Sorry Ed I must have responded to your reply to the author on this one.
>
>Second, there's a considerable expense in infrastructure to implement thin client solutions, like the terminal servers themselves and the necessary software licenses.

Agreed but considerable lessing of software development costs.


>If the existing workstations are adequate and the LAN is not a significant bottleneck, there's little compelling argument for thin-client solutions.

Agreed however the WAN is a compelling argument.


>In general, software that won't deploy under NT/Win2K at all elminates WTS as a platform, it requires considerably more administration than an existing network of Win9x boxes, and introduces new issues in deployment. It's not a panacea.

Um not sure this is generally accepted yet.

>Burdening the terminal servers with heavy I/O can kill them,

Agreed so no GIS/CAD systems on them for a while, but that is not the bulk off what we do.

>and deploying other server software, like IIS/SQL Server/SMS on the same box >has a significant impact on the performance of the terminal servers.


Thin client doesn't make the need for C/S environments any less necessary; if you need the added services of of a database backend, nothing has changed; VFP still doesn't secure tables, implement a security model stronger than file-level access control enforced by the OS, and the data files are just as fragile.


Agreed but for the smaller scale applications where VFP still has a strong position ie less than 10 users then the security problems are no worse than before and an imdeiate benefit has been gained.

Though you are quite right I should have qualified my statement a bit more.

I is not the end of C/S or even the beginning of the end, but it may be the end of the beginning:)