If at first or second you don't succeed, you may be Microsoft: Hold off installing re-released Windows Oct Update The 1809 build of Windows 10 and Windows Server is fast becoming infamous. Microsoft pulled it shortly after release when it started deleting people's files, and stumbling in other ways. Redmond reissued the software on Tuesday, and today it's clear you shouldn't rush into deploying it, if installing it at all, in its present …
Happily for MS there will be another October in the'2019 time frame', so all is not yet lost.
I remember when a science-fiction club I used to be in had some particular piece of business that *had* to be taken care of by a particular month. That month's meeting ran out of time to handle business just then, so they made a motion that the meeting would be put on hold, to be resumed on the 45th of October (or whatever month it was).
The windows code base in particular seems to have so much debt accumulated that I suspect (very strongly) micro$oft will never be able to repay it in full.
In the end windows will have to go, windows will be left to rot alone ending its miserable life as a piece of abandonware among many others. micro$soft will get tired of all the painful meddling windows takes and will let it go. That day is approaching fast!
I'm old enough to remember when major new features in Windows were so disruptive that MS had to throw them all away and start again. Even second time around, the result was Vista, so ghod alone knows how bad it was when they killed it first time.
The difference is, that first round of development is now being done "live, on customer systems", because some moron decided that this was the modern way of working.
>unbelievably stupid
Agreed, but that does have some nuances. There's a blog post somewhere titled "You're not Google" and it's about why solutions that work for Google may not work for you (the emphasis was on scaling-before-all not being a fit for everyone).
For FB (not a fan myself), moving fast, breaking and fixing may in fact be a good approach.
IF:
- you can re-deliver working software very easily. This is very true for websites, less so for PCs and quite a lot less for embedded. Steam games and smartphone apps are pretty good there too.
- the consequences to your users of an outage are trivial. For FB or a game, this holds true. For an online banking website it's not.
- you are largely read-only (Netflix)
- your users do not mind having to wait a while until things are fixed. This is not true for a computer's OS.
If you take steps to preserve data and privacy (let's not mention FB here) and if you can segment new code with A/B testing, then, yes, I'd argue this is not a bad approach. Netflix is on the extreme end as they deliberately break things to make them more resilient.
But most companies are NOT operating within those parameters. Stupidity happens when someone drinks the coolaid, takes an approach that works under certain circumstances and applies it willy-nilly.
Frequent build devops CAN be great on a website. It can also be great if it pumps new builds into a rigorous QA process, well-isolated from customers. I know I am dumb enough that getting something right takes me zillions of runs and builds - I'd have been fired in punchcard days - but then it usually stays working from that point on.
It sucks when it delivers failing code to your end users, destroys their data and keeps them from using the system.
Advice to MS: you are not FB*. Spend some $$$$$, hire some software quality supremos (maybe someone heading Linux/BSD builds), give them power, and fix your stupid practices that are making you a laughingstock, not just to the IT geeks but more and more to everyone else as well.
* holds true for your effin telemetry too.
Back in the 1990s, the joke used to be that "At Microsoft, quality is job 3.1" (based on Ford's then-current slogan "At Ford, quality is job one").
Unfortunately, things have gotten worse, not better. BTC (before the cloud), customers would look at a next-gen product blowing up in deployment, and go "hmm, nope, we'll stay with what we have, thanks". Today, that's not an option, as MS is pushing people off of their existing systems, and into the new tech, whether they like it or not.
That could be beneficial, if MS was releasing solid, secure improvements to the customer stream. But they aren't. They're flailing around, in mad panic, adding features that they want, not what the customer has asked for. And worse, by concentrating on the shiny new features, they're ignoring the core features of the operating system.
I though IBM's OS/2 1.3 CSD 5050, known as "the CSD of doom" was the nadir of consumer operating system rollouts. Even WinME wasn't that bad. But MS has now surpassed it.
When customers are not only unwilling to implement vendor fixes, but discussion fora are filled with customers discussing how to not only bypass, but defeat vendor patches because they're afraid of them, there's a serious problem. Unfortunately, Microsoft doesn't seem to care all that much, based on how they're doubling down on their rollout strategy.
Win-10-nic forced updates! HA HA HA HA HA HA HA HA!
The October November December Who knows WHEN they'll get it right update. For Win-10-nic. Cramming it up your computer's peripheral slot following a multi-hour intarweb download near you!
Doesn't it *FEEL* good to have Win-10-nic on your computer?
(I'm glad I'm still using 7 for my windows things)
This gift just KEEPS ON GIVING!
Got Linux?
I believe its a reference to the Titanic, a ship which was claimed to be the most advanced ever developed, and completely unsinkable, but turned out to be a disaster. Intel's disastrous Itanium processor was dubbed the Itanic and Microsoft's disastrous OS is dubbed the Win-10-nic.
Ah, many thanks for the explanation.
I wasn't parsing it phonetically, when I saw "NIC" and "forced download" in the same sentence, I was thinking of the NIC as being the network interface controller doing the download. Unsurprisingly, the Win-10-NIC makes no sense whatsoever in that context.
I just bought a new gaming laptop with the intent of putting Linux on it.
So far not all the keyboard fn keys work (eg brightness). Not a biggie but annoying.
Optimus isn't working properly so until I figure out how to get Bumblebee (and maybe VirtualGL) working right. I have to log out every time I switch graphics chips.
HiDPI is mishandled by the majority of distros. I had to try several before I finally settled on Linux Mint Cinnamon. (And even then I need to play with env variables for certain apps)
Still trying to find decent equivalents of basic software like a PIM suite. The software landscape for business-quality apps on Linux is frustratingly barren. I did at least find a good quality PDF Viewer/Editor to replace Adobe: Qoppa PDF Studio. None of the usual suspects (Evince, etc) can handle a PDF form with embedded javascript.
So yeah, Linux has a long way to go before it can come close to being a viable alternative to Windows or Mac for the average person.
And to save time: No, there is no equivalent laptop with the desired specs, that is sold with Linux on it. If there had been, I would have bought it! The closest is the Oryx Pro from System76, but it's battery is nowhere near large enough.
So yeah, Linux has a long way to go before it can come close to being a viable alternative to Windows or Mac for the average person.
She said, having given a list of requirements which are not those of the average person. I've been using Ubuntu since 2006 for work and the only thing missing from my life is a Skype for Business client.
Linux is missing so many opportunities to eat into Windows because the distro communities are so utterly divided. Ubuntu tried but failed to become the distro of choice.
Mint is the best contender but until hoardes of developers from other distros jump ship and support Mint to become good enough to become a powerful enough contender then it will continue to struggle.
I won’t even get into the issue of different desktops with Gnome, K, Mate, Cinamon, etc
Then we need to address applications. Even Libreoffice is regrettably not good enough yet to replace the ubiquitous MS Office for most people. Until Mint becomes the main distro, the likes of Adobe, SAP, and thousands of others will not invest their dollars into producing and maintaining a Linux version. They need a winner to risk their efforts on.
Then there’s the basics of printing and scanning which are still woeful.
Graphic drivers, sleeping, hiberbating, touchscreens, WiFi are all serious issues or shortcomings I’ve faced with Mint and others.
We need the community to finally stop masturbating on their own fantasies and instead join together in a creative orgy of cooperation on one distro to finally push Windows aside forever.
"... instead join together in a creative orgy of cooperation on one distro to finally push Windows aside forever."
That's not going to happen. See the main reason many people including me chose Gnu/Linux is because you have a choice. Choice of distro, desktop, applications and the freedom to get your computer to work the way you want it to, not having to work the way a monolithic, faceless entity makes you. After all, who is the expert in how and why you use your computer? Not MS or Apple that's for sure. Red Hat and Ubuntu have generated lots of criticism and controversy by trying to be the "one distro to rule them all". With Gnu/Linux you have a choice and can go off and fork any project if you need to. See Linux Mint or LibreOffice as excellent examples of that. With a single distro we would soon be in a similar position to that of IE users a few years back. Stagnant, insecure and being told "Like it or lump it"
Too many people appreciate what FLOSS and Gnu/Linux in particular have done for them to throw it all away in a bid to topple a increasingly incompetent, error prone, arrogant company like MS. They are doing a good job on their own.
Linux is missing so many opportunities....
@Cavehomme_
Sound s like basically you think 'Linux should become a corporate product, just like MS.
All the issues you have mentioned (apart from the vast choice in desktops, distros etc) are down to Manufacturers of hardware not having provided drivers (or just shit BETA ones) for their hardware, be it GPUs, Network cards, Scanners.
They often refuse or are difficult or reticent to cooperate with projects like Nouveau, or SANE to help 'Linux software engineers write their own native drivers to run their hardware.
At this point they are actively participating in locking users into the Microsoft ecosystem by continuing with this short-sighted limited outlook.
Unfortunately the Linux community is in denial about the reasons for the continued Ubiquity of Windows. The things that Reg reading techies value are not the same as in the wider world and we are a small minority. They see Linux zealots as Jeff Albertson supercilious bores and they are not far wrong. A different approach is needed. They are not interested in your openn source and your free choice and superior architecture. They just want to get on and get their work done.
I am not a advocate of any os. I use all three main desktop options daily, interchangeably .
"Unfortunately the Linux community is in denial about the reasons for the continued Ubiquity of Windows."
I don't agree. I think that the Linux community tends to be fully aware of those reasons. I think where people get confused is that they think there is a great yearning in the Linux community that it topples Windows. Outside of a small (but vocal, since tech journalists are a major portion of this) group, I don't think that the Linux community is willing to compromise Linux to the degree necessary to achieve that goal.
"We need the community to finally stop masturbating on their own fantasies and instead join together in a creative orgy of cooperation on one distro to finally push Windows aside forever."
I sincerely hope this doesn't happen. The variety of Linux distros is a strength, not a weakness. A distro aimed at being a drop-in replacement for Windows is likely to include an awful lot of things that will only make Linux worse for those of us who currently choose it. Having some distros with that as the goal is fine, as long as there are other distros that I can use.
yeah well this article
https://www.theregister.co.uk/2018/10/23/microsoft_windows_10_crisis/
contains a brilliant statement: "Windows 10 is officially a shit show"
I summarized a few links about this update in a usenet post a couple of weeks or so ago:
https://www.theregister.co.uk/2018/10/08/microsoft_windows_10_pulled/
https://www.theregister.co.uk/2018/10/10/microsoft_windows_deletion_bug/
https://www.theregister.co.uk/2018/10/04/microsoft_windows_10_1809_audio/
https://www.theregister.co.uk/2018/10/23/windows_10_problem_roundup/
posted here as well, for everyone's enjoyment!
Hey El Reg, would you mind adding this list [and maybe a few more similar links] as a bootnote? Or if this comment stays near the top then everyone will see them here, whichever. It's a nice chronology, of sorts, as documented in El Reg.
> I am old enough to remember that Quality used to be something positive.
Actually, "Quality used to *mean* something positive"
Unfortunetaly, MS is rather more old school, back to the roots here and refers to "quality" as in its original, latin form "qualitas". Which translates to "(the) state (of sth)".
And no one cany deny Windows 10 does indeed have a lot of state.
I am old enough to remember that Quality used to be something positive.
Remember that "control" can mean keeping something from growing or expanding too much. Therefore "Quality Control" means insuring the quality of a product doesn't get too high. Product testers were conflicting with that agenda.
I am old enough to remember before the abbreviation of Quality Assurance had an ampersand in it.
And I'm old enough to remember doing it when there was an ampersand in it..
(Don't get me started with Agile and QA - no really, don't get me started becuase I want to go home sometime today..)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
