On Fri, 10 Mar 2006 11:42:18 -0600, Rod Engelsman
<[EMAIL PROTECTED]> wrote:
Alexandro Colorado wrote:
How much will it cost to build a PIM for OOo, to be able to connect
with the most popular webservices, and also to have the level of
integration and put OpenDocuments and make Online OpenDocuments and
live documents through webservices.
How many work hours from a team of programmers and how much will this
programmers cost. This might be something that many Sun/Novell
employees might be more aware. But it might be interesting to have an
actual figure. And even more interesting thinking on ways to acomplish
this.
I won't comment on webservices because I don't know much about that
area, but as to a PIM for OOo, a couple of observations are in order.
1. OOo now includes an integrated relational database engine.
not just that, but also an email interface pretty much everything is
included except the POP Interface. This mean that OOo right now has
interfaces for smtp (sending, To, CC, Attachments, etc). Which is why it
wont really cost that much (developer hours persay).
2. At it's heart, a PIM is nothing more than a specialized database
application tied into some communications facilities. This holds for
email clients as well.
Correct even thought you also need to think on the actual services that
needs to plug into. Email doesnt just drop.
So it seems to me that the building blocks at least are already present.
In fact, a clever person could build a functional PIM right now using
Base. The only drawback is that the interface would likely be a bit
clunky and programming the thing in StarBasic would probably make it run
slow.
I think here you are wrong, StarBasic is only one option and is usually
just used for higher level scripting, Code for OOo can be done in Java,
Python, C# and C++ Which is why I put 15,000 dls in trainning.
Another thing to consider is that an email app built around a database
would be a very scalable entity for corporate use -- much more so that
mbox or maildir -- and if you can set such a thing up to use the native
HSQLDB then it shouldn't be too much of a stretch to hook the thing up
to a corporate DB like Oracle.
It can be both, you see email clients like Opera use Sqllite to manage the
emails and orders and suck but is definetly not holding the email. The
email is stored in different places. The database just manage the way data
is presented. (think on link vs hold).
What I'm saying is that every time this subject comes up in one way or
another, there is a chorus of protest that OOo's limited resources
shouldn't be squandered to produce "Yet Another E-Mail" app
This is true, and is exactly that the reason of mentioning money /
resources. Currently there is a lot of work to trim down OOo, and
integrated into the Linux platform at the same time of creating more
filters to different file format, let alone documentation, localization
and what not.
In the end all this consume a lot of time, so we definetly need more
developers, and developers cost money. The actual point is narrow down how
much money are we talknig about and how affordable or not can it be by let
say 500 SMB business willing to put money on this. My guess is that not
much than 1/2 of a M$O licence.
from scratch. But I think that's simply wrong-headed because 1) you
wouldn't be starting from scratch, and 2) the product that you end up
with could be very much more than "Yet Another" email app.
i will go ahead and think if we can send OpenDocuments as emails (not as
attachments) but as actual email, there was a push of this in the past
usually known as Xmail which wanted to XMLize the email as opposed to
dirty HTML 4.0 which is currently used by Thunderrbird, Outlook and other
email clients.
--
Alexandro Colorado
CoLeader of OpenOffice.org ES
http://es.openoffice.org
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
