• SONAR
  • NO MORE Monthly updates for me rant (p.10)
2015/12/04 22:39:33
Doktor Avalanche
John
No its an endless argument. One that at present I see no purpose for. Heck I'm not even sure what the thread is about anymore. 


Well then please stay out of the thread if you feel that way. Stop rocking the boat, the only aggression so far I'm reading is from you, it's been a fairly respectful, sensible and calm discussion, stating all sorts of viewpoints. Up until you joined in. You are clearly out to polarise once again.
2015/12/04 22:43:53
kevinwal
Let me try it another way.
 
Your notion is that a no-features stability release will carry fewer defects than the current release process, and it's an attractive idea. That might be true only because you won't be shipping any of the defects in the excluded new features. There will still be an appreciable number of brand new defects in the fixes-only codebase. The more defects that are fixed in a release, the more the potential for new defects. This is the point I believe Noel was making.
 
As you know, regression tests only test aspects of the code that were broken and then fixed. Unfortunately there are many usage contexts for which no test has yet been created, and regression tests will not uncover them because they are not regression defects. They are brand new. 
 
To make things more interesting, as more fixes are applied, the more defects that subsequently arise tend to manifest themselves in increasingly complex and subtle ways. These defects also fall outside the scope of regressions and are damned difficult to write unit tests for.
 
Of course, over time, these issues are eventually addressed, additional regression tests are created and product managers then nervously declare that portion of the software to be probably mostly somewhat reliable. Maybe. :)
 
Broadly speaking, since the number of bugs in a proposed "stability release" may well not be appreciably fewer than in a standard release, the decision to do one over another is not a technical decision, it is a business decision.
 
Fix-only releases tend to encourage deployment of the software by the existing customer base, which is of course a good thing. Microsoft was forever fighting to get enterprises to deploy new releases of software, so service packs were key to making that happen. For smaller, more revenue constrained outfits, it's the new features that matter. Balance.  
 
I once saw some math that covered a lot of this but I don't remember a bit of it.
 
2015/12/04 22:55:28
kevinwal
Doktor Avalanche
Anderton
But I think Cake's track record of fixing bugs that ride along with new features is, frankly, excellent. It's rare that a problem persists past the next update.


And I'm going to totally agree with you here Craig. Regression issues get fixed with new features pretty well.

The crutch of the issue is they get packaged with brand new features that in turn need their regression issues fixed every single month without fail. Roll back and you are in a similar scenario just with different regression issues to cope with (which become unfixed as you have rolled back).



I may be wrong about this, but you seem to be confusing regression issues with features that once worked, were fixed but are now broken again. They are different, but kind of the same, but not.
 
A regression issue is specifically an area of code that once worked but now does not. A feature may be broken by a bit of code, then fixed and have a regression test for it, then in the next round be broken again by completely different code that manifests itself in exactly the same way as it did with the earlier defect.
 
For the latter, no existing regression test exists, so additional regression testing will not uncover the defect. Rather, the defect must be discovered, then repaired, then a regression test developed for the new case.
 
This is a real challenge for small companies doing effective fixes-only releases.
2015/12/04 23:05:49
John
Doktor Avalanche
John
No its an endless argument. One that at present I see no purpose for. Heck I'm not even sure what the thread is about anymore. 


Well then please stay out of the thread if you feel that way. Stop rocking the boat, the only aggression so far I'm reading is from you, it's been a fairly respectful, sensible and calm discussion, stating all sorts of viewpoints. Up until you joined in. You are clearly out to polarise once again.

Nonsense. I'm expressing a view too.  A view that this thread has become tedious. It may not fit your agenda but it is a view.  
2015/12/04 23:07:26
kevinwal
John
Doktor Avalanche
John
No its an endless argument. One that at present I see no purpose for. Heck I'm not even sure what the thread is about anymore. 


Well then please stay out of the thread if you feel that way. Stop rocking the boat, the only aggression so far I'm reading is from you, it's been a fairly respectful, sensible and calm discussion, stating all sorts of viewpoints. Up until you joined in. You are clearly out to polarise once again.

Nonsense. I'm expressing a view too.  A view that this thread has become tedious. It may not fit your agenda but it is a view.  


lol, it's a subject only a geek could love. For the sanity of others I'll stop now.
 
Oh, and to make it clear, I love the monthly releases as they are now. I'm one stupidly happy Cakewalk customer here.
2015/12/04 23:12:11
Doktor Avalanche
kevinwal
Let me try it another way.
 
Your notion is that a no-features stability release will carry fewer defects than the current release process, and it's an attractive idea. That might be true only because you won't be shipping any of the defects in the excluded new features. There will still be an appreciable number of brand new defects in the fixes-only codebase.


I'll paste this again..

Doktor Avalanche
And it's iterative as well, meaning if the release brings up new bugs, they get fixed straight away until the release is deemed stable.


kevinwalThe more defects that are fixed in a release, the more the potential for new defects. This is the point I believe Noel was making.


Noel was referring to old codebase in the paraphrase, not new.

Noel was specifically saying it's much easier to keep things stable when fixing issues with NEW features rather than with old. I did supply quotes earlier. May I ask a favour in future that you quote rather than paraphrase it really does help avoid looping :)

kevinwal
Broadly speaking, since the number of bugs in a proposed "stability release" may well not be appreciably fewer than in a standard release, the decision to do one over another is not a technical decision, it is a business decision.


You are darn right there. And it's exactly the same for the customer. What they are going to be prepared to cope with or not. Without a stability release path I suspect some customers may end up voting with their feet. Others customers of course will forgive regression issues because they have shiney new features. I don't think either profile of customer is 'right' or 'wrong' (sadly too often the topic of conversation in these forums) but do I believe both profiles need to be catored for otherwise there could be more even pain along the way. I hope cakewalk sees the business argument here.
2015/12/04 23:21:04
Doktor Avalanche
John
Nonsense. I'm expressing a view too.  A view that this thread has become tedious. It may not fit your agenda but it is a view.  


You of all people should realise those sorts of statements really do get up peoples noses. Your 'view' isn't contributing anything at all except to derail the discussion as you well know. The very fact you've now mentioned the 'agenda' word speaks bucketloads, further pointing fingers, polariasing people and stiring up the muck. I guess this is what you want.

Are none of us are allowed to have a conversation unless you agree with it?

Can we please now continue with the thread sir??
2015/12/04 23:25:23
kevinwal
Doktor Avalanche
kevinwal
Let me try it another way.
 
Your notion is that a no-features stability release will carry fewer defects than the current release process, and it's an attractive idea. That might be true only because you won't be shipping any of the defects in the excluded new features. There will still be an appreciable number of brand new defects in the fixes-only codebase.


I'll paste this again..

Doktor Avalanche
And it's iterative as well, meaning if the release brings up new bugs, they get fixed straight away until the release is deemed stable.


 
This is done before the release, new code or old. Once it's shipped, it is what it is. And deeming a release stable is not an arbitrarily objective measure. It's based on too many factors to recount here, but the prime directive is Ship. 
 
Doktor Avalanche
kevinwalThe more defects that are fixed in a release, the more the potential for new defects. This is the point I believe Noel was making.


Noel was referring to old codebase in the paraphrase, not new. I've supplied the quote earlier.

Noel was specifically saying it's much easier to keep things stable when fixing issues with NEW features rather than with old. I did supply quotes earlier.

Yes, and so was I.
Doktor Avalanche
May I ask a favour in future that you quote rather than paraphrase it really does help avoid looping :)

No, I'm much too lazy to go to all that bother.
Doktor Avalanche

kevinwal
Broadly speaking, since the number of bugs in a proposed "stability release" may well not be appreciably fewer than in a standard release, the decision to do one over another is not a technical decision, it is a business decision.


You are darn right there. And it's exactly the same for the customer. What they are going to be prepared to cope with or not. Without a stability release path I suspect some customers may end up voting with their feet. Others customers of course will forgive regression issues because they have shiney new features. I don't think either profile of customer is 'right' or 'wrong' (sadly too often the topic of conversation in these forums) but do I believe both profiles need to be catored for otherwise there could be more even pain along the way. I hope cakewalk sees the business argument here.

 
And that's where the personal decision for you, the customer comes in, and thus the purpose of this thread. If you decide to bail on Sonar because Noel & Co. will not adopt your recommended release process, well, that's your call, Doktor.
 
I can tell you that from a product management team's perspective the decision on what to ship and not ship and how to do it is one that is agonized over. It's certainly not a decision that's taken lightly. So I respect the decision Cakewalk has made. It is what it is.
 
PS: I'd also add that we're barely a year into this model. From sitting outside, I'm quite impressed with the quality of the releases and the rapid turnaround of fixes to breaking changes. If we were four years in, I might change my mind, but for now, nope.
 
Cheers.
2015/12/04 23:34:44
Doktor Avalanche
kevinwal
I can tell you that from a product management team's perspective the decision on what to ship and not ship and how to do it is one that is agonized over. It's certainly not a decision that's taken lightly. So I respect the decision Cakewalk has made. It is what it is.


You don't have to tell me, I've worked for several software houses I agree they definitely are agonised after. You are probably aware of the politics between marketing / qa and developers as well...

When you say though 'I respect the decision Cakewalk has made. It is what it is' you are implying they are some sort of immovable force that will not be moved, that hardly makes good business sense.

I don't think they are myself, eventually they will adapt to market demands, one hopes they are able to predict rather than play catch up. BTW as I keep saying I do like the monthly release cycle if people think I'm disagreeing with it. Not my point at all.
2015/12/04 23:44:19
kevinwal
Doktor Avalanche
kevinwal
I can tell you that from a product management team's perspective the decision on what to ship and not ship and how to do it is one that is agonized over. It's certainly not a decision that's taken lightly. So I respect the decision Cakewalk has made. It is what it is.


You don't have to tell me, I've worked for several software houses I agree they definitely are agonised after. You are probably aware of the politics between marketing / qa and developers as well...

 
I was a product manager at a company many years ago, and I brokered many wrestling matches between the interest groups. So yeah. But both teams always wanted the best for the company, they just refused to acknowledge that they were both right.
 
Doktor Avalanche
When you say though 'I respect the decision Cakewalk has made. It is what it is' you are implying they are some sort of immovable force that will not be moved, that hardly makes good business sense.

 
Strawman argument, Dok. No, that's not what I'm saying. I am saying that they've made a decision and they will stick with it as long as they perceive that it's in their interest to do so. For example, if all their customers abandon the product for Bitwig or some such thing, they'll certainly make a move. But they haven't so far, so they must have some preliminary indication that things are going pretty well. And while they certainly listen, one or two noisey customers in a crowd of pretty happy ones won't cause them to spend a boatload of money to retool.
 
Doktor Avalanche
I don't think they are myself, eventually they will adapt to market demands, one hopes they are able to predict rather than play catch up. BTW as I keep saying I do like the monthly release cycle if people think I'm disagreeing with it. Not my point at all.

 
Yeah, I don't hear you complaining about the release cycle, only that they're not doing it right.
© 2026 APG vNext Commercial Version 5.1

Use My Existing Forum Account

Use My Social Media Account