Today's Windows 11 update. What a dog's breakfast!
-
Jeremy Falcon wrote:
thought I was supposed to use it to actually do some work.
Work? I thought you were a manager. :laugh:
I went back to doing contract work. One of the things I love about the business side is being able to make an impact, rather than having a boss "above" you that's too afraid to be honest with the business. But, it also consists of a bunch of useless meetings that waste people's time with folks that have no earthly idea about tech and are usually in it for the wrong reasons. I intrinsically lack respect for people who only wish to tell others what to do rather than do actual work themselves. I mean, there is value in the business side of course, but most of it is fluff and garbage. Ok, I'll stop ranting now. :-D
Jeremy Falcon
-
I went back to doing contract work. One of the things I love about the business side is being able to make an impact, rather than having a boss "above" you that's too afraid to be honest with the business. But, it also consists of a bunch of useless meetings that waste people's time with folks that have no earthly idea about tech and are usually in it for the wrong reasons. I intrinsically lack respect for people who only wish to tell others what to do rather than do actual work themselves. I mean, there is value in the business side of course, but most of it is fluff and garbage. Ok, I'll stop ranting now. :-D
Jeremy Falcon
-
Things were a lot better when MS had their internal testing lab. now the updates are released after dogfooding on their own VM's, not actual hardware. Things have been really bad for patches ever since.
Asking questions is a skill CodeProject Forum Guidelines Google: C# How to debug code Seriously, go read these articles.
Dave KreskowiakDave Kreskowiak wrote:
now the updates are released after dogfooding on their own VM's, not actual hardware.
That's been my argument as well...maybe their updates are well-tested on VMs, but virtualized hardware (and their well-known drivers) is NOT representative what most people are actually using, and then encounter problems that MS hasn't seen doesn't bother trying to seek out. For a multi-trillion dollar company, they can do better.
-
So did I, on the other hand, I did manage to be respected by them.
M.D.V. ;) If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about? Help me to understand what I'm saying, and I'll explain it better to you Rating helpful answers is nice, but saying thanks can be even nicer.
-
Dave Kreskowiak wrote:
now the updates are released after dogfooding on their own VM's, not actual hardware.
That's been my argument as well...maybe their updates are well-tested on VMs, but virtualized hardware (and their well-known drivers) is NOT representative what most people are actually using, and then encounter problems that MS hasn't seen doesn't bother trying to seek out. For a multi-trillion dollar company, they can do better.
You misunderstand. They are tested on the developers VM's. At one point, the old testing lab moved from actual hardware to a bunch of VM's before the entire lab was just shutdown and everyone laid off when Satya took over.
Asking questions is a skill CodeProject Forum Guidelines Google: C# How to debug code Seriously, go read these articles.
Dave Kreskowiak -
You misunderstand. They are tested on the developers VM's. At one point, the old testing lab moved from actual hardware to a bunch of VM's before the entire lab was just shutdown and everyone laid off when Satya took over.
Asking questions is a skill CodeProject Forum Guidelines Google: C# How to debug code Seriously, go read these articles.
Dave Kreskowiak -
I call it a dog's breakfast, but I am sure no self respecting dog will touch it. X| First installing it froze at 69% for almost half an hour. Then (after a restart) the computer froze at "You are 0% there." After waiting for about another 30 minutes, I forced the power off and restarted the machine. Upon the restart, it immediately resumed counting up to 100% quite fast. After another restart, I thought it best to run a sfc scan and sure enough it reported that it found and repaired some corrupt files. Now the machine seems to be working OK. This update robbed me of more than an hour out of my day! :mad:
Get me coffee and no one gets hurt!
I just continue to decline the upgrade to Windows 11. Its not like it will provide me with anything I actually need...
Steve Naidamast Sr. Software Engineer Black Falcon Software, Inc. blackfalconsoftware@outlook.com
-
abmv wrote:
switch to linux lol....
Yeah, 'cuz Linux never gets borked updates, no matter what odd combination of packages you might have, all coming from a bunch of devs that have nothing to do with each other.
I've had way more issues with Windows than I've had with Linux. The first couple years of Windows 10 was fairly bad for me. I've got 10+ machines running Windows and there'd be 2 or three that couldn't do the latest feature updates. Tried all the 'fixes' I could find on the web to no avail. Had to do a fresh install on the affected machines. And it wasn't always the same machines that would get borked on the next feature update. This went on for about 2 years. With my Linux updates, I've had one issue in the last 3 years. My email client (Thunderbird) wouldn't run after an update. Did a little research on the web and figured it was related to AppArmor preventing it from executing. Told AppArmor to ignore Thunderbird and then Thunderbird worked normally.
-
I just continue to decline the upgrade to Windows 11. Its not like it will provide me with anything I actually need...
Steve Naidamast Sr. Software Engineer Black Falcon Software, Inc. blackfalconsoftware@outlook.com
Hi Steve, I suggest Win 11's improvements to WSL might be worth looking at: arstechnica.com/gadgets/2021/10/the-best-part-of-windows-11-is-a-revamped-windows-subsystem-for-linux/ It's all about your use cases, of course. Given I spec/built my own tower, Win11 works well for me (except for the now fixed Ryzen 5600-TPM stuttering). What I *do* like about it is the relentless Fluent UI integration, which Just Works. - BTW, I'm an ex-MS 25+ year Senior Escalation Engineer (Win9x/NT services/Open Spec Team). Win 11 has more abstraction in the shell, which is good. There is more *fun* in the registry for the shell, not all of which is exposed in Settings, which is not so good. Time for some git gists. What would be great features for Win 10/11 would be tear-off tabs and preset tab sets in File Explorer. Bill Wesse Bill Wesse
-
I call it a dog's breakfast, but I am sure no self respecting dog will touch it. X| First installing it froze at 69% for almost half an hour. Then (after a restart) the computer froze at "You are 0% there." After waiting for about another 30 minutes, I forced the power off and restarted the machine. Upon the restart, it immediately resumed counting up to 100% quite fast. After another restart, I thought it best to run a sfc scan and sure enough it reported that it found and repaired some corrupt files. Now the machine seems to be working OK. This update robbed me of more than an hour out of my day! :mad:
Get me coffee and no one gets hurt!
I just got a brand new XPS 8950 and powered up for the first time on May 11. After connecting to the internet, Windows started to check for updates -- and hung! (The little blue arc on the circle just stopped) Finally Dell told me to do a clean shutdown and restart. Lo! Windows11 came up with no further difficulty, except that it is really ugly after ten years of Windows 7.
Joan F Silverston jsilverston@cox.net nhswinc.com
-
Same difference, updates are only tested on VMs, not real hardware. Whether it's being done by devs or a dedicated QA department really doesn't make much of a difference at that point - a lot of bugs are simply gonna be missed.
dandy72 wrote:
a lot of bugs are simply gonna be missed.
Versus when? Windows 98 had 13 million lines of code. Windows 10 has 50 million. So certainly then one should expect at least 5 times as many. And how many variations of hardware are there now? How many variations of software are there now? As an example Windows 98 was released in 1998. In 1999 99% of computers using the Web were using Internet Explorer. Keeping in mind of course there were quite a few less features then in any browser. But quite a bit less to test to insure it ran on Windows 98. And it was considered part of the OS.
-
And to think, they want to us blindly install updates immediately now. Trust the corporations without question. Big tech knows best. We're just plebes. As a side-ish note, even VS Code is getting odd. I swear every other day I open it up there's a new update it's bugging me to install. Fortunately, it doesn't crash or anything, but here I thought I was supposed to use it to actually do some work.
Jeremy Falcon
"I swear every other day I open it up there's a new update it's bugging me to install." Looks like you don't have a Linux box.
-
And to think, they want to us blindly install updates immediately now. Trust the corporations without question. Big tech knows best. We're just plebes. As a side-ish note, even VS Code is getting odd. I swear every other day I open it up there's a new update it's bugging me to install. Fortunately, it doesn't crash or anything, but here I thought I was supposed to use it to actually do some work.
Jeremy Falcon
-
Things were a lot better when MS had their internal testing lab. now the updates are released after dogfooding on their own VM's, not actual hardware. Things have been really bad for patches ever since.
Asking questions is a skill CodeProject Forum Guidelines Google: C# How to debug code Seriously, go read these articles.
Dave Kreskowiak -
I've had way more issues with Windows than I've had with Linux. The first couple years of Windows 10 was fairly bad for me. I've got 10+ machines running Windows and there'd be 2 or three that couldn't do the latest feature updates. Tried all the 'fixes' I could find on the web to no avail. Had to do a fresh install on the affected machines. And it wasn't always the same machines that would get borked on the next feature update. This went on for about 2 years. With my Linux updates, I've had one issue in the last 3 years. My email client (Thunderbird) wouldn't run after an update. Did a little research on the web and figured it was related to AppArmor preventing it from executing. Told AppArmor to ignore Thunderbird and then Thunderbird worked normally.
Cool story bro. Thing is, everybody has a different one. I'm glad it's working out for you. For me, if I were to just call it as I see it, my observations could be summarized as "different OS, different problems". And that's the only point I was trying to get across.
-
dandy72 wrote:
a lot of bugs are simply gonna be missed.
Versus when? Windows 98 had 13 million lines of code. Windows 10 has 50 million. So certainly then one should expect at least 5 times as many. And how many variations of hardware are there now? How many variations of software are there now? As an example Windows 98 was released in 1998. In 1999 99% of computers using the Web were using Internet Explorer. Keeping in mind of course there were quite a few less features then in any browser. But quite a bit less to test to insure it ran on Windows 98. And it was considered part of the OS.
jschell wrote:
Versus when?
A lot of bugs are simply gonna be missed when testing only on VMs versus when instead testing on the infinite "variations of hardware" that exists, as you put it. They can't cover it all, but they could at least try some subset. Not just VMs. That's all I was saying; I'm not sure where you were going with your line count comparison. If any comparison's to be made, it should at least be between two versions of the same product line (9x vs NT kernels have rather little to do with each other).
-
"I swear every other day I open it up there's a new update it's bugging me to install." Looks like you don't have a Linux box.
Peter Adam wrote:
Looks like you don't have a Linux box.
Frequent updates are actually worse in Linux because every day you log in, everything is updating daily. Even with distros such as Debian.
Jeremy Falcon
-
jschell wrote:
Versus when?
A lot of bugs are simply gonna be missed when testing only on VMs versus when instead testing on the infinite "variations of hardware" that exists, as you put it. They can't cover it all, but they could at least try some subset. Not just VMs. That's all I was saying; I'm not sure where you were going with your line count comparison. If any comparison's to be made, it should at least be between two versions of the same product line (9x vs NT kernels have rather little to do with each other).
dandy72 wrote:
I'm not sure where you were going with your line count comparison.
I started off my post with a statement that made it clear. "Versus when?" Your post seems to be comparing now to some other time. And I was asking that. Then I gave a comparison of different time periods that demonstrate the complexity of now versus then. You can provide a different time period if you wish.
-
dandy72 wrote:
I'm not sure where you were going with your line count comparison.
I started off my post with a statement that made it clear. "Versus when?" Your post seems to be comparing now to some other time. And I was asking that. Then I gave a comparison of different time periods that demonstrate the complexity of now versus then. You can provide a different time period if you wish.
-
I wasn't focusing on code complexity. I was focusing on the fact that MS only tests on VMs nowadays, and that's not sufficient as it's not representative of the real-world computers actual people use.
Except there is no way that Microsoft can test in any reasonable way using actual hardware. Any combination chosen is going to be a very, very small subset of what is out there. Not to mention that the introduction of the management software onto the machine which would allow them to automate the process would itself change the nature of the test environment and thus guarantee that no machine would match it.