>>Well time is money. If you spent 100 hours working on replication in VFP (I doubt you'd have much to show for it after 100 hours, but it's a nice round number) at $50/hour that's $5,000 in programming time. I think that's a very conservative estimate. I think the unlimited user license of SQL server is about $27,000. I'd be surprised if you could write a reliable, flexible replication tool in VFP for twice that even if you only used one of the many approaches SQL Server offers.
>>
>>>Hmmm...just took a look at the SQL Server pricing, wait and see I think.
>
> I agree. In a perfect world, I'd also be using some sort of SQL server. But there are definately cases where SQL Server is just too expensive, or when you need to add replication capabilities to an existing VFP program. I'd venture a guess and say a lot of VFP apps would take a significant amount time to make it work with a SQL server - that's, of course, based on the assumption that the programmers didn't plan ahead using views, etc.
>
> I'm in that boat right now. I would need to buy 10 copies of SQL Server (only 10 users at each site) plus do a near complete rewrite of their software to make things work. I think a product like FoxAudit and a bit of custom programming is going to fit the bill nicely. Granted, replication will only be in one direction - but that's all it calls for in my case.
>
> Having said all that, good luck getting two way replication to work well under VFP ;-)
>
>-Paul


I think if I price my software at $27000 + (what it normally costs), my client base is likely to shrink somewhat :-) - OK, I'm being silly but you get the idea: suddenly, instead of a no per-seat license product (VFP), and shrink-wrapped software I'm into forcing my users to buy SQL Server and related licences: I've got to sell them my product *plus* MS's product to do something that other packages do out of the box (e.g. Access packages, ACT).

SQL server has been suggested by most as the solution - I agree with Paul (well I would wouldn't I :) ), it's not always appropriate nor cost-effective: a one-time cost of (say) 10000 divided by a number of clients over a reasonable period is cheaper than adding 5000 per client, every client. However, it's certainly the majority view and with MS seeming to be pushing us all that way no doubt we'll be having to use it soon anyway.

I think I'll still wait and see: when it's built into every W2K desktop, that's the time to start using it...cheaply.