johnf wrote:
> On Wednesday 31 December 2008 04:42:45 am Wolfgang Keller wrote:
>>> There is also a major addition to the framework: the ability to
>>> deploy Dabo applications as true web apps.
>> To waste the effort just for Yet Another Web Framework was A Really Bad
>> Idea.
>>
>> Some things that would be actually useful instead:
>>
>> - Get rid of XML. Use only Pure Python instead. XML configuration files
>>   are for static languages like C++ or Java that don't provide other
>>   means to configure something at startup- or runtime. Python can use
>>   itself for that. And it's way better than XML, bvoth in terms of
>>   readability and power.
>> - A rich notification system which allows:
>>   - to notify other clients of inserts, updates, deletes so that all
>>     clients are always in sync (client->client). See the modeling ORM
>>     for an example: http://modeling.sourceforge.net/
>>   - to keep all clients in sync with the database server, WITHOUT
>>     POLLING, obviously (server->client). See the BaseTen framework for
>>     an example: http://basetenframework.org/
>> - Include the HyperTreeList widget:
>>   http://xoomer.alice.it/infinity77/main/HyperTreeList.html
>>   for example to do what the XTreeWidget can do in xTuple:
>>   http://www.xtuple.org/index.php?option=com_wrapper&Itemid=70
>> - Include the wxAdvTable widget:
>>   http://wxcode.sourceforge.net/components/advtable/
>>
>>> Imagine: being able to develop a rich internet app using nothing but
>>> Python on both the client and server!
>> Who cares for fat, slow, clumsy and un-functional trendy web crap.
>> Managers and administrators do. Users who need to get actual work done
>> efficiently rather run away screaming wildly as soon as they hear the
>> words "web application".
>>
>>> It's still early in the development process,
>> And I thoroughly hope that the development process will be stopped NOW.
>>
>>> so it's lacking a lot of the supporting tools, and almost no
>>> documentation has been created, but that will be coming in the next
>>> few weeks/months. When you deploy your app as a web app, all data
>>> access and business logic is on the server, and the framework
>>> automatically handles the communication between the client and
>>> server.
>> It can't, because HTTP is a totally braindead purely asynchronous,
>> client->server one-way-only protocol.
>>
>> Definitely the least suitable client-server protocol that one could
>> imagine for a database GUI application. But then, HTML isn't an actual
>> GUI framework, so abusing it as such is bullshit anyway.
>>
>>> The framework also automatically grabs file changes from the
>>> server, making UI updates seamless and quick.
>> Yeah. Welcome to POLLING. Technology from the early 80's. JUst a
>> billion times more bloated to make the applications unusably slow on
>> ANY hardware.
>>
>>> Lots more interesting stuff will be happening in this area in the
>>> near future, so stay tuned!
>> It would be extremely sad to see any further development effort wasted
>> for just Yet Another Web Framework, while there is NOT A SINGLE
>> actively developed cross-platform framework for actually usable (i.e.
>> non-web) database applications available for Python any more.
> 
> When I speak with others about Dabo they always ask if there is a web 
> solution 
> too.  So from my point of view there is a need.  
> 
> Dabo does not require XML files.  You can hand code anything in the 
> FrameWork.  
> 
> If your intent was to better Dabo in some way I'd have to say you failed 
> because I was offended by your last statement. 
> 
> 
> 
Hello Everyone

I am confused over this argument. Midwest brain freeze, I guess!!

I looked for years, hunting a "NON-WEB" centric framework to use
with Open Source Databases. It seems many Dabo users went through the same
process ending up finding wxGTK, Python, wxPython, Dabo. I use Rekall, also,
because I found it first. However it is no longer actively developed and
is an more of an application not a framework.

Having found Dabo, has been a unique experience. The group of developers
have been excellent in making up for the lack of polished documentation.
People very seldom have to wait for an answer. If there is a problem, they
quickly work to fix it or train you in what is happening.

Babo is completely extensible by the end-developer with the complete wxPython
and Python frameworks available.

While I do not like Web apps, I think the addition of what has been added to
Dabo is a good thing, even if I might never use those features.

Ed and Paul should be commended for their investment in this project.
So many others help. For many of us, it is EXACTLY what we could not
find anywhere else.

The rapid and open development cycle is both good and bad. It is nice
to have features as soon as possible, but sometimes it is hard to pick a point
to jump in and put out a stable release to clients.

End of soapbox ---------------------

Thanks Ed and Paul and everyone else that makes Dabo what it is!!!!!!


Paul McNary
[email protected]










































_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-users
Searchable Archives: http://leafe.com/archives/search/dabo-users
This message: http://leafe.com/archives/byMID/[email protected]

Reply via email to