03-30-2006 11:45 PM
03-31-2006 12:40 AM - edited 03-31-2006 12:40 AM
I agree with that and any application I develop for a customer will be an exe, that's a rule in the company I'm working for and I wouldn't change it ; but here is the point... "who will use the application ?"
becktho a écrit:This is the only way to go. Each testrig I had to write a program for, is controlled by an exe. No LV development environment. Just drivers, LV runtime and the simple exe. No need to ask "how can I remove the run-button" or "how can I avoid the user accessing the block diagram".
Message Edité par TiTou le 03-31-2006 07:42 AM
We have two ears and one mouth so that we can listen twice as much as we speak.
Epictetus
03-31-2006 09:22 AM
03-31-2006 09:42 AM
03-31-2006 11:17 AM
Our company policy is to issue software to the production floor in the form of LLBs made with all block diagrams removed. The vi's are then called by a TestStand like hombrewed sequencer. Therefore no one can change anything. For modifications, the developer changes the original vi's and has to build into a new LLB with block diagrams removed. Actually, we have build tools that make a cab file, and the cab file is extracted into an LLB and supporting files.
Why this method, I don't know. This was dictated long before I got here. I like the exe idea better. In fact, I recently developed a tool to report test set configuration and I made it into an exe. I didn't have to follow the policy because it is not being used to test product. The exe is easier to deploy.
03-31-2006 01:43 PM
Hi
We are working with a team of four people to support about 200..300 researchers (out of 1000). We decided always to give the source code with the software, so that the researchers were able to modify and adapt the programs. If we had to do all changes ourselves it would have taken more then 20 people doing the programming.
When big changes are needed or the mess gets too big we help or write a new starting point.
This makes it possible to have fast changes and we learn a lot of how bad programs can get. That also is the reason that we beleive in software tools that help to structure the program, not only in LabVIEW but also on disk and therfor we hope to see a real program manager in lv8.
The basic problem in research is that you try to find something that requires changing the measurement tool when you see the first measurements.
For repetitive measurement on a production floor or a standard system we also use runtimes but not much during the research phase.
I hope I did not make you too jealous.
By the way it's raining again....
Greetings from a wet country
04-03-2006 10:46 AM
04-04-2006 03:44 AM