Windows 10 Update KB4549951 Causes BSOD For Some Users

[mr moose]

So I searched the forum for news titles about updates issues.

 

https://linustechtips.com/main/search/?q=windows update&type=forums_topic&nodes=13&updated_after=any&sortby=newest&search_and_or=and&search_in=titles

 

And it appears there was a thread about a bug in February then thee next one before that was April last year.   Then before that was January.  So unless people aren't creating threads every time there is an update bug/issue it isn't every month.

[Delicieuxz]

Just now, mr moose said:

So I searched the forum for news titles about updates issues.

 

https://linustechtips.com/main/search/?q=windows update&type=forums_topic&nodes=13&updated_after=any&sortby=newest&search_and_or=and&search_in=titles

 

And it appears there was a thread about a bug in February then thee next one before that was April last year.   Then before that was January.  So unless people aren't creating threads every time there is an update bug/issue it isn't every month.

People don't create a thread every time there's an issue with an update.

[mr moose]

9 minutes ago, Delicieuxz said:

Huh? Microsoft states they're testing patches through Insider and Windows Update, just like they states they've laid off a huge number of programmatic testers. And offloading testing onto developers doesn't accomplish the same thoroughness of testing as having dedicated testers.

 

Not security updates. and the windows IP is only one of 7 QC steps.  please read the link I provided.

Quote

It doesn't have to be assumed because it's stated: "Prior to these cuts, Testing/QA staff was in some parts of the company outnumbering developers by about two to one. Afterward, the ratio was closer to one to one. As a precursor to these layoffs and the shifting roles of development and testing, the OSG renamed its test team to "Quality." "

 

Again, QC has not shrunk, it has been moved. Instead of having 2 testers for every dev they  basically made the dev test his own work.  and had a tester for the end product. This is not an unreasonable thing when you have bloating that doesn't actually provide anything.  No point in have 2000 testers if they aren't testing anything.

 

Quote

And: "The intent of these releases is to provide visibility into, and enable testing of, the non-security fixes that will be included in the next Update Tuesday release"

 

And other statements clarifying that Microsoft is out-sourcing testing to unpaid Windows owners, including the existence of the Insider program.

 

As for where the testing jobs went: "For those of you whose jobs are impacted by this, I want to thank you for your contribution to Microsoft and our customers, and wish you the best."

 

As for whether the quality of Windows updates has degraded...

 

January 2018: Windows 10 will not start/boot after windows update

March 2018: Total Meltdown?

October 2018: Microsoft's latest Windows 10 update is reportedly wiping user data

October 2018: New Windows 10 1809 bug: Zip data-loss flaw is months old but Microsoft missed it

October 2018: HP users report BSOD after Tuesday patch

November 2018: Hold off installing re-released Windows Oct Update - networked-mapped drives not working, display driver problems

November 2018: Microsoft Blocks Windows 10 Version 1809 on Some PCs Due to Bad Intel Drivers

November 2018: Microsoft confirms File Association bug in Windows 10 will be patched on Tuesday

November 2018: Re-released Windows 10 October 2018 Update breaks Apple iCloud

November 2018: Et tu, Office? After pulling Windows 10 update, Microsoft does the same for Office

November 2018: Latest Windows 10 update breaks Windows Media Player, Win32 apps in general

December 2018: Microsoft pulls Windows 10’s optional November update following BSOD reports

March 2019: Latest Windows 10 Update Kills Performance in Some Games

 

 

And on, and on, and on...

 

So over the space of a year you had 5 issues and most of them were 3rd party caused, and you want me to believe that things have gotten worse specifically because of something MS did or are not doing.  How many update issue were their before these apparent cuts to QC?

 

EDIT: and I ask again, how many people did they effect? what was the failure rate?

[mr moose]

3 minutes ago, Delicieuxz said:

People don't create a thread every time there's an issue with an update.

But they are certain it happens al the time.  Even your list did not have issues for 8 forced updates.

[StDragon]

https://www.askwoody.com/

LOL, love the "MS-DEFCON" (Microsoft Patch Defense Condition) level. Currently at lvl 2 btw.

 

 

[Salv8 (sam)]

Oh shit my computer has downloaded and installed it...

oh fuck...

2 hours ago, Rainbow Dash said:

Now not only many users are reporting that this update fails to install, but few other users who installed it are even facing serious issues like deletion of files from C drive, disabling of microphones, camera & USB ports, defender shutting down and many other issues like slow performance.

thank god windows allows me to uninstall updates!

i'm disabling windows update for a good while!

yay! livin on the edge!

[Delicieuxz]

15 minutes ago, mr moose said:

Not security updates. and the windows IP is only one of 7 QC steps.  please read the link I provided.

 

Again, QC has not shrunk, it has been moved. Instead of having 2 testers for every dev they  basically made the dev test his own work.  and had a tester for the end product. This is not an unreasonable thing when you have bloating that doesn't actually provide anything.  No point in have 2000 testers if they aren't testing anything.

I don't know why you'd think that all those fired OS testers weren't testing anything. Offloading bug-testing onto developers only increases the burden of developers and means they have to divide their time between developing and testing, which means that there is a cut to both efforts.

 

There is a large net loss in testing happening in that situation. Cutting testers per developer from a ratio of 2:1 to 1:1 means that the developers would have to work 100% more hours with 50% of their time dedicated to testing in order to make-up the difference.

 

And it also has to be factored in that with Windows 10, Microsoft are making changes to the Windows OS at a rate that never before happened. So, there's a much larger volume of changes to be tested. This means that Microsoft is now testing more software changes and additions than ever before, but with fewer testing hours than they previously were employing.

 

Quote

So over the space of a year you had 5 issues and most of them were 3rd party caused, and you want me to believe that things have gotten worse specifically because of something MS did or are not doing.  How many update issue were their before these apparent cuts to QC?

None of the issues were caused by 3rd-parties, because they were caused after Microsoft updated Windows and broke 3rd-party software functionality with their changes.

 

That list, which I put together from news articles mostly in this forum, also hasn't been checked for comprehensiveness.

 

13 minutes ago, mr moose said:

But they are certain it happens al the time.  Even your list did not have issues for 8 forced updates.

I don't know what people assume, but the fact is that there are routinely serious issues with Windows updates. Some larger than others. And I bet that if you list the 8 forced updates you are referring to, that there are some additional issues to be identified that were reported with them.

[mr moose]

1 minute ago, Delicieuxz said:

I don't know why you'd think that all those fired OS testers weren't testing anything. Offloading bug-testing onto developers only increases the burden of developers and means they have to divide their time between developing and testing, which means that there is a cut to both efforts. There is a large net loss in testing happening in that situation.

 

Because no one has given me any statistic to show the number of issues has risen or fallen or done anything.

1 minute ago, Delicieuxz said:

None of the issues were caused by 3rd-parties, because they were caused after Microsoft updated Windows and broke 3rd-party software functionality with their changes.

That's not how this works.  Updates are necessary and  is actually quite well know that 3rd party apps frequently are the cause of hte problem.  remember KB4524244, which was kaperskys fault:

 

1 minute ago, Delicieuxz said:

That list, which I put together from news articles mostly in this forum, also hasn't been checked for comprehensiveness.

Then why did you post it?  trying to prove a point with information you haven't validated?  why am I not surprised.

1 minute ago, Delicieuxz said:

I don't know what people assume, but the fact is that there are routinely serious issues with Windows updates. Some larger than others. And I bet that if you list the 8 forced updates you are referring to, that there are some additional issues to be identified that were reported with them.

And you still haven't shown me how many, what the failure rate is etc.  How can you claim there are more issues and that those issue are due to one specific thing when you can't even show that those issues are necessarily; higher than before and nor that QC numbers within MS are adequate?

 

 

[Delicieuxz]

51 minutes ago, mr moose said:

Because no one has given me any statistic to show the number of issues has risen or fallen or done anything.

 

That's not how this works.  Updates are necessary and  is actually quite well know that 3rd party apps frequently are the cause of hte problem.  remember KB4524244, which was kaperskys fault:

 

Then why did you post it?  trying to prove a point with information you haven't validated?  why am I not surprised.

And you still haven't shown me how many, what the failure rate is etc.  How can you claim there are more issues and that those issue are due to one specific thing when you can't even show that those issues are necessarily; higher than before and nor that QC numbers within MS are adequate?

The information I've posted has proven that what I've said is true. I can't do anything about you not liking that fact. Your protestations are coming from a position that is simply false.

 

Something isn’t right with Windows 10 testing

Quote

None of this is excusable. I wrote last Friday that issues like the webcam problem would "inevitably recur" due to the problems of Microsoft's current testing regime: lack of internal testing (the people who did this were laid off); Insiders not testing on real systems (because they're advised not to use it on their primary PCs); and Insiders tending to give poor feedback (they're not professional testers, and Microsoft's very weak release notes give no indication of what things have been changed and hence need testing in the first place).

 

Critics warn Microsoft it needs to fix broken update process

Quote

To many, there is a straight line between Microsoft's 2014 dismissal of a large number of internal testers and the decline in quality. In response, Microsoft points to Windows Insider, the program that previews feature upgrades - but not the monthly security and non-security updates - to a large pool of Windows 10 volunteers. The 15 million Insiders, the plan goes, report problems, which Microsoft engineers fix before the code reaches production PCs.

 

Two Windows 10 feature updates a year is too many

Quote

After watching Windows 10 in the real world for more than three years, I am convinced that the current incarnation of "Windows as a service" is unsustainable and needs to change.

 

Microsoft needs to refocus on Windows 10 fundamentals, not just new features

Quote

Windows as a service, as it currently exists, isn't working. Microsoft's plan to release two feature updates to Windows 10 annually, as Microsoft has been doing since 2015, is not seen as a plus by many customers. After the debacle around the Windows 10 October Update/1809, the perception that Microsoft has lost control of its Windows 10 update strategy is even more apparent.

 

Windows as a Service Isn’t Working

Windows Isn’t a Service; It’s an Operating System

An open letter to Microsoft management re: Windows updating

Microsoft’s problem isn’t how often it updates Windows—it’s how it develops it

 

 

Anyone having used Windows for the past 15+ years who has paid attention and is being honest with themselves will be aware that post-update issues with Windows have been more prevalent with Windows 10.

 

If you'd like to change the subject from the fact that Microsoft has largely out-sourced the Windows OS beta testing to Windows owners, resulting in a decrease of patch quality, to the subject of exactly what percentage Microsoft's OS updates have degraded by and what the relative issue-rate with them now is, you can. And if that's your interest, then you should also do the research to find out - I'd be interested to know the conclusion, myself. But to try to put that onus on me is you shifting goalposts and attacking a straw-man because your argument that Microsoft hasn't offloaded testing onto Windows owners has already been proven false.

[mr moose]

32 minutes ago, Delicieuxz said:

The information I've posted has proven that what I've said is true. I can't do anything about you not liking that fact. Your protestations are coming from a position that is simply false.

 

Something isn’t right with Windows 10 testing

 

Critics warn Microsoft it needs to fix broken update process

 

Two Windows 10 feature updates a year is too many

 

Microsoft needs to refocus on Windows 10 fundamentals, not just new features

 

Windows as a Service Isn’t Working

An open letter to Microsoft management re: Windows updating

Microsoft’s problem isn’t how often it updates Windows—it’s how it develops it

 

 

Anyone having used Windows for the past 15+ years who has paid attention and is being honest with themselves will be aware that post-update issues with Windows have been more prevalent with Windows 10.

 

If you'd like to change the subject from the fact that Microsoft has largely out-sourced the Windows OS beta testing to Windows owners, resulting in a decrease of patch quality, to the subject of exactly what percentage Microsoft's OS updates have degraded by and what the relative issue-rate with them now is, you can. And if that's your interest then you should also do the research to find out. But to try to put that onus on me is you shifting goalposts and attacking a straw-man because your argument that Microsoft hasn't offloaded testing onto Windows owners has already been proven false.

So still no numbers?

 

Lots of media articles full of money making talk, but no hard numbers.  The closest thing to numbers is a survey run in a forum.  the rest of the articles are re hashed  or articles written about other articles. 

 

However hear are some quotes from the articles you provided:

 

This quote the author admits the overall result is better than before:

Quote

Most of the inevitable issues that arise with each update have been resolved within the first two months, which is amazing by historical standards and a tribute to the engineering processes Microsoft has put in place for Windows 10 development.

 

This one acknowledges that the problem may well be very small but the effects of the internet echo chambers means that we are facing an optic issue.

 

Quote

But even if it's just "one one hundredth of one percent" of customers hitting a particular issue, the optics around Windows 10 quality and reliability are not good.

 

 

So where are the numbers?  just reading a bunch of media articles is not actually presenting any evidence for your claims, in fact the above two quotes highlight they know full well that it's not a massive failure, but they have to write an article on something. 

 

 

[leadeater]

35 minutes ago, Delicieuxz said:

If you'd like to change the subject from the fact that Microsoft has largely out-sourced the Windows OS beta testing to Windows owners, resulting in a decrease of patch quality, to the subject of exactly what percentage Microsoft's OS updates have degraded by and what the relative issue-rate with them now is, you can, but if that's your interest then you should also do the research to find out. But to try to put that onus on me if you attacking a straw-man because your argument has already been proven invalid.

Seems last time this came up and corrections to the narrative around testing weren't taken on-board. There is the popular online narrative supported by opinions of those that do not believe in the strategic structure change, some of those no longer working for the company, but those opinions don't actually mean what they are saying is correct or the cause of anything.

 

The simplistic view of "Microsoft fired the testers" really is not a proper picture of what happened and the number of affected employees in that area that did actually lose their job is much smaller than these nice grands statements being made. Of the 18,000 at the time most were Nokia employees, that left 5,500 Microsoft employees. Of that 5,500 a good chunk of that was the Windows Mobile team as Windows Phone OS no longer exists and the other big one was the Xbox Entertainment Studios team. The rest were across the wider company under the restructure changes being made, teams were being renamed, employees being moved in to other or new teams etc.

 

For a company at the time with 130,000 employees there was no public information about how many people were actually employed in the old Programmatic Testers team, neither do we actually know how many of that team actually did lose their job rather than being reassigned.

 

When a company undergoes a change of leadership and instigates a strategic change the company structure will also change and it's during these times it's really easy to make up nice soundbite sayings which may not be accurate at all. Just because some teams were merged and responsibilities redistributed doesn't actually mean functions like testing were reduced.

 

Quote

The organizational changes Myerson mentions as being completed in the U.S. yesterday are focused largely around decreasing the number of testers as compared to the number of developers, resulting in layoffs of a number of those testers. This "combined engineering" plan, meant to undo the functional management structure that has been in place at Microsoft for the past few years, is new to OSG, but not new to other parts of Microsoft, including the Applications and Systems Group and the Cloud and Enterprise Group.

 

Quote

The changes around test were not completely sudden, I hear. A few months ago, Microsoft changed the job of the OSG test organization, renaming it "Quality," and refocusing the team from writing tests to "measuring quality." That meant the job of writing tests was already moved to the dev team, leaving many of those in the Quality team with less to do -- and clearing the way for the tester cuts.

https://www.zdnet.com/article/microsoft-layoffs-operating-systems-group-chief-myersons-memo-to-the-troops/

 

Microsoft is shifting to a more modern development model using more modern processes and team structures while also developing better tools and workflows to accommodate it, things other companies are already doing. Microsoft has not left critical testing to the public or insiders.

 

Where problems do come from and have always come from is new features and function changes. In the past those were saved for new operating system editions or large service packs which would bring about a huge influx in bugs and patches to fix those and that would be ongoing for many months. Service pack deployments for operating systems were delayed for months, many months while companies tested it and kept an eye on rate of issues and fixes. Service packs were also not flagged as Critical or Mandatory in Windows Updates on release and were optional for a good while.

 

Now Microsoft has changed to releasing new features and functional changes when those have been completed so we have rolling changes and with those comes rolling fixes if there are issues. These are also not released as optional and Microsoft doesn't really do optional updates for current editions of Windows anymore, part of their philosophy change.

 

So is it really as simple as pointing to 'all the testers got fire' as the root cause for all the problems, not really as it's a lot more complex than that and also not really an accurate statement anyway.

 

I don't think there are more issues with updates, not by number, but with the rolling release these get spread out so it feels like there is more rather than getting hit with them and a 3-6 month clean up with a good 1.5 to 3 year break until the next flood gates of bugs and fixes i.e Service Pack release or Operating System Edition release.

[Delicieuxz]

42 minutes ago, leadeater said:

Seems last time this came up and corrections to the narrative around testing weren't taken on-board. There is the popular online narrative supported by opinions of those that do not believe in the strategic structure change, some of those no longer working for the company, but those opinions don't actually mean what they are saying is correct or the cause of anything.

 

The simplistic view of "Microsoft fired the testers" really is not a proper picture of what happened and the number of affected employees in that area that did actually lose their job is much smaller than these nice grands statements being made. Of the 18,000 at the time most were Nokia employees, that left 5,500 Microsoft employees. Of that 5,500 a good chunk of that was the Windows Mobile team as Windows Phone OS no longer exists and the other big one was the Xbox Entertainment Studios team. The rest were across the wide company under the restructure changes being made, teams were being renamed, employees being moved in to other or new teams etc.

What I've said isn't founded in a claim that a specific number of testers were fired, but that a large number of them were (or reassigned), and that a lot of testing was offloaded onto Windows owners while at the same time the volume of software changes needing testing increased dramatically. And those things are indisputable.

 

And as I wrote above about the shifting of testing responsibilities onto developers:

 

"Offloading bug-testing onto developers only increases the burden of developers and means they have to divide their time between developing and testing, which means that there is a cut to both efforts.

 

There is a large net loss in testing happening in that situation. Cutting testers per developer from a ratio of 2:1 to 1:1 means that the developers would have to work 100% more hours with 50% of their time dedicated to testing in order to make-up the difference.

 

And it also has to be factored in that with Windows 10, Microsoft are making changes to the Windows OS at a rate that never before happened. So, there's a much larger volume of changes to be tested. This means that Microsoft is now testing more software changes and additions than ever before, but with fewer testing hours than they previously were employing."

 

Quote

For a company at the time with 130,000 there was no public information about how many people were actually employed in the old Programmatic Testers team, neither do we actually know how many of that team actually did lose their job rather than being reassigned.

If they were reassigned, they would still count as losses to the testing team. So, it is inconsequential to the fact that Microsoft has lost a lot of dedicated testers.

 

Quote

I don't think there are more issues with updates, not by number, but with the rolling release these get spread out so it feels like there is more rather than getting hit with them and a 3-6 month clean up with a good 1.5 to 3 year break until the next flood gates of bugs and fixes i.e Service Pack release or Operating System Edition release.

My personal experience with Windows has encountered an overflow of major issues with Windows updates since Windows 10, of which I had no parallel to with other Windows versions (for example, I never had a Windows Update update brick my OS before Windows 10). That personal experience is echoed in the widespread reports and professional analysis of Microsoft's current update regime. It's hard to deny personal experience that is corroborated by overwhelming user testimonies and professional pundit affirmation.

 

But if you think that Windows QA hasn't gone downhill, then just point to the pre July-2014 Windows 7 updates that were as problematic and numerous as the many Windows 10's disaster feature updates.

[Bitter]

2 hours ago, Semper said:

Substantial enough that more than once, Microsoft has delayed, and ultimately pulled updates for investigation because of critical flaws that hafe affected more than just the "outliers" and "one-offs"
Numbers enough that even Microsoft likely does not have an accurate numerical value to, but has been forced to take notice of.

I'd love to get my hands on internal stats from MS themselves of how soon they apply their own updates company wide or how much of a delay they're on.

[leadeater]

32 minutes ago, Delicieuxz said:

that a lot of testing was offloaded onto Windows owners while at the same time the volume of software changes needing testing increased dramatically. And those things are indisputable.

No that part easily is.

 

32 minutes ago, Delicieuxz said:

There is a large net loss in testing happening in that situation. Cutting testers per developer from a ratio of 2:1 to 1:1 means that the developers would have to work 100% more hours with 50% of their time dedicated to testing in order to make-up the difference.

This doesn't account for an increase in developers and those who's job titles changed. 

 

32 minutes ago, Delicieuxz said:

If they were reassigned, they would still count as losses to the testing team. So, it is inconsequential to the fact that Microsoft has lost a lot of dedicated testers.

No it isn't because if you are trying to claim less testing is done then less actually needs to be happening, not a change in job title and reporting line and a change in responsibilities but you're still doing testing.

 

Not having a team called "Programmatic Testers" anymore literally only means a team of that name no longer exists, what evidence do you actually have that testing has actually reduced rather than being done differently at the same capacity as before.

 

32 minutes ago, Delicieuxz said:

My personal experience with Windows has encountered an overflow of major issues with Windows updates since Windows 10, of which I had no parallel to with other Windows versions

Vista, Windows 8 (8.1 being almost an entire re-release). Two recent parallels there, Windows 7 SP1 wasn't issues free either.

 

32 minutes ago, Delicieuxz said:

That personal experience is echoed in the widespread reports and professional analysis of Microsoft's current update regime. It's hard to deny personal experience that is corroborated by overwhelming user testimonies and professional pundit affirmation.

Like I said that is more likely a perception issue because of the way feature and function changes are released, not that there are actually more issues you just no longer get breaks between changes of high likelihood of issues.

[mr moose]

17 minutes ago, Delicieuxz said:

What I've said isn't founded in a claim that a specific number of testers were fired, but that a large number of them were (or reassigned), and that a lot of testing was offloaded onto Windows owners while at the same time the volume of software changes needing testing increased dramatically. And those things are indisputable.

You are still insinuating the same cause for an unqualified condition.  Have you got a link to evidence that the volume of software "increased dramatically" while the total man hours went down. Or are you just assuming it?  It seems to me there is a lot of information missing in these claims and the ccusations are all based on media articles that only hint at things and do not provide any substantial reasoning.

 

17 minutes ago, Delicieuxz said:

And as I wrote above about the shifting of testing responsibilities onto developers:

 

"Offloading bug-testing onto developers only increases the burden of developers and means they have to divide their time between developing and testing, which means that there is a cut to both efforts.

 

There is a large net loss in testing happening in that situation. Cutting testers per developer from a ratio of 2:1 to 1:1 means that the developers would have to work 100% more hours with 50% of their time dedicated to testing in order to make-up the difference.

 

And it also has to be factored in that with Windows 10, Microsoft are making changes to the Windows OS at a rate that never before happened. So, there's a much larger volume of changes to be tested. This means that Microsoft is now testing more software changes and additions than ever before, but with fewer testing hours than they previously were employing."

 

If they were reassigned, they would still count as losses to the testing team. So, it is inconsequential to the fact that Microsoft has lost a lot of dedicated testers.

 

Again, on what grounds is any of this true?  based on what numbers?  has the workload changed, has the number of developers changed, What was the workflow before and after and has the work per man hour changed?  without that qualifying data all your claims and quotes and meaningless.  they are just assumptions and accusations based on hearsay.

 

17 minutes ago, Delicieuxz said:

My personal experience with Windows has encountered an overflow of major issues with Windows updates since Windows 10, of which I had no parallel to with other Windows versions (for example, I never had a Windows Update update brick my OS before Windows 10). That personal experience is echoed in the widespread reports and professional analysis of Microsoft's current update regime. It's hard to deny personal experience that is corroborated by overwhelming user testimonies and professional pundit affirmation.

 

But if you think that Windows QA hasn't gone downhill, then just point to the pre July-2014 Windows 7 updates that were as problematic and numerous as the many Windows 10's disaster feature updates.

And my personal experience with windows 10 has been almost perfect.   So what?  anecdotal is anecdotal,  you don;t see me or any of the other who don't have problems running around trying to convince everyone that there are no problems.   Just because you experience problems doesn't mean there is a major issue with more people experiencing problems. 

 

Also,. just in case the irony has escaped people, but there might actually be something in the fact you are promoting not updating and boasting about never updating yet at the same time you are complaining about having and "overflow" of issues with windows update.  I mean seriously which is it? or is the issue your having with updates because you keep messing with it and trying to avoid it. 

 

[Delicieuxz]

51 minutes ago, leadeater said:

No that part easily is.

With what? The Insider program's existence, and the statements by Microsoft that people who manually tell Windows to update are receiving beta patches, and that Windows 10 Home and Pro are used as test-beds to prepare patches for wider deployment all state that Microsoft is using owners of Windows 10 as beta testers.

 

Your arguing that it doesn't represent an offloading of testing means you take the position that Windows owners are only being used to increase Microsoft's amount of testing rather than replace some portion of it. And if you're going to assert that, then you should supply some evidence for it.

 

Quote

This doesn't account for an increase in developers and those who's job titles changed.

No it isn't because if you are trying to claim less testing is done then less actually needs to be happening, not a change in job title and reporting line and a change in responsibilities but you're still doing testing.

 

Not having a team called "Programmatic Testers" anymore literally only means a team of that name no longer exists, what evidence do you actually have that testing has actually reduced rather than being done differently at the same capacity as before.

Claiming that some job titles changed doesn't account for whether they changed to jobs that are no longer dedicated testers.

 

Quote

Vista, Windows 8 (8.1 being almost an entire re-release). Two recent parallels there, Windows 7 SP1 wasn't issues free either.

Neither Vista or Windows 8 suffered from patch issues on the scale of Windows 10. And Windows 7 SP1, which is just one update from Windows 7, wasn't a disaster like Windows 10's worst updates. You can compare the list of issues reported with them.

 

But if you feel otherwise, then just start pointing out the harmful update history for any of Windows Vista, 8, or 10 that rivals Windows 10's.

 

Quote

Like I said that is more likely a perception issue because of the way feature and function changes are released, not that there are actually more issues you just no longer get breaks between changes of high likelihood of issues.

If, with previous versions of Windows, there were the same number of issues only spread-out across longer break periods between updates, then that would mean the fewer update periods would be all the more notorious as they would contain massive numbers of huge bugs. But they aren't known for that, and they aren't known for being as bad as Windows 10's worst ones.

 

And I'm sure that the records of prior Windows version patch release issues haven't disappeared. We can compare them side-by-side to Windows 10's.

 

 

 

47 minutes ago, mr moose said:

You are still insinuating the same cause for an unqualified condition.  Have you got a link to evidence that the volume of software "increased dramatically"

Since the volume of feature additions and OS changes with Windows 10 patches are apparent for all to see, that's like asking for evidence that people breathe air.

 

You can easily find and peruse Windows patch notes.

 

Quote

while the total man hours went down. Or are you just assuming it?  It seems to me there is a lot of information missing in these claims and the ccusations are all based on media articles that only hint at things and do not provide any substantial reasoning.

The statement is qualified: "Cutting testers per developer from a ratio of 2:1 to 1:1 means that the developers would have to work 100% more hours with 50% of their time dedicated to testing in order to make-up the difference."

 

The one article's report of cuts to testers mentions the 2:1 ratio being reduced to closer to a 1:1 ratio. And the rest of what I said regarding the increase in workload based on that ratio of change and the report of shifting testing workload to developers is straight-forward mathematics.

 

Quote

And my personal experience with windows 10 has been almost perfect.   So what?  anecdotal is anecdotal,

But I didn't defer merely to personal experience. The difference is that your personal experience is offset by widespread user and pundit reports and confirmations of issues, while my personal experience is corroborated by the same widespread user and pundits reports and confirmations of issues. So, if your personal experience is being reported accurately, then two things can be claimed as true: You have had a near-perfect experience with Windows 10 updates, and Windows 10 has had an extreme number of series issues with its updates that have affects many people.

 

Your experience and my comments are not necessarily contradictory.

 

47 minutes ago, mr moose said:

Also,. just in case the irony has escaped people, but there might actually be something in the fact you are promoting not updating and boasting about never updating yet at the same time you are complaining about having and "overflow" of issues with windows update.  I mean seriously which is it? or is the issue your having with updates because you keep messing with it and trying to avoid it. 

 

Huh? Actually, I've had zero issues due to Windows Update since disabling Windows Update. So, disabling Windows Update was the cure-all for all of those issues. There goes your theory.

[NumLock21]

Just went to check my Windows Update history and for me windows update KB4549951 was installed on April 15, 2020. No problems here.

How many users actually have the problem, because it seems it's just a selected few, which means there really isn't a issue with the update.

[mr moose]

1 minute ago, Delicieuxz said:

 

Since the volume of feature additions and OS changes with Windows 10 patches are apparent for all to see, that's like asking for evidence that people breathe air.

You are conflating a companies total output with man hours per workload. You are also trying to insinuate that more products or output from a company is intrinsically linked to the number of issues with that product.  Boith are wrong and cannot be determined from each other.

 

1 minute ago, Delicieuxz said:

You can easily find and peruse Windows patch notes.

But you aren't doing that,  so get back to me when you have solid figures on the number of update issues now versus at any other time in history or for any other operating system. 

1 minute ago, Delicieuxz said:

The statement is qualified: "Cutting testers per developer from a ratio of 2:1 to 1:1 means that the developers would have to work 100% more hours with 50% of their time dedicated to testing in order to make-up the difference."

No it's not, because you don't know how much work they were doing before and if the structure change reduced the workload or not.  There are way to many variable in this to be so simple and absolute.  

 

1 minute ago, Delicieuxz said:

But I didn't defer merely to personal experience. The difference is that your personal experience is offset by widespread user and pundit reports and confirmations of issues, while my personal experience is corroborated by the same widespread user and pundits reports and confirmations of issues. So, if your personal experience is being reported accurately, then two things can be claimed as true: You have had a near-perfect experience with Windows 10 updates, and Windows 10 has had an extreme number of series issues with its updates that have affects many people.

 

Your experience and my comments are not necessarily contradictory.

 

 

The issue here is you are blinkered, because you choose only to accept the information that supports your feelings on the matter you think that your experience is a widely held observation.  Again, show me some numbers not just your opinion.   

 

 

1 minute ago, Delicieuxz said:

Actually, I've had zero issues due to Windows Update since disabling Windows Update. Disabling Windows Update was the cure-all for all of those issues. So, there goes your theory.

So why did you say you have an overflow of issues if you don't have any?   Are you just here for the argument?

[Syaoran]

I don't think this should be news anymore, every update is broken.

[Delicieuxz]

20 minutes ago, mr moose said:

You are conflating a companies total output with man hours per workload. You are also trying to insinuate that more products or output from a company is intrinsically linked to the number of issues with that product.  Boith are wrong and cannot be determined from each other.

No. You are again trying to shift your argument because you're caught in a corner by your own words.

 

I wrote: "the volume of software changes needing testing increased dramatically"

 

You wrote: "Have you got a link to evidence that the volume of software "increased dramatically" "

 

The record of update changes for Windows Vista, 7, 8, 10 are available. You can peruse them if you are willing to. There is no parallel in prior Windows versions to the number of software additions and changes that have been enacted in Windows 10 following its 2015 release. It's ridiculous of you to argue otherwise because that is also inarguable.

 

Quote

But you aren't doing that,  so get back to me when you have solid figures on the number of update issues now versus at any other time in history or for any other operating system. 

Documenting the exact number and ratio of issues has been your concern, and so it's your responsibility to look after.

 

Quote

No it's not, because you don't know how much work they were doing before and if the structure change reduced the workload or not.  There are way to many variable in this to be so simple and absolute.  

My statement is qualified because it is founded on known variables. I don't need to know whether workload changed to calculate how offloading tester workloads onto developers would impacts productivity when the before and after ratios of developers and testers are given.

 

But if you are going to argue that there is less work to be done with Windows 10 than with prior Windows OS, then that's a brazen claim to make and you should back that up with some evidence.

 

Quote

The issue here is you are blinkered, because you choose only to accept the information that supports your feelings on the matter you think that your experience is a widely held observation.  Again, show me some numbers not just your opinion.   

 

Calling the record of user and media reported and Microsoft-acknowledged issues bad opinions because they disprove your assertions isn't a good policy.

 

Quote

So why did you say you have an overflow of issues if you don't have any?   Are you just here for the argument?

Wow. Talk about projecting.

 

But why do you not actually read what you respond to, but instead assume, or worse, twist and mischaracterize what you respond to with something you invented yourself?

 

If you can't be bothered to actually check what I wrote, here it is again: "My personal experience with Windows has encountered an overflow of major issues with Windows updates since Windows 10"

 

And some of the major issues I've had with Windows 10 are documented in this forum, while many others are not.

 

And here is what I later wrote: "Huh? Actually, I've had zero issues due to Windows Update since disabling Windows Update. So, disabling Windows Update was the cure-all for all of those issues. There goes your theory."

 

If you find contradiction in those statements, then your reading comprehension is not that great. But if your reading comprehension isn't the problem, then that would mean that you're just here for the argument.

[mr moose]

1 minute ago, Delicieuxz said:

No. You are again trying to shift your argument because you're caught in a corner by your own words.

 

I wrote: "the volume of software changes needing testing increased dramatically"

 

You wrote: "Have you got a link to evidence that the volume of software "increased dramatically" "

 

The record of update changes for Windows Vista, 7, 8, 10 are available. You can peruse them if you are willing to. There is no parallel in prior Windows versions to the number of software additions and changes that have been enacted in Windows 10. It's ridiculous of you to argue otherwise because it is also inarguable.

 

Documenting the exact number and ratio of issues has been your concern, and so it's your responsibility to look after.

 

My statement is qualified because it is founded on known variables. I don't need to know whether workload changed to calculate how offloading tester workloads onto developers would impacts productivity when the before and after ratios of developers and testers are given.

 

But if you are going to argue that there is less work to be done with Windows 10 than with prior Windows OS, then that's a brazen claim to make and you should back that up with some evidence.

 

Calling the record of user and media reports of issues bad opinions because they disprove your assertions isn't good policy.

 

Wow. Talk about projecting.

 

But why do you not actually read what you respond to, but assume, or worse, twist and mischaracterize what you respond to with something you invented yourself?

 

If you can't be bothered to actually check what I wrote, here it is again: "My personal experience with Windows has encountered an overflow of major issues with Windows updates since Windows 10"

 

And some of the major issues I've had with Windows 10 are documented in this forum, while many others are not.

 

And here is what I later wrote: "Huh? Actually, I've had zero issues due to Windows Update since disabling Windows Update. So, disabling Windows Update was the cure-all for all of those issues. There goes your theory."

 

If you find contradiction in those statements, then your reading comprehension is not that great. But if your reading comprehension isn't the problem, then that would mean that you're just here for the argument.

My god, this whole post is just turning comments around, how about addressing the points instead?

 

facts you need to provide:

1. actually failure rates, that is a percentage of all updates that result in an end user issue.

2. Work load statistics from MS about each specific devision and job, without them you can't make claims about how much work each person is doing

 

Now I know the above are hard if not impossible for the average person to find, I am not asking for them because you can't provide them, the reason I am asking is to highlight that the claims you are making are simply derived from your own interpretation of media reports.  Without the above information you cannot be so absolute, in fact if we are rational about this the best you can do is argue there is room for improvement, although we don;t know how that would be done if at all possible given the current conditions.

 

 

[leadeater]

23 minutes ago, Delicieuxz said:

And if you're going to assert that, then you should supply some evidence for it.

Well I did, the link I provided and the explanation. Problem for you is, and this equally affects your own arguments, Microsoft is not and will not release detailed employee information so you can only evidence your claims as much as I can. The fact remains Microsoft underwent a company wide restructure and the simplistic soundbite "The test team was fired" is not accurate.

 

26 minutes ago, Delicieuxz said:

Claiming that some job titles changed doesn't account for whether they changed to jobs that are no longer dedicated testers.

You don't need dedicated testers as long as the same testing capability is retained. The way software development is done has changed, no modern company is doing it the way Microsoft was in the early 2010s.

 

27 minutes ago, Delicieuxz said:

Neither Vista or Windows 8 suffered from patch issues on the scale of Windows 10.

You're joking right? They both suffered huge amounts of issues, one of them still is a contender for the worst OS release reputation. Memories fade easily, Vista was total garbage and had lots and lots of issues and took forever to resolve them, I'd argue they never were, Windows 7 was the fix for that OS.

 

30 minutes ago, Delicieuxz said:

If, with previous versions of Windows, there were the same number of issues only spread-out across longer break periods between updates, then that would mean the fewer update periods would be all the more notorious as they would contain massive numbers of huge bugs. But they aren't known for that

They were known for that, that's exactly why service pack releases and migrations to new operating system editions were delayed for so long. I pointed to this already.

 

Btw I'm not saying the experience has been good or better than the old way, just that rolling issues will always seem more painful.

[mr moose]

19 minutes ago, leadeater said:

 

Btw I'm not saying the experience has been good or better than the old way, just that rolling issues will always seem more painful.

I think one of the articles linked earlier said exactly that too, the security side of updates is exceptional given the circumstances, however the rolling upgrades to the OS and providing it as a service has left a bitter taste in the mouths of those who need to manage many systems.

 

I think no matter what MS do there is always going to be people who think they have done a lousy job and things are worse.   And naturally there will always be examples of that.  I just feel that the overall experience for the domestic consumer is so much better than it was.  Security seems to be much better, updates are having an impact on that I am sure, and new upgrade features twice a year is value add for most of us.  It's like receiving a little box of treats for being a good boy.

[Delicieuxz]

38 minutes ago, mr moose said:

My god, this whole post is just turning comments around, how about addressing the points instead?

Points related to things I've claimed have been addressed by me. It's not my duty or my interest to engage with mischaracterizations of and irrelevancies to what I've said.

 

Quote

facts you need to provide:

1. actually failure rates, that is a percentage of all updates that result in an end user issue.

2. Work load statistics from MS about each specific devision and job, without them you can't make claims about how much work each person is doing

No, I don't, because nothing I've said depends on knowing those specific figures.

 

Quote

Now I know the above are hard if not impossible for the average person to find, I am not asking for them because you can't provide them, the reason I am asking is to highlight that the claims you are making are simply derived from your own interpretation of media reports.  Without the above information you cannot be so absolute, in fact if we are rational about this the best you can do is argue there is room for improvement, although we don;t know how that would be done if at all possible given the current conditions.

Whether genuine or not, the concern of what the exact issue rates are, as opposed to the detail that there are more issues with Windows 10, is one that only you have shown an interest in. Therefore, it is for you to look into if you want to have that exact rate comparison.

 

What we can be stated with absolute understanding due to an amalgamation of having records of patch changes, and understanding of the scope and complexity of features and changes, as well as knowledge of patch sizes, is that there have been far more software additions and changes in Windows 10 updates than there were in updates for previous Windows versions. We also can acknowledge that previous Windows updates didn't have bi-annual major feature releases, and that the service packs for previous Windows OSes do not reach the scope of what has been released for Windows 10 in its major feature updates.

 

Here is an overview of Windows 10's feature additions and changes: https://en.wikipedia.org/wiki/Windows_10_version_history

 

But there is way to compare issue rates between OSes - and it's one I've also mentioned to leadeater: Compare the documented issues with prior Windows version updates to those with Windows 10 updates, and see where the worst and most numerous issues are. If you are wanting exact figures, you can get them by collecting the reported issues from the most notorious updates from previous Windows OSes and comparing them to the same for Windows 10. If you want a usable general overview, the biggest issues with the most notorious patches will provide that.

 

 

 

31 minutes ago, leadeater said:

You're joking right? They both suffered huge amounts of issues, one of them still is a contender for the worst OS release reputation. Memories fade easily, Vista was total garbage and had lots and lots of issues and took forever to resolve them, I'd argue they never were, Windows 7 was the fix for that OS.

I am honestly not joking at all. I'm well aware of the reputations of those OSes - and also that the reason Vista had one of the worst reputations was largely not due to update issues. Vista's reputation as a "worst" Windows is actually largely a perception issue caused by its relatively-high system requirements for the time it released in and its lack of included 3rd-party drivers, and maybe the UAC which had draconian behaviour at its default setting.

 

But as I said, the records of the issues those OSes had are still available and they can be compared side-by-side to the records of Windows 10's updates. If you think that any of those previous Windows OSes have as bad an update track-record, just point to the notorious updates that challenge Windows 10's notorious update history in number and severity.

 

Moose would have me bring-up firm statistics for things that aren't even relevant to my comments. But the issues prior Windows version updates have are not tangents to this topic, and they aren't hard to locate, either. So, this is entirely a doable proposition.

 

Quote

They were known for that, that's exactly why service pack releases and migrations to new operating system editions were delayed for so long. I pointed to this already.

Taking a long time before releasing a service pack meant that a lot more testing was done on them before they were released. And the fewer number of service packs also meant there was less to go wrong with throughout an OSes lifespan.

 

That Windows 10 has major updates frequently, and that the major feature additions and updates in Windows 10 updates have already greatly outscaled the scope anything contained in all the service packs for any previous Windows OS, means that there's been a lot more that needs testing in Windows 10 updates and is a part of why Windows 10 has had more issues.

[Arika]

1 hour ago, Delicieuxz said:

But I didn't defer merely to personal experience. The difference is that your personal experience is offset by widespread user and pundit reports and confirmations of issues, while my personal experience is corroborated by the same widespread user and pundits reports and confirmations of issues

Which will be offset by a much larger number of people who don't experience issues which doesn't get reported, because no one reports on things that go right