OO did a bunch of crap like this to me recently. Setting the entire spreadsheet to text still didn’t seem to permanently work. Given the slightest edge due to a change, it would try to “interpret” my text for me. :mad:
I want to make a merely technical point, of no relevance to any of the other discussion.
A SQL server database application consists of a front end – perhaps a web browser – and a back end – perhaps postgreSQL. connected by a channel. The protocol includes a method for selecting transmitting and receiving “records”.
An Access database application consists of a front end (MS Access) and a backend (a file system), connected perhaps by a network redirector, SMB protocol, an SMB service, and a couple of caching systems. The protocol includes a method for selecting transmitting and receiving “records”.
An important difference between the two is that the Access system is optimised for a small number of users, by locally caching indexes (originally, it was characterised for 32 simultaneous users, and had an absolute limit of 256). The SQL Server system is optimised for a larger number of users, and the most important facet of that is that indexes are kept in memory at the server end.
Locally cached indexes mean fast local indexing, but all of the copies have to be updated for any change. Remote indexing means heavy computational load at the sever, but no need to send or receive index updates.
You will occasionally see people claiming that a shared Access database is not a “client-server” database system. If pressed, they will say that a shared Access database is not “true” client-server database system. If pressed, that will say that it’s not “true” because it doesn’t use a client-server protocol, and that the server is not a service. If pressed, they will just ignore you… All this is at best unnecessary, and a worst rather misleading, since when making technical decisions somebody needs to understand what services are running, what protocols are used, what the licensing limits are and how the design decisions affect the use case.
There are other important differences as well, (and not just authentication and security), but none of these are well captured by describing Access as a ‘single process’, or ‘not a client-server system’.
Memory only holds the subset of the database (pages of data and indexes) that can fit in memory at any given moment, unless the database is small enough that the entire thing fits in memory.
Many years ago I created an Access database for my daughter’s Girl Scout cookie orders.
I have found that Access is often a great solution when you have two (or more) systems that don’t “talk” to each other but each have the capability to export Excel files. I can usually find a way to “marry” the various files into a cohesive report. Of course my preference would be to get the company to invest in SAS but when you have a week to deliver and no budget Access works very well.