Inventory and POS/Cash Solutions
William Park
opengeometry-FFYn/CNdgSA at public.gmane.org
Tue Dec 7 02:20:45 UTC 2010
On Mon, Dec 06, 2010 at 05:55:33PM -0500, Christopher Browne wrote:
> On Mon, Dec 6, 2010 at 5:23 PM, Peter <plpeter2006-/E1597aS9LQAvxtiuMwx3w at public.gmane.org> wrote:
> > Anyway doing all this by hand could be an interesting learning
> > project but also a harsh and possibly painful entry into CRM land
> > (and SAP territory).
>
> Yep.
>
> SAP territory is where "everything's deeply integrated into one giant
> database." There's a lot of neat stuff that can be done with that,
> but it imposes a lot, too.
>
> I rather like the "have independent systems, and pull data from them"
> approach; if you can periodically pull the data underlying the cash
> register tape out, a lot can be done with that, *without* forcing
> writing BDC sessions
> (http://www.sap-img.com/abap/learning-bdc-programming.htm) and such
> awfulness.
>
> If you want tight integration, that's hugely costly. But having a
> bunch of independent islands isn't without its costs, too :-(.
Cost is more labour, of course. But, retail/pos market is so
customized, simply because different merchant has different needs, that
there is no such thing as "standard POS", "standard backend database",
or "standard ERP" solution.
Some of the benefit of separated stages are as follows:
- easier development and debugging -- since each stage has "input" and
"output", you can do automated regression test for each stages as well
as the whole process.
- paper trail that can be verified, checked, and audited -- Programmers
poo-poo this, but I came to appreciate its value only after working in
Retail/POS industry. With "in" and "out" files, you get lots of
things for free, something you otherwise have to write code for. This
is particularly important for customer support.
- easier to upgrade, or change vendor.
Even though my company sells Linux backend and Linux frontend (both
Redhat) for Retail/POS/ERP market, I do admit there is nothing "open"
and "free" about what we're selling. That's because, our business model
is customization and support. Having said that, I do see the need for
"POS screen" for Linux platform. But, the itch is not great enough...
:-)
--
William
--
The Toronto Linux Users Group. Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists
More information about the Legacy
mailing list