LabVIEW new version Every year - Why?

[P@Anand]

I just want to be clear with one point: why LabVIEW version every year?

Well it looks healthy but from a software development side in the industries its bit tough for us to switch to the new version every year?. 

It would be better atleast we have a time of 2 years to play with the new version.

[Darin.K]

Well I guess nobody is making you upgrade, I happen to have LV10 but rarely use it. All of the version stomping and lack of setting portability issues come up yearly now instead of every few years. There used to be a new major version every 2.5 years on average, that means we are due in LV12 since I count LV9 as the last major release. BTW you can see how long it took me to complain here http://forums.ni.com/t5/LabVIEW/Introducing-LabVIEW-2009/m-p/956321#M428752 And you can throw your Kudos away here: http://forums.ni.com/t5/LabVIEW-Idea-Exchange/Revert-to-old-LabVIEW-version-naming/idi-p/1047557

[P@Anand]

But for the new programs we ultimately have to go to the new version as it has lot of advantages. The tough part comes when we need to work on the old and the new as well. we cannot simply upgrade the old as it has lot of procedures inside. But as an individual I welcome the update.

[tst]

---

@P Anand wrote:

But for the new programs we ultimately have to go to the new version as it has lot of advantages.

---

So what if it has advantages? That doesn't mean you have to upgrade. You can wait two years and then get the advantages of both versions, and at least here you have the benefit of all the features that went into the previous version going through testing in the real world, which you wouldn't have if a new version was only released every two years.

[pallen]

There are plenty of reasons for a regular upgrade, but probably the main driving force is new hardware.  NI is also a hardware company, and they've got lots of new goodies coming out all the time. 

Supporting multiple projects that span over many versions can become cumbersome.  Everyone has there own way of doing this.  Either by installing multiple versions of LV on the same PC.  Or by maintaining virtual machines that have the correct LV versions installed.  

The latter method seems to be getting a lot more popular.  VMs are portable and easy to backup and maintain.  Perhaps not adaquate in every situation, but some experienced developers have had great success this way. 

[DianeS]

I think that a new version annually is too frequent, simply because most of my clients don't maintain their SSP and therefore, don't upgrade.  This means I have to maintain more than one machine (virtual or physical -- I do both, actually) so I can continue to support them.  Of course I realize that I can back-save, but none of my clients want to install new versions of Run-Time engine.  Therefore if I originally developed the code in 8.5, I have to build the executable in 8.5.  Therefore I have to maintain 8.5. 

I can't force them to upgrade...and, since what they have works for them just fine, they have a valid argument for not going through the annoyance of upgrading.  As their developer, the burden falls upon me to make sure I can continue to support them.  They don't want to risk anything that holds up their production lines, so they are adamant about not changing anything.  I understand the "it works -- leave it alone" philosophy.  There are enough problems out there without unnecessarily adding new ones, and upgrading LabVIEW every year is considered unnecessary by the majority of my clients.

Then there's the problem of hardware drivers -- the DAQmx drivers which support 2011 don't support 8.5.  So the two versions can't reside on the same machine and still work properly.

Certainly this will continue to be a problem, but it would be a lot less of a problem if the time between new releases of LabVIEW was longer (say, 2 years).  I also find that the vast majority of the changes between versions are oriented towards beginning users, limiting their usefulness to professional developers.  Perhaps that's changing now, since the advent of the Idea Exchange.

My two cents. 

[PaulG.]

When I first heard NI was going to start releasing a new version ever year (LV2009) I had a bad feeling about it. Sure enough the hits just kept on coming. 2009 needed a service pack from the start. 2010 was a bit wonky. 2011 seems a lot more stable but from there to hear it has been anything but fun. I suppose more of us should beta test more and complain less.

[Ben]

---

@DianeS wrote:

I think that a new version annually is too frequent, simply because most of my clients don't maintain their SSP and therefore, don't upgrade.  This means I have to maintain more than one machine (virtual or physical -- I do both, actually) so I can continue to support them.  Of course I realize that I can back-save, but none of my clients want to install new versions of Run-Time engine.  Therefore if I originally developed the code in 8.5, I have to build the executable in 8.5.  Therefore I have to maintain 8.5. 

I can't force them to upgrade...and, since what they have works for them just fine, they have a valid argument for not going through the annoyance of upgrading.  As their developer, the burden falls upon me to make sure I can continue to support them.  They don't want to risk anything that holds up their production lines, so they are adamant about not changing anything.  I understand the "it works -- leave it alone" philosophy.  There are enough problems out there without unnecessarily adding new ones, and upgrading LabVIEW every year is considered unnecessary by the majority of my clients.

Then there's the problem of hardware drivers -- the DAQmx drivers which support 2011 don't support 8.5.  So the two versions can't reside on the same machine and still work properly.

Certainly this will continue to be a problem, but it would be a lot less of a problem if the time between new releases of LabVIEW was longer (say, 2 years).  I also find that the vast majority of the changes between versions are oriented towards beginning users, limiting their usefulness to professional developers.  Perhaps that's changing now, since the advent of the Idea Exchange.

My two cents. 

---

Mega-ditto!

That is almost* exactly my situation approach and thoughts.

You saved me a lot of typing. Thank you Diane.

Ben

* I have four towers in my cube with multiple boot partitions to support BV 2.1 (LV 4.1) throgh 2011.

[DianeS]

Always happy to save you typing, Ben! 

[GuenterMueller]

---

@DianeS wrote:

I think that a new version annually is   [...]  Perhaps that's changing now, since the advent of the Idea Exchange.

---

Me too, Diane. Thanks for pointing it out. Your post is fully fitting.