Posts by billdehaan

Microsoft teases targeted Copilot removal for admins

Does this count towards Copilot metric

Yes.

If you go to Office.com, it now calls itself the "Microsoft 365 Copilot app" rather than "Office". People quote this as proof that Microsoft is abandoning Office. Others laugh and say that those people just don't understand.

Microsoft isn't dumping Office, but they are shovelling Copilot into every single nook and cranny within Office. By branding it as the "Copilot app", they are basically doing what's known as Hollywood accounting. They may only have 20 million Copilot activations, against 300 million Office activations (not the actual numbers), but by rebranding this, they can now claim 300 million Copilot installations.

When you use Office with Copilot, you're counted in the AI user numbers. Whether companies disable Copilot or not, they are still counted, because they purchased Office with Copilot.

Years ago, at IBM, companies used to be able to get OS/2 dirt cheap, because IBM would sell things like site licences of 100-1000 users for something like $10 more than the 10-99 user licence. If your company had 80 users, it made sense to spend the extra 10 bucks, just in case you grew beyond. IBM then quoted that they'd sold 1,000 licences, because they had.

It's a standard industry trick to pump numbers up to make a product look popular.

The MS internal polling must be really bad

It's no secret that users dislike the direction Microsoft is taking Windows. The forced online account, the telemetry and invasion of privacy, and the advertisements are all objectionable, but it's the injection of AI everywhere, even/especially where it's not needed that has has caused the most pushback from users.

The criticism to all this has been deafening, but Microsoft has completely ignored it, dismissing user all concerns and forging on ahead. They have not only ignored user criticisms, they are doubling down and accelerating their push to make Windows an Agentic (or "AIgentic") OS.

Until now.

This is the first indicator that upper management in Microsoft is aware of how much user dissatisfaction there is. They can dismiss home users, because they are a captive market. Sure, some will switch to MacOS or Linux, but the majority of users are locked in, and will take what they're given. They'll complain, but they'll stay. But the corporate world is a different matter. Microsoft actually cares when they start looking at alternatives, and many are.

A lot of EU governments are running pilot projects to test the viability of using non-Microsoft solutions. So are many companies, both publicly and quietly. One of the arguments for that has been to get away from the forced AI that Microsoft has been pushing. By allowing corporations (but not end users) to disable it, Microsoft can now claim that argument is no longer relevant.

For them to have done that, the corporate feedback they've received must have been pretty scathing.

Whether this will trickle down to home users is another story. Personally, I'm doubtful it will.

What the Linux desktop really needs to challenge Windows

Re: What Linux needs to be pre-installed on PCs

I agree 100%.

Back in the 1990s, OS/2 was a better OS than Windows 3.x and 9x, in terms of technical merit. But it didn't matter, because it was starved for applications and hardware driver support. And back then, Linux was great for servers, but immature on desktops. I still remember dealing with /dev/audio trying to get speakers to work late into the night, never mind non-SCSI tape systems.

Today, it's a different world. Linux may not have the market share of Windows, but unlike 20 years ago, it's easy (in many cases almost trivial) to install, it uses less resources than Windows, and while driver support lags behind Windows, it's still excellent. And most importantly, the catalogue of applications available has no real coverage gaps, not to mention that unlike 20 years ago, many applications are web-based services now.

In many ways, I liked using OS/2 more than Windows, but I couldn't use it for everything because of gaps in application software. That's not the case with Linux. There are only three things I could do in Windows that I can't do in Linux, and two of them run in Wine (the third is a hardware GPS that doesn't work under Wine or in a VM).

So what, exactly, does Linux need to do to "challenge" Windows? And why should it? Linux advocates point out that the user base size has risen from 3% to 5%, but what difference does it make? If it hits 10%, or 14%, or even 40%, nothing will change for the end user.

I'm glad it exists, I like people to know about it, especially the people who have been abandoned by Windows 10 and can't afford a new machine simply to run Windows 11, but it's not a crusade to replace every Windows or Mac desktop out there.

What Linux needs to be pre-installed on PCs

People keep pointing out that installing Linux is easier than it's ever been, so users won't have any problem installing it.

They're right that it's easier than installing Windows 95/98/2000/XP or OS/2 was, back in the day. They're wrong that it means users won't have a problem with it.

The problem isn't how difficult Linux installations are, it's that it requires installation at all.

The modal Windows user today didn't install Windows, it came pre-installed on his machine. Oh, he configured it when it first booted up, certainly. He gave his name, his location, his (now mandatory) Outlook account, and the like, but he didn't partition and format the hard disk/SSD, he doesn't know how much memory he has, and he certainly didn't look at, let alone set, jumper settings on the motherboard like "the good old days".

Linux in 2025 is unquestionably easier to install than Windows XP was in 2001. But it's not 2001 any more. Comparing ease of installation from media isn't the point, we're comparing installing Linux from media against pre-installed machines.

Sure, there's System76 and Tuxedo, but if Linux is really going to compete on the desktop, it's going to need vendors who offer PCs with Linux preinstalled, and with support.

Microsoft is continually shooting itself in the foot and driving many users away, but even the thought of installing an operating system from scratch nowadays is intimidating to many people. Sure, many will give it a shot, and many of them are surprised at how easy it was. But many more have no more interest in installing a computer operating system than they have of doing their own oil change on their car.

Re: snapflatimages

One VM for each and every program.

Sadly, I know people who seriously do this. Unsurprisingly, all three of them are software architects at their companies.

One has a desktop with 192GB of ram, 32TB of NVME SSD disk space, four Ryzen processors, no less than six NICs. Everything is a Proxmox instance. Email is a 4GB Linux VM with Thunderbird. Browsing is a 32GB VM with Firefox, running an average of <cough> 300 </cough> tabs. And he considers that to be a perfectly normal desktop, not excessive at all.

What Linux needs to challenge Windows is advocates for it who realize that the average computer user isn't a computer scientist, and doesn't want to become one. They just want their computers to do what they need it to do.

When a friend's power supply blew and smoked her entire PC, these architecture astronauts were speccing out $3,000 computers for her, babbling on about Swarmkit, Kubernetes, orchestration, and how easy it is to inject a JSON stream into a Docker instance.

All of which meant nothing to her.

When someone asks you for advice on what car they should get, don't give them a 15 minute lecture on timing chains and fuel compression ratios.

I grabbed a used I5 with 8GB of ram and a 256GB SSD, for her, installed Mint on it, set up Ungoogled Chromium (she was used to using Edge), Thunderbird, KeyPassXC, and a couple of other apps, all for under Cdn$150 (about US$110).

It does what she wants, it does what she needs. And really, that's all that Linux needs to compete.

Smartphones face a memory cost crunch – and buyers aren't in the mood

We want you to pay us more for features you don't want

Not exactly the most compelling marketing slogan I've ever heard.

The marketplace is littered with corpses of things the industry pushed as the Next Big Thing. A decade ago, 3D televisions were all the rage, until the industry realized that people didn't care, and weren't interested in shelling out a ton of money to buy something they didn't need in the first place.

Unpopular features, like car touchscreens replacing manual controls, home computers that require internet accounts to work, normal TVs that don't require internet access sell eventually, when the previous models break down and need replacing. And for markets like cars, home computers, and TV sets, where the average consumer may only one every decade or so, the cellphone market is based on annual, or biannual upgrade cycles.

Sure, there are gearheads who get a new car every 18 months, PC gamers who buy a new rig every year, and video geeks who bought 8K TVs the first day they were available, but for the average home user, the four year old car, PC, or TV set does what they need and they don't see a need to shell out for a new one.

Not so with phones. The average consumer is conditioned to expect that they have to replace their phone every 2 years, or 3 at the outset. Even though all they get is camera improvements you'd need to be Yousef Karsh to appreciate, bezel that are 2 microns narrower, or features they've never needed, used, or wanted.

Cell phones have gone from $300 flagships to $500 flagships to breaking the $1,000 barrier. And once that was psychological barrier was broken, the sky was the limit. There are foldable phones that sell for $2,000 now, and that's before the AI injection, and the corresponding ram upgrades.

I'm using an 8 (almost 9) year old Samsung S8. It has an audio jack, and removable storage, two features which have been removed from current phones. Friends try to show me what their new $1,200 2025 phones can do that my old hand me down phone that I got for free can't, and they struggle to.

With luck, some enterprising vendors may offer slimmed down, AI-free (or at least AI-lowered) phones at lower price points, and bring some sanity to the market.

Not everyone requires recording studio quality audio/video tools, and Cray level artificial intelligence tools.

It's nice that there are $25 iced mocha pumpkin spice lattes using water from arctic glaciers, and $30 aloe vera shaving injected quadruple roasted Kenyan espresso infused with almond milk, but a lot of people would be just fine with a $2 cup of coffee, if they could find one.

Latest Windows 11 updates may break the OS's most basic bits

I really try to not be that guy

I switched to Linux at home full time two years ago, and while I'm happy to help others who are curious about switching to it, or want help, I try to not be one of those insufferable twits who talk about "Windoze Exploder", and how Linux systems are sooo much better, etc., etc.

Microsoft is making it really difficult, though.

There's the normal given and take of how Linux is better at this, and Windows is better at that, and depending on user needs, Windows or MacOS will be a better choice than Linux for many people.

Nonsense like forcing an Outlook account is bad enough. Issues with telemetry and tracking are worse. Uploading user data to the cloud without permission is terrible, and Recall was a privacy nightmare that's a class action lawsuit waiting to happen.

But at least, for all the garbage layered on top of it, the operating system itself kept functioning, at least.

At least it did. Recently, the file explorer became so bad that Microsoft is pre-loading it into memory to get around performance issues (performance issues? in a file manager?). Then last week, they screwed up the lock screen, making the sign-in icon invisible for many users, locking them out of their own computers. And now they've broken the start menu, the task bar, and basic explorer functions.

This is the lowest application level of the UI, the "don't mess with this unless you know what you're doing" level, and they're modifying it like crazy. End users who purchased a supposedly mature operating system are being used as guinea pigs to beta test whatever Microsoft feels like doing this week. I'd be very hesitant to put a production system on Windows 11 under they can prove they have this churn under control. Right now, it's just chaos.

Swiss government says give M365, and all SaaS, a miss as it lacks end-to-end encryption

No European government body should use anything Microsoft, it's as simple as that.

I'm not reflexively opposed to everything Microsoft automatically, but their telemetry, user tracking, and increasingly strongarm tactics forcing users to register an online account in order to use a supposedly offline operating system certainly disqualifies Windows from consideration for anything confidential or private.

One of the issues that was raised in Canada recently was the issue of warrants. If the police want to look at your PC as part of an investigation, whether of you personally or of a person or institution you may have interacted with, they require a warrant. However, they can also make an informal request. Since it's just a request, doesn't have to be recorded in the public record, whether it's agreed to or rejected.

The thing is, that request doesn't have to be made of the PC owner, It does if they want access to the physical computer, but if all they want are files on the computer, then CSIS/RCMP/OPP/local police can call Microsoft, and informally ask if they have the files in question. Microsoft can, and by policy should, reject requests that aren't accompanied by warrants, but there's nothing stopping them from turning over your unencrypted online backups to the authorities if they ask.

And even if they do say "come back with a warrant", obtaining a warrant for a third party, like Microsoft, is granted much easier than a warrant to search an individual premises.

If the police serve a warrant to Microsoft to get your OneDrive backups, neither they, nor Microsoft, have any obligation to tell you about it.

I've had non-technical friends be absolutely horrified to discover that confidential customer data from "their" PC was uploaded to OneDrive without their knowledge or consent, because they stored those files in "My Documents" folders, which Microsoft backed up to OneDrive without being asked. And once they're in the cloud, they're staying in the cloud.

A lot of my friends are lawyers or doctors, they are bound by client and patient confidentiality laws. Microsoft isn't, however. And if you don't want discussions of your client's medical histories to be online, you damned well don't want your governments nuclear launch codes to be one warrant away from showing up in a twitter post.

End to end encryption is not enough

“Most SaaS solutions do not yet offer true end-to-end encryption that would prevent the provider from accessing plaintext data”

I sincerely hope that doesn't mean that if/when Microsoft does enable full end to end encryption in M365, that the Swiss government would then start using it for confidential government data.

The fact that the Swiss government is talking about end-to-end encryption rather than zero trust is a bad sign.

All end to end encryption does is prevent the data from being decrypted in transit. The SAAS recipient, in this case Microsoft, still has access to the unencrypted data.

People aren't concerned about Windows Recall because they're worried about man in the middle attacks, they're worried that Microsoft will have access to their data. Even with end to end encryption, Microsoft could still access M365 data. There can be all sorts of legalities stopping them, and internal processes, but physically, Microsoft employees could access the M365 data.

Unless they're committing to zero trust systems, I wouldn't trust any SAAS vendor. And I'd only trust them with zero trust because, by definition, zero trust assumes they can't be trusted.

LisaGUI recreates Apple's innovative computer OS, without emulating it

Re: I actually used a Lisa, and I almost got one for free

That was the inference everyone made, and hence my comment that I hoped he was joking.

I actually used a Lisa, and I almost got one for free

We got one in our university lab, and everyone had the same reaction.

They said "cool", sat down and played with it for a bit. Then they got frustrated, said it was way too slow, and asked if they were doing something wrong. When people replied that "no, that's just how it is", they'd sigh, and go work on one of the two IBM PCs we had.

A few years later, I briefly worked at a Unix shop that a bunch of oddball machines. They had big iron (Siemens and Nixdorf) but they had dozens of one-offs of desktop machines few people today have heard of, like the Victor 9000 and Hyperion, to do testing. There were several PCs, a couple of Macs, some Acorns, BBC Micros, and one solitary Lisa. When the company lost its' major contract, and went belly up, people were basically told that the company was folding at the end of the week, they weren't getting paid, and there was no need to return any company assets that people had taken home to work with, they could just keep them.

The big iron was leased and would be seized, but what about the machines in the office that the company owned? The owner felt bad about not paying people, and knew that the receiver would be lucky to get 15 cents on the dollar, so he literally said "help yourself". And people did. I was the fourth person to eye the Lisa. Three others powered it up, made sure it worked, but after 15 minutes of playing with it thought the better of it. So did I. Apple had already discontinued it at that point, so there would be no new software, no improvements, and no repairs.

The guy who did eventually take it was a sailor. When we asked why he picked it, he just said "I have a boat". I'm at least hoping he was joking that was the reason.

Linux admin hated downtime so much he schlepped a live UPS during office move

Tempting, but dangerous

In 1992, I was migrating a server room to a different floor in the same building. The company was expanding, so they'd rented the full 10th floor, and were abandoning the half of the 3rd floor they had occupied for about six years.

The new server room was built, the power, A/C, HVAC, and halon systems were all approved, tested, and good to go, so now it was just a matter of moving 150+ machines. They were mostly Sun IPCs and IPX machines, but there were a few HP machines, and even an SGI box or two. All the networking and power connections were set up, so all that should be required for each machine was doing a proper shutdown, unplug all the wiring, throw it in the elevator, go up 7 floors to the new server room, find the properly labelled bench, plug everything in, turn it on, and everything would be hunky dory.

It was booked for a long weekend, all users were informed, so all user workstations were shut down, there were no open files or pending requests on the servers, etc. Everyone left on Friday night, and Tuesday morning, they'd go to their new office, where everything would be the same.

We had a fireman's line with one group disconnecting machines and loading them on a cart, a second team moving the cart to the new floor and unloading the machines, and the third team putting the machines in their new location and setting them up. In theory, it should only take a few hours to move all machines.

We all know the difference between theory and practice.

As it turned out, nearly 40% of the machines failed to turn on. We tried bringing them back down to the old floor, but no luck. So, we powered down some of the machines, and tried to power them back up again on the same floor, and the 40% number held. Almost half the machines wouldn't power back up after a shutdown.

Fortunately, all of the machines that were DOA were Suns, so we only had to deal with one vendor. Even more fortunate, it was a big customer with lots of money, so it had enough clout to get a Sun tech person on site within an hour (3am on a Saturday morning).

The culprits were the power supplies. Having never been power cycled for half a decade, many had faults occur that were never detected. Well, they would be detected on the next power cycle, but because the admins were competing to see whose subnet of machines could have the highest total uptime, most of which were over 2,000 days at this point, they went to extraordinary lengths to never reboot. Patches were hotswapped in, and other voodoo was performed to keep the uptime going.

Fortunately, it wasn't the hard disk, and the machines were off the rack configuration, so they could pop the disks out of a unit with a bad power supply into a new machine and get it up and running quickly. All that we had to do was get about 80 new Sun workstations at 4am, have them delivered, swap out hard drives, and that was it. Easy peasy.

The Sun rep saw a multi-million dollar account being lost on his watch, so he moved heaven and earth, and actually got us the machine in under four hours. I think he aged a decade in those four hours. We were amazed, and speculation was that he had pictures of Scott McNealy with a sheep, or something. But he managed to pull it off.

Some security configurations that used MAC addresses had to be modified, but other than that, it was surprisingly uneventful.

Our "should only take a few hours" migration that was supposed to be complete by Saturday noon, leaving Sunday and Monday for testing ended up taking until 9pm Sunday. We got everything working, but we were glad that we're migrated the small server room of 150 machines, not the major data center with over 2,000 machines.

Before doing that undertaking, a new rule was instituted that all servers would, on rotation, be cold booted every 60 days, specifically to avoid a repeat. And that data center with 2,000+ machines? It's admins didn't have uptime competitions, so failure rates were under 40%, but they were still around 5%.

Multi-year uptimes are great, but make sure you've got redundancy, and factor in the possibility that you may have to take everything off line at the same time at some point.

Developer battled to write his own documentation, but lost the boss fight

And where is the company wide style documentation approved by all?

It's been "under review" for over a decade. It requires approval from different departments which have contradictory intractable requirements. Their superiors attitude could be best summarized as "we don't have time for this, work it out amongst yourselves", followed by an annual "haven't you people resolved this yet?" review. It's then sent back to the departments for another year of pointless haggling and stalling.

As I may have subtlety implied, it wasn't the most well run company I've ever worked with.

We won't let you publish a user's guide. What do you mean it's a not a user's guide it's a manual?

I worked at a company which could have been called Jobsworth, Inc, and should have been.

I developed a fairly technical new network analysis tool, and I wrote up a 30 page user's guide to explain it. The documentation group refused to approve it because it didn't follow the "proper" format for a user guide. What was the proper format? There were three competing style guides, all of which were incorrect, of course. The official style guide was "talk to Joe". Joe, the wizened Tech Writer, would decide which arbitrary rules were to be enforced for any particular project.

Unfortunately, their style guides (ie. Joe) were all based on their historical projects, which were not networking tools. So my tool had sections that didn't map to their format, and their format required sections that were completely irrelevant to a networking tool. I was required to add dozens of sections and fill them with "not applicable to this tool", as a software tool didn't really need installation safety precautions or weight allowances, but they had to be there, or Joe wouldn't allow it to be published.

After that, it had to be approved by the Publications department, a second Kafka-esque entity with equally obscure undocumented rules. One version of my document was rejected because it used Courier 12pt font when it should have used Garuda 12pt. Two versions later, the same passage was rejected for using Garuda 12pt, and should have used Courier 12pt, instead.

I finally cornered one of the Publications team, who admitted that they were required by their management to always reject every submission at least three times. Usually that wasn't an issue, because usually there were three or more actual problems, but for ones like mine, where nothing was really wrong, they just had to invent an excuse.

Finally, I got past the lower level drones, and reached the Boss Level. This was the Project Approval, an administrative fascist who rejected my document and said it had to use a template that followed the company product structure. The fact that it made no sense to document a software tool as it if was a construction project made no sense, but he was adamant. User Guides were User Guides, sacred texts that must follow the holy template. My document was fine "for what it was" but it was not a User Guide. Deviation was forbidden

So, in the end, I decided my tool didn't actually need a User Guide after all, and simply didn't publish one. The same Kafa-esque rules that dictated the format also stipulated which projects required user guides, and mine didn't meet any of the criteria, so I was off the hook.

I took the original 30 page "User's Guide" document, renamed it "User's Manual", and attached it as a Word document, an RTF document and a PDF to the distribution media, so anyone who got the software also got the manual along with it, without having to order it from Publications.

Project Approval, Publications, and the Documentation group were all aghast, and one even tried to file a formal complaint. Unfortunately, I'd followed the letter of the law, and they could do nothing to stop me.

On the up side, I got thanked by end users who said they appreciated that our company "finally" documented something in a way that they could actually understand and use it, unlike the "Chaucer on crack" that they'd come to expect from us.

'Windows sucks,' former Microsoft engineer says, explains how to fix it

The problem isn't Windows, it's Microsoft

Back in the early 1990s, the fight for OS dominance was between OS/2 and Windows. Apple was busy making itself irrelevant, and Unix vendors weren't a viable option for the home market.

In a technical comparison, OS/2 beat Windows in almost every way. And yet OS/2 was steamrolled by Windows, and within 4 years, IBM threw in the towel.

Although there were lots of conspiracy theories, the simple fact was that it wasn't Windows that beat OS/2, it was IBM's treatment of OS/2, and the OS/2 user base that did it. IBM viewed consumers as defective corporations rather than customers. They rejected user criticisms and suggestions for improvements, and instead told customers the OS was right, and they were wrong.

I read both IBM's and Microsoft's respective releases for the operating systems. Microsoft talked about system features, improvements, and end-user benefits, where IBM talked about licencing, distribution, centralized auditing, and control. In one 14 page writeup, IBM never used the word "user" a single time, talking only about "seats", instead.

And that's what Microsoft is doing today. Invasive telemetry is a benefit to data brokers Microsoft can sell user data to, not the end user. Likewise, advertisements in the start menu are a boon to advertisers, not the user sitting in front of the computer. And shoehorning AI into everything, even AI, is not something that customers asked for.

Unlike 25 years ago, the consumer operating system market is very mature now. For the general user, there's not much that can be done on MacOS that can't be done on Windows, or Linux, or vice versa.

Back in 1992, Windows 3.1 was technologically inferior to OS/2, never mind Linux. Today, that's not really true. Windows is just as mature and capable as MacOS or Linux. The differentiator is no longer the technical merits of the system, but the priorities, such as privacy, and data collection. That's where Microsoft is losing, not because of bad OS foundations.

Win10 still clings to over 40% of devices weeks after Microsoft pulls support

It's not as bad as it sounds - it's worse

As many people will no doubt post, they've switched to Linux (or MacOS, or in a few cases, BSD). That's fine for individuals, but it doesn't help the larger problem.

I switched all of my machines except one almost two years ago, and the one remaining machine is not connected to the internet.

While we may not have to worry about our insecure machines being attacked, unpatched Windows 10 machines are botnet nodes waiting to be exploited, which is still going to be a pain. DDoS attacks are already a pain; this is going to make them a lot worse.

I'd like to think some enterprising people will offer third party security updates (like 0code) to Win10 holdouts, but even if it's free, I doubt most users will take advantage of it.

'Keep Android Open' movement fights back against Google sideloading restrictions

Re: And this is why I keep my old cell phone

Really? I didn't see any Moto G ports when I looked, or maybe there were none for the specific version of the phone I have. I'm in Canada; sometimes we get European versions of a phone, sometimes we get American versions.

There's been more than a few videos where the presenter of a how-to video will say at step 3 or step 4 that "if you have the Canadian version of the phone, you're out of luck because the bootloader is locked" or something similar.

I'll take another look.

And this is why I keep my old cell phone

I made the mistake of accepting Android updates on my 2014 Moto G, and they made the phone utterly unusable. The phone technically still works (until October 31st, when Canada discontinues the 3G network), and of course Google and Motorola both say it's obsolete and not supported. But if I hadn't upgraded it, I'd still be using it.

I am still using my 2017 Samsung S8. With the exception of security updates (which stopped years ago [1]), I don't update any of the Google or Samsung apps, because I don't use any of the Google or Samsung apps, with everything coming from F-Droid. If this is the path Google is taking, there's no way I'm getting a "certified" Android device that only allows Google-verified developers. I'll buy a used S8/S9/S10 and flash the ROM to LineageOS or the like.

If I wanted to be in a walled garden, I'd by an iPhone.

[1] Yes, I understand the security implications. My phone has a Gmail account which is only for the phone. I use the phone for phone calls, SMS, camera, offline music playback, GPS, and Dashcam. If someone breaks in, at worst they'd get my SMS messages, my phone contacts, and mp3 and FLAC playlists. There are no financial apps or anything like that. If something is even remotely confidential or private, I don't put it on the phone.

Programmers: you have to watch your weight, too

Managers have to weigh in

As a contractor, I worked at dozens of companies, and I saw the same trends, over and over.

Architects in all companies loved to over complicate things, often to a ridiculous extent. Unfortunately, a lot of it is because the architect read up on a cool new language feature and he was going to use it, whether it made sense or not. Other times, it's to show off how clever he or she is. If they can use arcane features that no on else is familiar with, it gives them bragging rights about how smart they are.

And sadly, for many of them, unintelligible code is job security. At more than one company, the architect was deeply disliked, but couldn't be let go because he or she was the only one that could maintain their company's overly obscure lifeblood application.

In some companies, management reigns them in. In others, they're allowed, or even encouraged, to run wild. I've seen arcane code in well-run companies, but it's been accompanied with design documentation explaining why the obscurity was necessary, and the code is heavily commented. In less well-run companies, the code is just an intelligible blob that no one understands, management tells developers to ask the architect if they don't understand, and the architect tells management that the problem isn't the code, it's the quality of the developers.

Often, when proper documentation standards are enforced, the architects suddenly discover that they can implement their functionality is a less complex (and less difficult to document) manner.

I still have a copy of the email chain where the system architect dismissed complaints about the complexity of his system, saying that the problem was that developers were instantiating his library functions incorrectly. His constructors were protected abstract virtual base pure const virtual private, they were not protected virtual abstract base pure virtual private const, for god's sake. When his new manager told him he had to document the difference, a week later there was a new library where his constructors were now all virtual base pure const. Eight layers of abstraction were reduced to four. The complexity was lowered exponentially simply by forcing the architect to document.

It's always a shock when a manager from a well run company moves to a poorly run one, and vice versa.

Managers from poorly run companies are honestly shocked to see meaningful documentation, and that their developers are actually able to use an architect's framework without constant handholding.

And when managers from well run companies see developers being told to use chelate bijective manifold private association references in polymorphic spinlocks and notice that there's not a word of documentation, they call the architects on the carpet for it. And usually, the architects are outraged, because they've never been asked to document things in the X years they've been there.

A software-defined radio can derail a US train by slamming the brakes on remotely

Remotely take over a brake controller

"You could remotely take control over a train's brake controller from a very long distance," he explained. "You could induce brake failure leading to derailments or you could shut down the entire national railway system."

Not in any passenger CBTC system I've ever worked on, and I've worked on many of them.

First off, CBTC systems have to meet CENELEC compliance, and that includes a cyber safety assessment.

No train I worked on ever had brakes connected directly to a radio. The radios were connected to OBC (on board computers/controllers) that decoded incoming telegrams from the wayside and station operators. Even if an attacker knew the telegram format (which tend to be very complicated) and knew the correct IP address of the sender and spoofed it and sent a valid command to EB (emergency brake) the train, the telegram would have to be WPA2 (or better) encrypted.

So this may very well be an issue for freight vehicles, which is bad, but if anyone's worried that the London tube is going to be taken over by kids with cell phones, I wouldn't worry too much about it.

We're number 1! Windows 11 finally overtakes Windows 10

Re: The real spike will come in September

I worked at a company once where the Windows 8 to Windows 10 migration planning was held up by the fact that the company was still in the process of migrating from Windows XP to Windows 7. It was particularly troublesome because they discovered that 80% of the PCs in the company didn't meet Windows 7 minimum requirements.

But no worries, IT assured upper management it would be fine, and they could switch to Windows 10, no worries.

As for Windows 11, I think a lot of companies are waiting for the "turn off all that AI garbage first before we install it" version before switching...

The real spike will come in September

Not just because October is EOL, but because that's when the executives return from vacation and can start signing purchase orders.

I lost count of the number of companies I worked at that spent the summer queuing up tech rollouts, only to have them sit on loading docks for weeks or months because every single required signature was on vacation.

Of course, the companies that don't make the deadline, because executives kept stalling, only bothered to look at it now, and are shocked that a company with 100,000 seats can't be upgraded on two weeks notice, will be paying Microsoft for Windows 10 extended support.

How to get free software from yesteryear's IT crowd – trick code into thinking it's running on a rival PC

Several fake company names, indeed

Microsoft has several fake company and product names it uses for illustrative purposes in its documentation.

How many IBM old timers <raises hand> read this and thought "ah, PanootaCalc"?

IBM had an iron-clad rule about not naming competitors. Because they were so powerful, any mention of another vendor could be seen as an implicit recommendation, so mentioning other companies by name was forbidden. That's why their modems weren't Hayes compatible, like every other modem was, Instead, they used the "industry standard AT command protocol".

An imaginary competitor, PANOOTA, was created for documentation purposes. There was PanootaCalc, PanootaWrite, PanootaDB, etc. Of course, there was no actual Panoota company (they checked), but it was ubiquitous in (internal) IBM documentation.

When I asked a co-worker where the name Panoota came from, he explained that when they were brainstorming for a name, someone said they should "Pull A Number Out Of The Air", and so the term was born.

IBM may be the stodgy, humourless old fart today, but they had their moments, back in the day.

Microsoft dangles extended Windows 10 support in exchange for Reward Points

So basically backup supported security updates instead of ad supported

I've had a Hotmail/Live/Outlook/whatever account since around 1998. I just checked, and I've earned... 18 points. So it would take me over a millenium to "earn" a year of security updates. Oh, well, it's not like I'm using Windows anyway.

I have to admit I'm a bit conflicted over this.

On the one hand, Microsoft is withdrawing free security updates and basically blackmailing the most vulnerable people, the ones whose machines won't/can't run Windows 11, and who can't afford a new machine into tying themselves to the Microsoft ecosystem.

On the other, it is a ten year old operating system, that has been superceded, and free support for a decade matches or exceeds the industry standard. Ubuntu 14.04 was released in 2014, its' free support teir ended in 2019, and its' paid expanded security maintenance option ended in 2024.

So Windows free security support is already about twice what we have on Linux, and they are just entering now the paid support tier. And they are offering consumers a free option in exchange for those users... doing something that's beneficial for them and that they should be doing in the first place.

All products have a lifespan, after which you can't really expect support, and it's not realistic to expect Windows 10 to be supported indefinitely, especially when security issues are the result of architectural issues that have been replaced in later versions.

So, I'm gonna have to give Microsoft a thumbs up for this one.

The AIpocalypse is here for websites as search referrals plunge

Be careful what you wish for

In 2023, Canada enacted the Online News Act. If it sounds Orwellian, good. The government decreed that major search engines should pay the news sites that they link to.

To no one's surprise, except the Canadian government and the news sites themselves,, Google wasn't interested in paying sites that they were giving publicity to, but if they had to, they'd just not link to them. So, they simply removed the links to those news sites.

Even less surprising, the smaller news outlets suddenly found their traffic drop by as much as 90%. It turns out demanding that big search engines pay for the right to link to you isn't really a compelling argument when there are literally millions of other news sites that they can just as easily link to for free.

Windows 11 migration heats up... on desktops

Re: September is going to be a great time to buy refurbed PCs

a refurbished can’t do this they have to clean and install a new Windows licence

That's where the real bargains tend to show up.

The hardware is a fixed cost to administer, it's the Windows licence that's a pain to keep track of.

A refurbed Windows PC where they can't account for the original licence (because the original owner re-used it, or whatever) means all sorts of juggling to try and sell it with a licence, or, just say "screw it", and sell it as-is, making the end user responsible for the licencing of it.

When $25-$25 of a $150 PC is the Windows licence, that leaves a lot of "as-is" discounts available to non-Windows (Linux, BSD, etc.) users.

Given how much time and effort is spent on the licencing of the refurbs, units that they can dump, licence free, make life easier for the seller, and discounts for the buyer.

September is going to be a great time to buy refurbed PCs

Unlike a decade or two ago, previous generation PCs are no longer dog slow and barely usable compared to current machines. Back in the 2000s, there wasn't as much demand for used PCs because old model machines simply couldn't keep up with current software.

Today, a two or three year old machine can be had for a fifth of what they originally cost. I routinely see machines that sold new three years ago for $900 going for $150-$200 used, and for the average home user, they may even be overkill.

With Windows 11 mandating the death of perfectly usable machines in the Windows corporate environment, I expect to see a deluge of fairly recent machines showing up in the used market. All of them are perfectly usable running Linux, and the more that are being dumped, the lower the resale prices will be.

Techie traveled 4 hours to fix software that worked perfectly until a new hire used it

Been there and done that. Literally.

I wrote an application for a bank that was notorious for losing documentation. They also had high turnaround. When I left, I not only handed the source code and the documentation in troff format (it was a Unix shop), I also printed off five copies and gave it to them.

Naturally, three months later, I got a call from the new manager claiming my app didn't work, no one knew how to use it, and since there was no documentation, they were going to sue me into the ground, etc., etc.

Although my contract had included on-site support for the followup period, it also had a stipulation that bank caused errors were chargeable at three times my development rate. I'd anticipated this. So they had to approve my coming in, because it was chargeable hours, including commute time, before I'd even look at it.

They agreed, and I was greeted by no less than four bank types who were there to follow me around and document every single thing I did, as well as clock how much time I spent, so they wouldn't be overcharged.

I sat down at the system (System B) and they all watched me do the magic:

$ ./System_B_App

System B Application (C) Copyright Bank of BlahBlah 1998.

For use by Bank of BlahBlah only.

Usage: System_B_App inputfile.csv outputfile.xyz

error: input file not specified. Type "System_B_App -h" for help

$ ./System_B_App -h

This application takes the $DATA CSV formatted output file generated on System A, imports it into System B, and generates an output file usable by the $APP application. If no output file name is specified, the root name of the specified input file will be used.

Usage:

System_B_App -d: show date data - collection date, ratification date, etc.

System_B_App -h: this help

System_B_App -m: generate user manual

System_B_App -t: generate user manual in troff format

System_B_App -v: verify checksum of the CSV file

<cracks knuckles>

"The printer over there is lpt58, right?"

$ ./System_B_App -t | prn lpt58

I walked over to the printer, took the seven page user manual (there wasn't much to it, just explaining the file format fields), and handed the manager the output. I asked him if he wanted more copies, or if he could photocopy them himself, and left.

Of course, they didn't want to pay, but even their legal team had to admit that "we didn't know how to generate the documention because we never even bothered to look at the screen output telling us how to" fell under the "bank caused" error stipulation, and I was paid quite a bit of money for showing them how to type "-h" for help.

Logitech's latest keyboard and mouse combo is wired, quiet, and suspiciously sensible

I'm typing this on my 2002 Logitech Internet Keyboard (the one with the mouse buttons and scroll wheel) and using my 1999 Logitech Mouseman Wheel Mouse.

I'd say that 23 years for a keyboard (two, actually, I got another one for work, and it's still going strong) and 26 years for a mouse is a hell of a long half life.

European consumers are mostly saying 'non' to trading in their old phones

Not just Europeans

I'm Canadian, and still happily using my 2017 era Samsung Galaxy S8.

Why am I running a 7 year old phone? Because the 8 year old S7 phone that I bought used cracked the screen. While it didn't bother me that much, a friend who was upgrading to an S9 just gifted me his S8, and I didn't see any need to turn down a free phone.

Yes, I know about security issues. All the phone is used for is calls, SMS, and offline things like the (excellent) camera, offline GPS, and PDA functions like reminders. I don't use the GMail account for anything, and I certainly don't do online banking or the like. Anyone who did crack into it would be disappointed with what they find.

I'm paying $15/month for unlimited calling and text, and I think I get something like 200MB of data, which I barely even touch. And that's up from $25 a year for 30 cents a minutes calling, but no data. I rarely need any data, but when I do, it's usually important, so I splurged for the $15/month plan.

Not only do I not see any reason to trade my phone in, I'm considering spending money on replacing the 8 year old battery. I still get a day to a day and a half on it, but that's down from the five days I got on it in 2020.

Given how little I use the phone, (a) why would I want to upgrade to a more expensive gadget when the current one does just fine, and (b) who wants to buy a seven year old trade-in anyway?

Like PCs before them, phones have become commodities, and it's no longer necessary to have the current year's model, because the previous year's version is no long obsolete.

The trendline doesn’t look good for hard disk drives

There is a lot of "non-active" data out there

There was a joke decades comparing the speeds of different dedicated music processors, where the market leader bragged that "Our processor can play the Minute Waltz in 466ms; our nearest competitor takes more than 800ms".

Of course, in the real world, the Minute Waltz should play at a speed of... one minute (hence the name). The value to the customer of the improved performance was somewhat questionable.

SSDs are fantastic for performance, they're also shock proof (a major benefit in industrial areas where vibration kills hard drives like no one's business), and both the form factor and power consumption benefit from not requiring a motor.

But, for all the benefits, SSDs have some downsides, too. Cost, for one. While the small (1TB) size SSDs are price competitive with HDDs, at higher densities, HDDs are significantly cheaper. At 4TB, the HDD is 1/4th to 1/3rd the price of an HDD, and the higher you go, the cheaper HDDs are relative to SSDs.

The other issue is that SDDs aren't great for long term storage. A damaged HDD can often be recovered, even if only partially, but I don't think I've ever seen a damaged SSD recovered. Once they go, they're gone.

I absolutely configure all desktop PCs with SSDs, but my video server has 20TB of videos on HDD, not SSD. And backup solutions aren't SSD based, either.

With more desktops being replaced with laptops, the trend line for HDDs is definitely going down, but until SSD reliability, MTBF, recoverability, and cost become competitive with HDDs, I don't see data centers phasing hard drives out any time soon.

Google Cloud goes down, takes Cloudflare and its customers with it

The cloud is just someone else's computer

Unfortunately, that someone else is in Nebraska, and he went out to eat just before his machine crashed, so your internet won't be back until he finishes his lunch.

Whenever a customer refused to pay for redundancy/backup because "we've never had a problem before/how likely is that to happen", I named their server spof.companyname.com or projectspof.companyname.com. Invariably, some curious executive would ask what "spof" mean, and I'd explain it mean "single point of failure". They'd then ask what would happen when, or if, that computer failed, I'd do a rendition of Monty Python's dead parrot sketch, and they'd demand to know why I "allowed" such an oversight.

Then I'd show the emails ordering me to not install a redundant setup, there would be a flurry of activity, and a few days later, budget would suddenly appear for redundancy.

It looks like Google managers are at the dead parrot sketch phase.

I'm sure they have huge redundancy accounted for, but it's always the weakest link in the chain.

Ubuntu 25.10 and Fedora 43 to drop X11 in GNOME editions

As a fellow Mint user, the general answer is no.

Assuming your Mint version is current, you can actually select Wayland on the logon screen rather than the default X11.

I did a few times, and I came to two conclusions.

First, that Wayland isn't ready yet. It may be better architecturally, more memory efficient, and more secure, but it was still too buggy for daily use, in my experience.

Second, that while applications and most programs weren't affected, things like desklets and applets were, and of course any application that connected with the desktop directly, such as the desktop automation tool autokey wouldn't work.

The main impact will not be the applications, but the desktop. LibreOffice will work fine. But things like onscreen calendars and the like that integrate with the desktop will have to be cleaned up, and some won't be compatible at all.

That's why they're dropping support for X11 specifically in Gnome - it will force the Gnome desktop tool developers to test/update their apps to work on Wayland.

Microsoft slows Windows 11 24H2 Patch Tuesday due to a 'compatibility issue'

Re: I know that MS Windows runs on a wide variety of hardware ...

Windows actually has a far smaller range of supported hardware.

If you're talking about legacy computers, yes. Windows won't support 80486 or 32 bit machines any more that Linux does, absolutely.

But when you're dealing with peripherals, it's a different story.

If you buy any USB device, it's expected to work on Windows. Device creators write device drivers for their new USB music synthesizer, or Bluetooth security dongle, when they release it. Few of them bother with Linux.

One of the first thing Linux users have to get used to is confirming Linux support for any peripheral because we can't make the assumption that the hardware will work on Linux. Windows users can.

Re: I know that MS Windows runs on a wide variety of hardware ...

Years ago, I was hired to do the server side of a project, and the customer hired another company to do the web front end. The differences in attitude between the two groups couldn't have been greater.

The back end people mostly came from avionics, which had a "if the software crashes, so does the plane" mindset. Every permutation was tested, and the front end complained that we were holding them up.

Once we shipped, we were done. The web side said they were done, too, so the customer then changed the site passwords, and they couldn't get in. And then the bug reports started to flood in.

They weren't back end issues, they were all web problems. Of course, the web developers said, you changed the passwords so we can't administer it. And by "administer", they meant debugging. To those of us in the back end, administering meant making backups, adding hardware for load balancing as demand grew, and things like that. But to the web team, software was never actually finished, it was a live thing that they just kept changing.

I see the same difference in mindset between the Windows and Linux arenas.

Windows has a much larger user base, a much larger range of hardware and software to support, and a far less technical user community, all of which makes testing more difficult. But they also have a different mindset of "if it doesn't work this week, we'll just put it in next week's patch".

Floppy disks and paper strips lurk behind US air traffic control

I worked on a number of these systems back in the 1990s

Why are a lot of avionic ATC and rail CBTC systems still running on old hardware (80386s, 80186s, and in many cases 8085s or Z80s) and software (usually proprietary, but sometimes DOS based, going back to as old as DOS 3.1)?

Because they work, they are reliable, and they don't suffer the complexities of modern systems.

One customer I worked with in 2008 recently decided they wanted to upgrade their existing system. What was wrong with their current system? Nothing. It met all safety and reliability standards, the operators were trained on it, and it's been running for 17 years without any real issues. Was there any functionality they wanted added? Was something missing, or not working?

No, the supposed problem was that interface wasn't "current", and there were newer, shinier technologies available now. I mean, why would anyone want to still use Ethernet when Wifi 6 is available?

Uhh... because it works?

Outdated and obsolete tech should absolutely be replaced before it fails. But too many (as in, almost all) of the infrastructure upgrades I see are the result of VIPs going to trade shows, seeing new technology, and having the tech presenters say "you're not still using that old technology are you?", when in fact there's nothing wrong with it, and there's no issue to be solved.

There is a very real problem being dependent on hardware parts that are no longer made. But 5.25" inch floppies are still being made and sold (hell, here's the Amazon link for some), as are a lot of older tech. They may not be used by home users any more, but there's enough of a market for a lot of this older stuff to keep at least a couple of suppliers going.

Unfortunately, a lot of the drive to upgrade is simply a desire to appear current. I had a customer want replace a perfectly functional 2012 era Windows XP system with a more modern Windows 11 architecture. The XP system was air gapped, and not connected to the internet, so there were no security issues. But for Windows 11, they decided that they needed cloud access for some unknown reasons. The result was that the budget for cyber security alone for their new system was about 60% of what the original 2012 system cost in the first place.

They're spending almost as much money on their new system to prevent problems that wouldn't even exist if they weren't upgrading in the first place.

KDE targets Windows 10 'exiles' claiming 'your computer is toast'

Re: Alarmist?

There is a reason Apple became popular among creative types who aren't in IT, long before the hardware itself became a lifestyle brand: Apples Interfaces were, to many people, more intuitive and easier to navigate than the alternatives.

To quote my younger niblings, "This is the way".

When the iPhone came out, I compared it to my Nokia 5800, and the iPhone was inferior in every way. It cost more, it had terrible call quality, battery life was terrible, and it was unreliable. The Nokia had every feature the iPhone had, and was often far more advanced.

And it didn't matter in the slightest, because only geeks could use the thing.

A female friend bought an iPhone and was talking about how awesome it was. One of the features was the weather. Sure, the Nokia had that, too. All you had to do was go into this app, enter your city code, get the GPS co-ordinates and city code, remember them, exit that app, go back three screens, then down two menus to the weather app, start it up, then enter the city code or GPS co-ordinates, then query the weather.

On the iPhone, she clicked the weather icon, it asked her if she wanted to enter the city or just use the local weather, she clicked local, and she got the weather. No digging through layers of menus, no GPS co-ordinates, no codes, just instant gratification.

Sure, you only had to do it once on the Nokia, but you had to do it. But every app was like that. Settings were sprayed all over the place. The calendar and the alarm clock were completely different apps. The iPhone had a common design philosophy that end users could use without training.

It doesn't matter if Nokia, OS/2, Linux or other technologies "are better" on technical merit if the users can't figure them out. It's like point and click cameras. Professional photographers despise them, and will talk about how horrible they are in terms of picture quality. But an unskilled person could pick one up, point it, and take a picture with it without having to configure the F Stop and aperature of the DSLR that was simply too much effort for them.

It really is easier to install Linux in 2025 than it was to install Windows XP

The problem is that it's no longer 2001.

Linux users (of which I am one) are right when they say that installing Linux is easier now than it's ever been, and it is indeed easier to install than Windows was in 2001, or 1995, or 1990, or 1980. And it is definitely easier to install than OS/2 ever was. I say that as an OS/2 developer who worked at IBM on OS/2 projects.

The point they are missing is that it doesn't matter. It's no longer 2001. "Linux is easier to install the Windows used to be" doesn't mean anything to users who have never installed an operating system in their life.

The days when PC users knew their memory map, configured their UMB and LIM settings in config.sys manually, set the dip switches for the LUNs on their SCSI devices, and knew their memory map are long gone. The modal PC user of today would likely not even understand the previous sentence.

Most Windows machines today come with Windows preinstalled by the vendor. Oh, they configure it, certainly. On first bootup, they're asked their location, language, and other things, most notoriously their Outlook network account, so in that sense, they're installing the OS. But they're installing the version that's already customized for their PC. It already has all of the correct drivers in the image.

Linux installs great for a lot of users, and a lot of machines. But if you have an Nvidia card, you'd better know what you're doing when you run the install.

And then there are applications. There are lots of people who absolutely can use Linux for everything they do. I've helped dozens of people who just use their PC for email, web browsing, online banking, Netflix, and things like that. There's nothing there that Linux can't handle, and in many cases better than Windows. But let's not kid ourselves that everyone is like that.

In 1992, IBM tried to sell OS/2 with the slogan that it was "A better Windows than Windows", because it had a more advanced kernel, better memory protection, and could therefore run Windows apps better than Windows. It couldn't. It did for a lot of things, but many Windows apps didn't behave correctly under OS/2, or didn't run at all. Long winded explanations that it was the fault of the app for not following proper coding standards and the like didn't matter to end users; what matter was that the apps didn't run.

Wine is a great technology, and it runs a lot of Windows apps. But it's not a cure-all. It won't run Adobe software. And forget getting LibreOffice to connect to Sharepoint.

I think Linux is fantastic (most of my machines are Mint or LMDE, with one switching between Kubuntu and TuxedoOS at the moment), and it does everything I need it to do. But let's not oversell it and pretend it's flawless and that it's a plug compatible replacement for Windows, because it's not.

Please tell us Reg: Why are AI PC sales slower than expected?

I know of only one person who wants to run an LLM

And he is an architecture astronaut that the rest of the group uses as a barometer of what not to do.

He doesn't trust ISPs or mail providers, so he hosts his own mail server, which has been down for months.

He won't use an FTP server, preferring to set up a microservice architecture, with SwarmKit to orchestrate it in a VM, of course.

And he needs an $8,000 four-processor Ryzen desktop with 128GB of ram, 16TB of SSD, and 6 NICs so that everything can be run in a dozen separate VMs.

Naturally, he's trying to run his own LLM on his desktop.

That's the target audience for desktop AI. The people whose lights dim in their house when they fire up a compiler because their computers use more power than their air conditioners.

Everyone else is using web-based ProtonMail or the like for mail, they use a shared DropBox account to share files, and they run $300-$1,000 computers (for the gamers) that get just as much done, if not more, because they're not rebuilding the internet service infrastructure on their desks.

The average user doesn't care about running Proxmox, or Kubernetes, or setting up an industrial data warehouse. And they don't care about having an LLM on their desktop, either.

Computer makers are salivating over AI processors because it means being able to sell high margin machines again, and since the market doesn't currently have AI chips, there's a potential market of billions of machines to be sold. The only thing missing is any indication that end users asking for it.

Does anyone else remember 15 years ago, when TV makers were pushing 3D television sets? You couldn't find a TV without 3D for a while. How many do you see today?

Sure, AI and LLMs are important, and people will use them. Remotely. End users love email, too, but they're not running out to install and configure SMTP servers on their desktop, they just use a remote server that provides it for them. The same will be true for LLMs.

I'm sure that AI features will creep into regular desktop chipsets; that's probably inevitable. But needing a dedicated AI chip is a solution in search of a problem.

Windows 11 market share stalls ahead of Windows 10 cutoff

Wait for corporate year end

The home users will migrate when they have to. Some are hanging on for dear life and won't switch until midnight the day before they have to, but most that would already have.

Other users will switch when their machines die and need to be replaced, or when their favourite app no longer works on Windows 10 and they have to upgrade to keep using it.

But the main driver is corporate PCs, and they move in bulk at year end. A lot of fiscal year ends are on June 30th, with the new year purchasing cycle starting on July 1st. I won't be surprised to see a ton of new PC deployments in September, as executives return from vacation and approve migrations.

Also, Microsoft could start a push campaign, as they did with XP to Win7 and Win7 to Win10, to inform home users about the impending death of Windows 10 and what they can do about it.

Microsoft patches the patch that put Windows 11 in a coma

This is rapidly turning into a Monty Python sketch

Python: "We apologise again for the fault in the subtitles. Those responsible for sacking the people who have just been sacked have been sacked."

Microsoft: "We apologise again for the fault in the patch. The patch for the OS has been replaced with a patch for the patch to the OS."

Of course, this is the same day that El Reg reports Windows 11 market share stalls ahead of Windows 10 cutoff, with numerous reasons given, but Microsoft's trashing the OS not even making the cut.

Microsoft's plain text editor gets fancy as Notepad gains formatting options

Everything is old is new is old is new again

We saw the same nonsense with car models.

In the 1980s, the Honda Civic was the generic term for a small and cheap car. They also sold the Accord, which was a bigger car relative to the Civic, but still small by North American standards.

In the 1990s, both Civic and Accord grew, and by the early 2000s, the Civic was now larger than the Accord had been in 1980. And so, Honda introduced a new car, the Fit (or Jazz, as it's known in many markets), which was closer to the size of the 1980s Civic. And then the Fit started to grow, and three versions later, it's also almost as large as the Civic was in 2000.

What the brain trust at Honda and Microsoft both don't understand, or don't care to understand, is that these minimal entry level products are popular for a reason. If they are upscaled to the next level, it doesn't move the market with it, it makes the market look for a new product to replace the gap left by the product leaving the space.

I don't know of any text editor for Windows that has fewer features than Notepad. There are dozens, and probably hundreds, if not thousands of lightweight freeware editors that all have at least one feature more than Notepad. Notepad's only real purpose was that it (a) was guaranteed to be on any Windows machine, so if you needed to edit a config file on a lab PC, you could be sure it was there, and (b) it was lightweight and fast.

People didn't use it to develop thousand line C++ files or XML configurations in it, but they would use it to quickly load one of those files to make a quick one line change rather than wait for Visual Studio or a heavyweight XML editor to load.

Microsoft ditched WordPad as being useless because it bigger and slower than Notepad, but not as full featured as Word. If Notepad is going to have markups and font support, as well as integrating AI, we're going to be in the weird situation where an editor called Notepad++, which has language-sensitive editing features and text highlight will actually be the lightweight editor compared to the stock "light" editor that comes with the operating system.

What next? Will vi get AI and people start using Emacs because it's faster?

AI agents don't care about your pretty website or tempting ads

Re: Soon you'll need to browse the web through an AI to filter out the clutter

That's likely because it's being blocked at the relay level, rather than at the end user.

If you look at the SMTP gateways, you'd be amazed at the amount at how smart they've become to detect incoming spam and not pass it on. At one of my accounts, as the end user, like your GMail, I see maybe three or four spam emails a month. But when I looked at the SMTP logs, holy god, in April there were over 60,000 spam sent to my account that were simply rejected, both by regional blacklists and by heuristics.

Soon you'll need to browse the web through an AI to filter out the clutter

In the 1990s, the web was chaotic, in a wild west sort of way, Around 1998, the advertising companies got involved, and many web sites started to look like video games. This was in the days of not only sidebar ads, but interstitial ads, popup ads, popunder ads, delayed popup ads (so you couldn't trace which site/tab had triggered it), etc. And, at a time when cable and DSL were just getting started, and the majority of people still had 56kb (or 28kb) modems, and had metered internet access, many ads were playing bandwidth killing music and video.

I was working in a Microsoft shop where everyone else was running Internet Explorer, but I ran Opera to filter all that crap off. I remember once being asked sent a URL to look at, and I couldn't see any problem with it. When I looked at a co-worker's computer with it in IE, it was a completely different page. At least 90% of the page was noise, with animated characters bouncing around, music playing in the background, three popups going on, etc. It to a minute for him to load it. I could load the same page in under 2 seconds because the content was barely 5% of the page.

In those early days, only a minority of users ran ad blockers, but eventually the web became so loud and obnoxious with ads that some browsers started integrating ad blocking in their releases, and not requiring the user to install a plugin or addon (which is outside the skill level of many users). And so many web sites now check for ad blockers, and it's become an arms race.

AI scraping the web is going to learn how to bypass all that clutter. And that's going to trigger a new arms race.

Web servers will start trying to find ways to insert ads that AI can't ignore, which will make things worse for humans, so AI in browsers will be used to counter the AI in the web servers.

People who used the internet prior to 1994 will remember the infamous Canter and Siegler green card scam, where a pair of lawyers flooded Usenet with advertisements. They were the first major spammers/scammers. Prior to that, email and usenet were both pretty insecure, because no one really cared enough to lock things down. After the spammers started, SMTP servers had to be secured, and usenet relays has to be locked down to prevent scammers.

Today, email spam is so overwhelming that actual email is essentially statistical noise at this point.

AI web crawlers could be the Cantor and Siegler of the web. Today, email admins spend more time dealing with spammers than supporting actual email, and the same could happen to the web.

Three ways to run Windows apps on a Linux box

Or if you have a space PC lying around, just run remote desktop

There are only a few Windows applications that I still run because I haven't found equivalents for in Linux. There's the Kedit editor, simply because I have so many macros written over the years, there's the Take Command shell, for the similar reason of having hundreds of scripts (some many hundreds of lines long) that aren't easily, or possibly, rewritten in Bash or Python. Both work fine under Wine.

The only other Windows app I used is something called Gentibus, an app that archives media, ie. CD/DVD/hard disk contents in a proprietary database format. Although there's an app called Katalog that does the same thing, sort of, a catalog of 400,000 elements on a SATA 2TB disk that takes 5 seconds to bring up in Gentibus takes almost 15 minutes to load in Katalog.

Unfortunately, it won't run in Wine, because installing it includes a DLL that co-exists with Windows just fine, but crashes under Wine.

I tried running Windows in a VM, but discovered that for some reason, both Windows 7 and Windows 10 wouldn't set screen resolution to anything over 1280x1024, and the Windows 10 disk was thrashing constantly, which appears to be common, but not universal problem.

Fortunately, I had an old machine lying around, so I found a much simpler solution. Just configure the Windows machine with a static IP in your local subnet, but remove the gateway address. Windows will be able to communicate only within the local subnet, and not to the internet. Install OpenSSH, and you can remote in via SSH for terminal applications. If you have a Windows Professional licence, you can access it via Remmina; if it's a Windows Home licence, there's VNC, Anydesk, or NoMachine. Make it a SAMBA server, or make your Linux machine a SAMBA server and make the Windows PC a SAMBA client, and you can run the Windows apps without issue.

Running things under Wine or a VM is preferred, but I've seen people spend literally weeks trying to get a troublesome Windows app running properly under Wine, or getting the VM configuration right, when they had an old PC just lying around that they could have repurposed and solved their problem in an hour.

Empire of office workers strikes back against RTO mandates

Like it or not, RTO is here to stay

The reality for employers is that many potential employees will expect, or demand work from home, and the reality for employees is that many employers will demand being in the office.

Having been both employer and employee, there are reasons for both. I've worked in defence and other high security positions where I worked on classified material, and there was absolutely no way that I could access that material remotely, no way, no how. In some cases, even working at home on it by writing things down would actually be a criminal offence. And, of course, if you're a lab technician or work with other physical assets, you have to be in the office or on site.

The flip side is that I've had positions where I was the one-man support team for a few European customers. Being based in Canada, if there was a crisis in Berlin at 2pm their time, it made much more sense for me to deal with it from my home office at 8am Toronto time. But no, corporate rules required that I drive to work, even if there was a snowstorm, meaning I could leave at 8am and get into the office at 10:15, more than two hours later, so I could then remote into Berlin and look at the problem. I didn't interact with any local co-workers, there were no resources in the office that I needed, and it wasn't so confidential that I couldn't do it from home. It was simply corporate policy.

Companies that demand everyone be in the office without a legitimate reason "just because" are going to find that's a turn-off for potential employees, just as employees are going to find that demands to work from home full time aren't going to be acceptable to most companies, either.

I find most low and mid level managers are flexible, with most taking an "as long as the work gets done" attitude. If the company allows work from home for six months and sees a 20% drop in productivity, they've got a strong argument when employees ask why they don't allow it, or why they're more restrictive about it. Likewise, if they allow it and productivity is unchanged, or even goes up, there will be employee pushback if the company demands everyone be back in the office full time for no stated reason.

The problem, as usual, is when upper management simply blasts out edicts company wide, without considering the implications, and simply expects complete compliance.

Microsoft dumps AI into Notepad as 'Copilot all the things' mania takes hold in Redmond

There's nothing a quick, lightweight editor needs more

than an embedded LLM.

People take screenshots and throw it into MS Paint to redline the issue rather than GIMP or Photoshop because Paint is lightweight and can do the job in a few seconds.

Likewise, software developers edit 30 line text files in Notepad rather than Visual Studio because it's quick and efficient.

Notepad has no benefits over the likes of Notepad++ or any other freeware/open source editor other than the fact that it is lightweight and blazingly fast. No one uses it for coding or writing long works, there are better tools for that.

Notepad's sole use case is quick and dirty changes to small files. By embedded a gargantuan LLM into it, it invalidates Notepad's sole use case. It Notepad+LLM is going to be as bulky (if note moreso) than Notepad++, why would anyone use it?

User unboxed a PC so badly it 'broke' and only a nail file could fix it

Have you fixed tech with an unlikely tool? Have I ever

I have gone to customer sites to:

- pry a floppy disk out of a disk drive with tweezers so the hard drive would boot properly

- used a ball peen hammer to move a rocker switch that was stuck because someone poured orange juice in it, and it was stuck in "off" position.

Both of those were $300+ support calls, by the way. And I can't count the times I'd driven across the city (and twice flown across the country) to end up basically just plugging something in properly.

My best "can you fix it?" was when an VIP executive who heard I was "good with computers" came into my office in 1985 with half of an 8088 processor, and asked me if I could fix it.

When I say half of a processor, I mean it was literally broken in half.

My first question, naturally, was "what did you do?". This was the mid 1980s, and the IBM XT had been released to replace the PC. The differences between these two IBM machines was that the PC had 1 63W power supply, three slots, and no hard disk. The XT had a 130W power supply, five slots, and a 10MB hard disk.

The VIP had a PC without a hard drive, and he wanted a hard drive. So, he bought a book (I think it was a Peter Norton book) explaining how you could upgrade a PC to an XT. You couldn't just put a hard disk in a PC, there weren't enough slots, there weren't enough bays, and the power supply wouldn't take it. An ingenious vendor came up with something called a HardCard, which was basically a low power hard disk on the card, so it didn't need either a bay or an upgraded power supply.

The VIP didn't go that route, however. He read the book, and the first step was to replace the 63W supply with a 130W one. So, he unmounted the power supply, took it to a local electronics store, and said "give me one twice as powerful as this one". And they did. This is the point where things went wrong.

The company didn't have actual IBM PCs, as it turned out. They used a local clone vendor that was half the price. Their machines weren't awesome, but they worked. And their engineers make some different design decisions. For one thing, they thought that a 63W power supply was stupid, so they put 150W in every machine, PC or XT, by default.

So, when the VIP asked for "twice as powerful", the electronics store said "sure", and gave him a 295W supply. Oh, dear.

The VIP then installed this overpowered power supply in his PC clone. However, it had an extra ground wire he didn't know what to do with. And because the PC originally supported RF modulators, there was an RF line to the motherboard which looked kind of, sort of like the ground wire...

Long story short, he grounded the 295W to the motherboard directly via the RF port.

When he turned the PC on, the results were quite spectacular, as evidenced by the half of the 8088.

When I inspected the damage, I discovered that:

- the CPU had exploded up off the motherboard, with half going into the monitor on top of the PC

- the monitor, fortunately imploded rather than exploded, so the VIP wasn't hurt

- not only did the CPU split in half, the motherboard did, too

- the solder actually melted; some of it actually coagulated into a small ball that was rolling around

- the surge blew out the printer that was connected via the serial port

- you could make out where the PC had been on the desk by the scorched silhouette outline

The "unlikely tool" used to fix the problem was a directive that "nobody upgrades PCs on their own any more without talking to a tech first". It was one of the very few companies I ever worked at where I was able to requisition new hardware without getting grief from senior management.

Neptune OS is Debian made easy but, boy, does it need some housekeeping

Re: Psudo security

The idea is that if the machine is compromised and the threat actor gains login access to the user account, he still needs to root password to perform system functions. If the machine was logged in as root when the machine was compromised, then the threat actor has access without needing to find any passwords.

While TV and movies have people imaging that the typical threat is Russian hackers in bunkers cracking through a firewall over the internet while fully equations float in the air around them, the more typical real-world scenario is a user at a coffee shop leaving his laptop unattended for 30 seconds to get his coffee, and having it snatched. If he was logged in as root to do some admin work, the thief would have unlimited access to everything, but as long as he'd not run sudo in the past 5 minutes, the thief would be restricted to user level access.

This sounds like Mint with extra steps

As happy as I am to help others migrate off Windows prior to the upcoming Windows-10-pocalyse, this seems like something that will appeal to a niche collection of users.

There's nothing wrong with that, hell, there's a Justin Bieber distribution of Linux that apparently has a user base, but it's not what I'd recommend to Windows refugees.

For people coming off of Windows looking at Linux for the first time, Neptune is a new distribution, with a small team, and by using the KDE desktop, as the article states, it's going to need a lot more housekeeping/maintenance than other OOB (out of box) distributions like Mint or Zorin OS.

I'm glad to see it, and I might give it a try in a VM, but I don't think that this is really the ideal choice for new users. If a new user wants to try KDE, I'd recommend Kubuntu (a larger community with greater support) or TuxedoOS (likewise, and they even sell PCs with it pre-installed and preconfigured).