LabVIEW
How do I make a program that doesn't need to be installed?
That's an interesting idea. It sounds logical it should work in all versions, since all DLLs needed are located there. Now i'm getting curious! 🙂
/Y
As long as you don't do anything fancy it may still work, but chances are that it gets more and more difficult for many LabVIEW features. Obviously don't try to do that with an application that does any DAQmx or other IO driver related operation. That definitely never will work and in fact never has.
@N1LAF wrote:
Sometimes, companies get a little too big for some of these little things... just look what happened to IBM.... NI take note.
IBM is still doing fine. They don't run their lung out of their body to catch the small fish but rather do the heavy stuff where they can garner millions with a project but it's a business strategy that works quite well. You may not like that they don't really worry about small fish like you and me, but hey we are in a free market and every company is free to go after the part of the market it likes most. I
I'm not aware of many companies who do equally well on small scale end user sales and big scale corporate customers. In the end it is "share holders" value for most companies and the customer is more of a liability than an asset with that.
@N1LAF wrote:
"IBM is still doing fine" - You missed my point. When companies stop listening to their customers, they may lose their market share, or even the market. Where is IBM in the PC market, that they basically started? Remember the attitude 'why would computers need graphics and sound'? Then Windows overtook MS-DOS. After that most of their products would not take. PowerPC is not a major player, and OS2 is non-existent. Microsoft may be starting to be at this point, taking features away and incompatibilities with older windows programs in Windows 7(including LabVIEW products), and the poor reception of the public towards Windows 8. Bottom line, in the PC world, IBM is not doing fine.
Companies change directions all the time and it's legit, or do you think 3M should have insisted on staying in mining business just because it was what they started with, just to name one of many examples?
In PC hardware market you can't really make a lot of money anymore. The competition with low wage countries is simply way to hard. So why invest billions of dollars in a business that changes with the week, and delivers margins that go into red in the time the products need to ship from Asia to the west? It's simply no good business and just because they invented the PC doesn't mean they have to cling on it like a desperate child. Markets get evaluated continously and decisions are made about selling off businesses that don't seem important. Sometimes such decisions might seem wrong to an outstander, sometimes they turn out to be wrong, but the alternative is to not make any decisions and go bankrupt for sure. 
If you like those decisions or not is another story but no business has any obligations to do any business they don't feel like doing. We don't live in a communist planning economy as you are probably quite aware of.
As long as you don't do anything fancy it may still work, " - Didn't work for a simple "Hello World" executable in LabVIEW 2009, so that functionality was taken away at that point.
You make it sound like that functionality was taken away intentionally. First NI never really has tried to support it, it just happened to be possible. Second NI has quite a few other things to tackle than intentially trying to find a way to try to break unsupported functionality. It may still be possible but requires more tickling due to increased intercomponent dependencies. Managing those dependencies is already a herculian task with a proper installer, trying to make it work without installer would be an even bigger undertaking and one that there is little business for. Most people want an installer that works with a single double click, not having to manually copy a huge directory with 100ds of MB of files.
They do listen but creating an installerless software is even for less complex applications than LabVIEW a challenge. And that has not been made easier with newer Windows versions. So I would not hold my breath for a fix of this.
As to what may be strategic mistakes or not, is even in hindsight often difficult to decide. Do you measure it by sales, profit, broadness of products or in fact rather the opposite of being the best in just one single thing only? Is being big a valid measure for it? IBM might be out of business if they had insisted on staying with their core business of mainframes at all costs. And most likely would be a minor market player if they had declared PCs as their core business to replace mainframes. And we may still work on 10000$ IBM XT computers if they had initially realized what a market it could get rather than thinking mainframes were still the future and throwing the PC design on the market for anyone to copy.
