We are using the Apache web server and PostgresSQL as the database for the
casework and accounting integration.

This solution was selected by the client over an alternative proprietary
investigations package we also offered to install and maintain.  The reasons
the client selected the OpenFlow solution are:
  1.. Flexibility to design a system to meet their process rather than
requiring them change to work with the process in a proprietary solution.
  2.. The proprietary solution was not being actively maintained to meet
current technology.  It was still client-server based, requiring Windows
Terminal Server for remote use.  Support from the software developer was
expensive and would have required much new coding.
  3.. There was no easy way to integrate the proprietary solution with
either their current accounting or future choices.  That meant all
accounting entries would be double work and prone to errors.  By using a
Windows server with the standard COM library, we could use the Windows based
Python to reach the COM object and integrate OpenFlow and the QuickBooks XML
interface which is designed for the Microsoft web services platform (.Net).
  4.. The Zope/PostgresSQL base is n-tier and can run on any common
operating platform.  The client has plans for growth and this platform is
extremely scalable.
  5.. The administration and hosting can be outsourced at a reasonable cost.
The client does not want to have to worry about having internal IT
resources.  We will provide those services.
  6.. The open source base means lower costs and wider options.  The cost of
a built to fit solution in this case is much less than the proprietary
solution and much more extendable.
It is important to understand we offered both the proprietary solution and
the OpenFlow solution side by side.  We costed hardware, licenses,
implementation, and maintenance for both solutions although we only sell
service.  Our competitors could only offer the proprietary solution and the
total cost of their proposals was far more than either of our proposals.  It
is also important to understand we came in discussing the business process
and not the eventual outcome.  When the client really looked at all the
things the proprietary package had in it that they frankly didn't need and
didn't want to add to their process, it was an easy choice.

Regards,

MIKE…
Make your Information your KnowlEDGE
Michael L. Dunham
Principal Consultant
2333 La Lima Way
Sacramento, CA 95833
Phone: 916-716-6073
Pager: 888-469-1963
Email: mic...@ke...
Webpage: http://www.kestrelworks.com/

Re: [Openflow-dev] How are you using OpenFlow ?

[Paolo B. <p.b...@ic...>]

Hi,

preferred language is english. 

However, if you can provide it in Italian, I will try to translate in a 
reasonable english.

Thank you

Paolo

> At 21.41 25/01/2003 +0100, you wrote:
> >Hi everyone,
> >
> >I would like to ask all the users of OpenFlow, if they are interested in
> >describing what they are doing right now with OpenFlow.
> >
> >Can you write a page or two, describing what you are using OpenFlow for ?
>
> It's ok also in italian language?
>
> regards, Maurizio Betti
>
>
>
>
>
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> _______________________________________________
> Openflow-dev mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/openflow-dev

-- 
Paolo Bizzarri - President - Icube S.r.l.
Address:   	Via Ridolfi 15 - 56124 Pisa (PI), Italy
E-mail:	p.b...@ic...	Web:	http://www.icube.it
Phone:	(+39) 050 97 02 07	Fax: 	(+39) 050 31 36 588

Re: [Openflow-dev] How are you using OpenFlow ?

[Maurizio B. <mb...@in...>]

At 21.41 25/01/2003 +0100, you wrote:
>Hi everyone,
>
>I would like to ask all the users of OpenFlow, if they are interested in
>describing what they are doing right now with OpenFlow.
>
>Can you write a page or two, describing what you are using OpenFlow for ?

It's ok also in italian language?

regards, Maurizio Betti

Re: [Openflow-dev] Workflow modelling techniques

[Paolo B. <p.b...@ic...>]

Hi Adrian,

we know petri nets, and if you see the model used in OpenFlow, you can find 
many analogies among the workflow model used in OpenFlow and the Petri Nets.

One of the first versions of OpenFlow was developed from Chautauqua:

http://www.cs.colorado.edu/~carlosm/chautauqua.html

that used a formalism called Information flow Control Net, which were similar 
to Petri Nets.

>
> I just recently discovered this excellent book:
>
>  Workflow Management
>  Models, Methods and Systems
>  Wil van der Aalst and Kees van Hee
>  MIT Press January 2002
>
>  http://mitpress.mit.edu/0262011891
>

[SNIP]

> The authors use petri nets as a formalisation for modelling workflow
> processes.
>
> Has anyone else found petri nets useful?  What other techniques do you
> find useful for modelling processes, especially highly parallel and/or
> conditional processes, and how do you go about translating the models
> to OpenFlow?
>

Using Petri Nets is useful for describing the general flow of work/actions 
inside the process. However, there is always the need to include further 
information (like, for example, the information requested and passed at each 
state). 

I guess the mapping among Petri Nets and OpenFlow processes could easily be 
one to one. Could be this useful for you ? We could work on how this could be 
accomplished. This could be one of the todo items we are trying to define for 
OpenFlow.

Best regards.

Paolo Bizzarri


-- 
Paolo Bizzarri - Presidente - Icube S.r.l.
Sede:   Via Ridolfi 15 - 56124 Pisa (PI), Italia 
E-mail: p.b...@ic...		WWW: www.icube.it	
Tel: 	(+39) 050 97 02 07		Fax: (+39) 050 31 36 588	

[Openflow-dev] Workflow modelling techniques

[Adrian v. d. D. <ad...@fl...>]

Hi openflowers,

I just recently discovered this excellent book:

 Workflow Management
 Models, Methods and Systems
 Wil van der Aalst and Kees van Hee
 MIT Press January 2002

 http://mitpress.mit.edu/0262011891/

If the site is still dead, use The Cache:

 http://www.google.com/search?q=cache:6gyPOGWI5PQC:mitpress.mit.edu/0262011891/

The authors use petri nets as a formalisation for modelling workflow
processes.

Has anyone else found petri nets useful?  What other techniques do you
find useful for modelling processes, especially highly parallel and/or
conditional processes, and how do you go about translating the models
to OpenFlow?

a.

-- 
 Adrian van den Dries                           ad...@fl...
 Development team                               www.dev.flow.com.au
 FLOW Communications Pty. Ltd.                  www.flow.com.au

[Openflow-dev] Feature wishlist for the future of OpenFlow

[<pi...@ic...>]

Hi everyone,

I would like to ask everyone on this mailing list to provide their "feature
request" on the future development of OpenFlow.

Right now, Icube have little resources to dedicate to the development of
OpenFlow. So we want to guarantee that what we are going to do is seen as top
priority by those who are using (or are interestend in using) OpenFlow.

Things could include:

- new specific features for OpenFlow itself;

- new programs to be used together with OpenFlow;

- more specific/detailed examples of using OpenFlow;

- updated documentation;

- everything else.

We will pick only the items that are seen as top priority, as we have little
time right now.

However, this could become an useful task list for whoever is interested in
contributing to the OpenFlow project.

Best regards

Paolo Bizzarri

-- 
Paolo Bizzarri - Presidente - Icube S.r.l.
Sede:   Via Ridolfi 15 - 56124 Pisa (PI), Italia 
E-mail: p.b...@ic...		WWW: www.icube.it	
Tel: 	(+39) 050 97 02 07		Fax: (+39) 050 31 36 588	

[Openflow-dev] How are you using OpenFlow ?

[<pi...@ic...>]

Hi everyone,

I would like to ask all the users of OpenFlow, if they are interested in
describing what they are doing right now with OpenFlow.

Can you write a page or two, describing what you are using OpenFlow for ?

We would like to publish a set of case studies on the web site of OpenFlow.

Many thanks to everyone who will contribute.

Best regards.
-- 
Paolo Bizzarri - Presidente - Icube S.r.l.
Sede:   Via Ridolfi 15 - 56124 Pisa (PI), Italia 
E-mail: p.b...@ic...		WWW: www.icube.it	
Tel: 	(+39) 050 97 02 07		Fax: (+39) 050 31 36 588	

Re: [Openflow-dev] Re: Fast Draw Demo

[Yuri C. <yu...@al...>]

Luca Arnaudo wrote:

>Of course, I thought I told the demo in the mail, I guess I'll write it on the 
>index.html. It's user demo pwd demo, tell me if anything goes wrong.
>
>Luca
>  
>
 I cannot view the example under mozilla and Linux :)

Re: [Openflow-dev] Process definition settings

[Adrian v. d. D. <ad...@fl...>]

On January 22, Adrian van den Dries wrote:
> What do the priority, limit, validity, waiting time and duration
> settings do within a process definition?  I did a quick grep and some
> of these appear not to be used within the engine.

Whoops!  These have been removed since 0.7.3 (I'm using that so I can
play with the demo).  Priority is the only one left in 1.0.3, and it's
actually used.

We return you to your regular schedule. ;-)

a.

-- 
 Adrian van den Dries                           ad...@fl...
 Development team                               www.dev.flow.com.au
 FLOW Communications Pty. Ltd.                  www.flow.com.au

[Openflow-dev] Process definition settings

[Adrian v. d. D. <ad...@fl...>]

Hi openflowers,  [damn, I love that]

What do the priority, limit, validity, waiting time and duration
settings do within a process definition?  I did a quick grep and some
of these appear not to be used within the engine.

a.

-- 
 Adrian van den Dries                           ad...@fl...
 Development team                               www.dev.flow.com.au
 FLOW Communications Pty. Ltd.                  www.flow.com.au

[Openflow-dev] Re: Fast Draw Demo

[Luca A. <luc...@re...>]

Of course, I thought I told the demo in the mail, I guess I'll write it o=
n the=20
index.html. It's user demo pwd demo, tell me if anything goes wrong.

Luca

On Monday 20 January 2003 17:57, you wrote:
> Luca -
>
> I am interested to see the FastDraw demo you mentioned
> in the note to the OpenFlow group.  We are working on a
> project in California with OpenFlow - actually we are just
> about to start the flow development. I tried to see the
> demo but it seems there is a login requirement beyond the
> SourceForge login. Is there some way I could see the demo
> you have?
>
> Regards,
>
> Michael Dunham
> Kestrel Works

--=20
Luca Arnaudo
mail: luc...@re...
Redesign srl
Tel: +39 02 39325385
Fax: +39 02 39314978

[Openflow-dev] From Luca: FastDraw

[Daniele <d.t...@ic...>]

Dear OpenFlowers,

it seems somebody experienced some trouble in submitting a mail to the list: 
here is a mail Luca Arnaudo tried sending to the list. Since the mail bounced 
back to him with a failure message I'm trying to send it to you all.

Luca, I hope your problems with the list will be over soon =). 
Congratulations for the FastDraw idea.

Daniele =)


On Sunday 19 January 2003 17:20, Luca wrote:
>
> [English]
> Dear openflow people,
> I hereby announce the release of FastDraw by Redesign SRL at
> http://www.sourceforge.net/projects/FastDraw
> FastDraw is a small and neat software working under Zope + CMF which
> permits graphical design of processes and flows via web, of course under
> GPL licence.
> Up to now we released the flow part, which supports Idef0 flow design
> simbology. We're now working on the process design plugin, which supports
> Idef3 process design simbology.
> The next job will be to export the designed processes to OpenFlow, our
> first supported and preferred workflow engine.
> There's an online demo at http://217.22.233.34:8280/fastdraw
> We use SVG technology, so to use our demo you have to download the proper
> SVG plugin of your favourite web browser. Up to now we tested the product
> on windows (IE), tests on linux are on progress.
> If you'd like, you're welcome to come and see our demo!
>
> Luca
>
> [Español]
> Queridos openflowsos,
> Esto es el anuncio de la creaciòn de FastDraw de Redesign srl, al sitio
> http://www.sourceforge.net/projects/FastDraw
> FastDraw es un pequeño software para el deseño grafico de procesos y flows
> por web, desde luego bajo licencia GPL.
> Hasta ahora esta completa la parte de flujos, que suporta la
> simbologia/metodologia de diseño de flujos Idef0. Estamos trabajando
> también
>
> sobre el plugin de diseño de procesos, que suporta la simbologia Idef3.
> Nuestro pròximo objetivo serà el export para OpenFlow, nuestro primero y
> preferito engine de workflow.
>
> Hay una demo online aquì:
> http://217.22.233.34:8280/fastdraw
> Utilizamos tecnologia SVG, entonces para poder ver la demo, hay que bajarse
> el plugin SVG del browser, desde http://www.adobe.com/svg/
>
> Si quiereis ver la demo, sois los bienvenidos!
>
> Luca

[Openflow-dev] [Zope-Annce] AdaptableStorage Update (fwd)

[<jm...@so...>]

Sry for the spam ...


---------- Forwarded message ----------
Date: Mon, 13 Jan 2003 12:35:30 -0500
From: Shane Hathaway <sh...@zo...>
To: zop...@zo...
Cc: d.t...@ic..., uw...@sc..., Paul Everitt <pa...@eu...>
Subject: [Zope-Annce] AdaptableStorage Update

A corrected release of AdaptableStorage has been posted.  Friday's
release had an accidental dependency on unreleased Zope code (the
"cannot import name z64" error) and was missing an entire subpackage,
gateway_sql.

Version 0.4.1 corrects these issues.  It also adds a README.txt with a
short step-by-step guide for mounting a filesystem directory into Zope.

Download it here:

http://hathaway.freezope.org/Software/AdaptableStorage

For the CVS savvy, you can also get it from cvs.zope.org at
Products/AdaptableStorage.

Shane



_______________________________________________
Zope-Announce maillist  -  Zop...@zo...
http://lists.zope.org/mailman/listinfo/zope-announce

  Zope-Announce for Announcements only - no discussions

(Related lists -
 Users: http://lists.zope.org/mailman/listinfo/zope
 Developers: http://lists.zope.org/mailman/listinfo/zope-dev )

[Openflow-dev] Re: FastDraw released, demo online

[Luca A. <luc...@re...>]

OPS!
Thanks Alastair, the correct link is http://217.22.233.34:8280/FastDraw
Please notice the F and D.

Bye!
Luca

--=20
Luca Arnaudo
mail: luc...@re...
Redesign srl
Tel: +39 02 39325385
Fax: +39 02 39314978