Chris G wrote:
Exactly, you can't use software to fix something that isn't working to start with. This is a very common error/problem and is often the reason for major software projects failing.
I think it is working at the moment, it's just not as efficient as it could be (nor as scalable as they want it to be) and they're looking to improve it. If I'm going to design something from scratch then it'll probably very much be a case of moving their existing paper-based system to a web-based (intranet) system to make it easier to generate reports etc.
But while I am able to do something like this from scratch I think it may be cost prohibitive and if there's an existing solution (or something that's close to it) I'd rather put my effort into adding whatever is necessary to that instead. It seems to me that this is a relatively common requirement amongst groups who would be unlikely to be able individually to afford a bespoke system and where a FOSS solution is ideal, and I'm constantly amazed just how many niche requirements already have a FOSS solution working.
In the meantime I'm looking at Xataface[1] as a quick way to build a database-based web package that gets close to what they want.
[1] xataface.com
Dennis Dryden wrote:
Sounds like they need more then just software they need a whole system! I'm guessing your working as there a systems analyst though? What do you have control to change and whats the current setup? For example can they have a unique identifier printed on each bag/leaflet? Do they have the people to process them when they come back?
I think this would be the right approach for a major organisation (eg if I was writing something for the Post Office to use) but is overkill for the scale involved here.
All I think I want is a system which allows manual route entry (by description, eg area, street name and house number range), a way of allocating routes to people, and a way of ensuring that all routes that get bags delivered get a collection on the date specified on the bag. Adding feedback to that (how many bags were collected in each area) should be fairly straightforward. From there, everything will work around printing out a "job sheet" for each person/group of people, and data entry based on what is scribbled on the printout by the collectors.
for the barcode stuff there is gnu barcode[1], kbarcode[2], libdmtx[3].
Although overkill for this job, these are useful to know about - thanks!