Well I for one am in for whatever, because I don't own any other DAW software, and even if I could afford it, I sure as hell don't have the time to learn it.  So I took the survey and used the money for a snazzy new Cakewalk t-shirt I can wear while reading spiky forum threads about phantom business practices and the pros and cons of public stock options and their relation to customer satisfaction.  This is great stuff....

When is Sonar X4 coming out?
 
The astrologics  say that X4 will be announced on saturday,
october 31 and released at  the 1. december 2014.
 

+1 Craig.
The free version of Survey monkey handles 100 users at a time. I doubt it would be beneficial to ask more than 1000 people (would anybody here like to spend time reading the comments of a 1000 people and send out 10 surveys?). . Also the whole point of a survey is to question customers not cakewalk. Finally it's probably none of our own darn business how cakewalk chooses to operate internally. As customers all we should care about is the product, the support around it and whether or not they strangle kittens in their spare time.

Finally Sonar will be released 1st April 2015. Perhaps.

CakeAlexS
As customers all we should care about is the product, the support around it and whether or not they strangle kittens in their spare time.

You'll be happy to know no animals were harmed in the making of Sonar X3, and the grays were very cooperative once we told them that kittens were not to be used as lawn ornaments. 

I've worked closely with both statisticians and actuaries in healthcare.  stat folk like representative samples, actuaries like the whole set of data.  the actuarial method works well with required data that is completely collected - things like procedures administered over time.  stats are a different beast.  if you ask the whole user base to take the survey you wind up with skewed results because the hardcore evangelists of the product will respond disproportionately.  typically a smaller random sample will yield better results because you will get a more even distribution of the hardcore users the medium users and the casual users.  
 
 

swamptooth
I've worked closely with both statisticians and actuaries in healthcare.  stat folk like representative samples, actuaries like the whole set of data.  the actuarial method works well with required data that is completely collected - things like procedures administered over time.  stats are a different beast.  if you ask the whole user base to take the survey you wind up with skewed results because the hardcore evangelists of the product will respond disproportionately.  typically a smaller random sample will yield better results because you will get a more even distribution of the hardcore users the medium users and the casual users.  

 
The things I learn from this forum...
 
I really hope Cakewalk finishes the forum-to-brain download adapter in time for X4.

There are three kinds of lies.
Lies, damned lies and statistics.

Wow lots of speculation and people reading between the lines and jumping to conclusions. As Andrew said it was just a survey to understand how our users use the software and understand expectations.
A primary intent of ours going forward is to improve agility. As many will attest there are situations where some N months down the road a problem will be detected by some users that impedes their workflow. Or more commonly, it could be a new plugin in the field that has some compatibility issues due to something in SONAR. 
 
I'll share some background into what we do today. In the model we (and may other retail software companies) have today, there is a major release at the end of the year and then a few maintenance releases following where we address any issues that might come up. We also may occasionally add a few new features in these point releases. After or during this period which is typically about 3-4 months post release, we transition fully to development of the next version. After this point it becomes extremely expensive for us to change gears and release an update for a single issue. A product plan is already scheduled and any change like this causes disruption and delays. To ship an update even if it has a single change has repercussions to development, QA, tech support, marketing, documentation etc. This is primarily why we don't do this after a few months post release. As some would attest, this model has can be a source of frustration to users who use the software for a living where an incompatibility can cause loss of workflow. TBH it has been a source of frustration for us in development as well - i.e. often we know how to fix something but can't get it to our users easily or quickly because of the above constraints!
 
We have experimented with a more agile "quickfix" system in the past but it didn't work very well primarily because a lot of thought didn't go into the infrastructure to do this well. So we wanted to rethink how we can design a more fluid system that provides value to our users and also works well for product planning and resource allocation internally. I think if we can figure this out it will be a huge benefit to users. i.e. A system where we can be more responsive to problems *and* provide value on a more frequent basis would solve many issues that the current "waterfall" based product life cycle model cannot possibly address. Its a bit ironic that many companies follow a pretty agile development model and yet the product life cycle still follows the classic yearly release cycle model that has been around since the beginning of time! I think most people will agree that at minimum its time to rethink this. That was the the main purpose of that question :)

Sanderxpander
There are three kinds of lies.
Lies, damned lies and statistics.

Statistics aren't lies, they just can be used to support lies and/or damned lies.
 
Regards,
Dan

Cakewalk's openness is always welcome and makes it easy to figure out how our needs and desires interact with yours.  However, there seems to be one aspect of all this that isn't getting much consideration, and that is the buy once and live with it for a long time scenario.  As I mentioned somewhere else, I bought Office 2010 Pro and it would never occur to me to upgrade it unless an eventual new computer/OS I had to buy could no longer run it.  This is the way I've always bought software.
 
I bought Sonar in the same frame of mind, after having enjoyed using Kinetic 1 for what, 10 years ?  My needs were simple and it worked just fine.  I don't even remember if there were ever any updates to it, I just bought the box in the store and used that.  Now Sonar is a lot more, but it also seemed to have a lot more wrong with it, back at the beginning of X2 (I got in through the tail end of X1).  Sonar to me has still not met my expectation of buy it and learn to live with it, though X3 may be very close (the sustain pedal issue has me worried).  Reading so much user feedback on these forums certainly distorts my opinion from what I'd form by myself with my current limited use, but the format already gives the impression that it's a type of subscription model.
 
The problem I have with a real subscription model, at least the way I imagine it would be and I'm not presuming what Cakewalk will do, is that it involves constant cash outlay.  I'm not yet on a fixed income (it's actually worse than that at the moment) but I've always bought software with the idea that it's a tool in my toolbox, to be used the best way I can.  I don't spend new money every year on my hammer which I bought 20 years ago.  I can easily imagine a professional situation where a subscription model would be desirable, but I've always thought of Cakewalk as being more of a musician / hobbyist oriented company, the people around here certainly give that impression.
 
Andrew Rossa's remark in the survey thread about perpetual licences is reassuring (though I'm not sure what that means exactly).  My only real fear is one day finding myself with a tool with a show stopping issue for which the only way to have it fixed is to jump into a perpetual payment plan.