09-28-2006 09:00 AM
09-29-2006 04:04 AM
Hi,
I realise that this has been going on for quite some time now, however the matter is still being investigated at an R&D level at this time.
Thank you for your continued patience
Sacha Emery
National Instruments (UK)
11-20-2006 10:54 PM
11-21-2006 12:38 AM
11-30-2007 01:47 PM
12-05-2007 12:31 PM
12-05-2007 02:43 PM
12-06-2007 03:39 PM
03-10-2008 03:31 PM
"The problem is that the failure occurs in InstallShields application and we can't debug into that application to see what's going on. Its not our source code. You would need to contact InstallShield and get them to debug into their own application and see what's failing. Then if they can show us that our merge modules have a bug in it, we would gladly be able to fix it."
I'm sorry, but that is not an acceptable response.
That's like Adobe telling its customers to contact Microsoft regarding Photoshop not working with Windows.
First, it's National Instruments' responsibility to get its merge modules to work with the most popular and powerful reference standard of installers on the market, InstallShield.
Second, "InstallShield is a black box" is not a valid software development excuse. Software developers have to deal with black boxes as a fact of life. Windows is a black box. InstallShield is a black box. LabVIEW is a black box. When an issue is encountered with a black box, the approach to take is to narrow down the problem. An empty merge module doesn't cause the problem. But the NI merge modules do. By gradually morphing the empty merge module into the full NI merge module, NI can determine what is triggering the problem and code around it.
This is not some arcane configuration we're encountering. It's InstallShield! And it's been going on for three years.
03-11-2008 09:33 AM