[Seaside] Re: [RFT] New web project

[Klaus D. Witzel]

On Sat, 11 Nov 2006 10:39:19 +0100, Philippe Marschall wrote:
> 2006/11/10, Jason Rogers wrote:
...
>> I didn't mean to imply that the strategies weren't different.  I was
>> speaking to the decision making process for using an RDBMS.  Rails
>> could have be done with an ODBMS, but then adoption would have
>> severely suffered because:
>>
>>     [1] most folks aren't used to it
>>     [2] it's not as easy to port an existing application
>>     [3] most folks don't have access to an ODBMS as readily as an
>> RDBMS (MySQL, SqlLite, PostGres, etc.)
>>     [4] other reasons.
>
> So basically the same "arguments" that speaks for Java and static typing.

Static typing? RoR is only the diameter of an atom away from constant  
typing *), almost the same as we used it in the early '80s of the previous  
century,

- http://www.support.unisys.com/linc/docs2/eae33/updates/zips/78616075.exe
*) well, this *is* a system written in itself ;-)
(on wintel platforms the .exe self-extracts into a harmless .pdf)

The success of RoR reflects the inverse of its distance from constant  
typing, just like it was for *) and competitors.

> And oh, if you think the persistence layer in rails is abstracted and
> you don't have to deal with it in the model code:
> http://www.firemoss.com/blog/index.cfm?mode=entry&entry=8F2D5D6A-3048-55C9-4315FAAD54617516

People seem to not have learned much 'til the '80s, in *) that's  
impossible even if you'd like to do it.

/Klaus

> Philippe