Posts by billdehaan

Moving the Linux is easier

It actually is. I have four machines, with one (32 bit) laptop running Linux, and the other three running Windows 10.

At least I did, until one of them could no longer download Windows 10 updates because the 32GB boot drive couldn't handle a 105GB "patch" that was almost four times the size of the OS plus applications.

I read the Windows support voodoo, which recommended all sorts of nonsense that pushed OS management onto the end user (clear this cache, delete this subdirectory, change the ownership of this file, then edit this registry entry, then reboot to the USB disk and copy this file to this directory, they reboot again, blah blah blah), and even spent a couple of days batting around with it, without too much success.

Then, I downloaded a few Linux ISOs, booted off them, installed Mint on the machine, set up a Win10 VM on the other disk in case I needed to actually run any Windows app on it (I haven't in the four months since I switched), and left it that way.

The second machine I switched to Mint, and likewise haven't touched.

My primary machine is still Win10, but I'm slowly migrating things off, and will easily be finished before October 2025.

Not one of my PCs would run Windows 11, according to Microsoft. All four of them run some variant of Linux. And since I can run Windows 10 in a Linux VM, all I have to do is disable networking for the VM and there's no worry about security, either.

Why should I throw away perfectly good working computers simply because Microsoft stops security updates for them?

Linux didn't pull me in; Windows pushed me out.

Always check cancellation procedures before signing up

Like most people, I've had horror stories about the difficulty, and sometimes near impossibility, of cancelling a service. Bell Canada stands out as one where their stores, phone support, and web site all pointed to each other as being responsible for cancellations. Despite their contracts clearly stating that the consumer "contact Bell" to terminate the contract, no one could actually clearly explain how to contact and how a cancellation could be achieved.

Despite no one in Bell having a clue how to cancel an account, once I did successfully manage to do it, I received a phone call from their retentions department less than 20 minutes later, and three followup calls within a week trying to get me to sign backup.

Of course, that's nothing compared to the guy who spent 20 minutes on the phone with his phone company repeatedly saying "cancel the account, cancel the account, cancel the account" to a service rep who simply refused to cancel it. Once he posted it to the internet and it went viral, he was able to cancel it, but the company had to be publicly bullied into cancelling an account. That's absurd.

Ever since my dealings with Bell, I've made a point of checking out cancellation procedures when I've considered signing up for any recurring service. I do a search for "cancel $SERVICE Canada", and it's surprising how many of those searches link to long lists of horror stories. I'm sure it's saved me money, as I've skipped signing up for a lot of things.

There are definitely reasons to not make it too easy to terminate an account, because it could be done accidentally (any service rep can tell you customer horror stories), but it should be no more difficult to terminate than it was to sign up for in the first place.

And before iPods, there was Usenet

In the mid 1990s, I worked in the USENET admin team for a fairly large (5,000 employees+) Canadian based corporation.

Bandwidth was metered in those days, and rates differed by usage and time of day. Much of what our team did was automate tasks to optimize bandwidth usage so that our Toronto to New York backup ran at 3AM, when the bandwidth cost was under 10% of what it would cost to run at 3PM, that sort of thing.

Although the web existed, it will still in its' infancy. Most of the traffic was FTP, and there were many Gopher, Archie, and Veronica users within the company. There was also USENET.

Originally, USENET was brought in as a programming resource for the software developers. We brought in the Comp, News, and Sci hierarchies to start with. Then Misc got added, and Rec, because (a) a lot of the users were book nerds, and (b) using Rec groups got them interested in USENET enough to learn it, and then they started using the News and Sci groups for their actual jobs.

But then, there was Alt. The Alt.* hierarchy was (and is) a snake pit that we tried to keep far, far away. Unfortunately, it was inevitable that some particular newsgroup was needed from it, and it wasn't easy to bring it in without bringing in the entire hierarchy, although we tried. We wanted to restrict it as much as possible, but were overruled, and so the entire thing came in.

USENET bandwidth usage exploded. It went from being ~3% of company bandwidth to ~70% in about two weeks. What was worse was that it didn't just increase by a factor of 25, it was also being used during peak business hours, when bandwidth rates were highest. Our $800-$1200 a month bandwidth cost, that we'd been trying to get under $500 a month, ballooned to something like $18,000 one month, and was projected to be over $30,000 the next month.

Management freaked and ordered an investigation.

What I found, not surprisingly, was that the bandwidth was almost all USENET, from the Alt.* hierarchy. Specifically, the Alt.sex.* hierarchy. Yup, people were downloading porn videos. Instead of 2kb text articles, lots of 40MB video clips were being downloaded. Repeatedly.

The next step was to identify who. And lo an behold, 90% of the usage was attributable to six users. Five users were responsible for about 20% in total, with the other 70% being from a single user. A single executive. An executive of the "reports directly to the Board of Directors" variety.

Awkward....

So, without naming names, we sent out an email blast to world+dog in the company, with a USENET traffic graph, showing how much was being transferred from the different groups.

Of the top 20 groups, 18 were alt.sex groups. Fortunately, there was nothing illegal, and we didn't have to deal with child porn or anything like that, thank god.

Unsurprisingly, when the users realized that we could see the login credentials and knew exactly who was transferring what, our bandwidth went back down to about $1100 a month again.

Working with defence contractors teaches you life skills

I wasn't military, but I worked with defence contractors, so in modern terms, I would be called "military adjacent" or somesuch.

The most basic skill when dealing with the military (any military) is CYA, or "Cover Your A**".

One of the reasons that the military has a reputation for staggeringly (over)complete documentation is largely due to the culture of CYA that developed, of necessity. In militaries where disobeying orders can get you executed, it's a good idea to have it recorded, repeatedly and in several different documents and locations that you "were just following orders" when you did what you did (sorting by first name, in Steve's case).

I personally had a team lead who was notorious for saying to do X (sort by first name), confirm it, double confirm it, triple confirm it, and then when it hit the fan, would deny to upper management that he had ever said that, and that I (or another member of the team) had done X on our own initiative. If someone refused to do the stupid thing (because it was stupid), he would tattle to senior management that the person was disobeying orders. If they had proof that he'd ordered them to do X, he's say that the person misunderstood his instructions. No matter what, the subordinate was always the one to blame.

As you can imagine, he was not beloved within the team for these reasons (and many others).

So, when he one day decided to order me to do something particularly stupid, I confirmed that he meant it. And doubly confirmed. But I waited until the meeting with the big brass that was scheduled for the next day to triple confirm it. I did the "explain it to me like I'm five years old" approach, and he condescendingly spelled out exactly what he wanted done, step by step, exactly what he wanted me to do. And so I did, exactly in the sequence he'd laid out.

The results were glorious. They resulted not only in invalidating a flight test and missing a ship date, they put the entire project at risk of cancellation. Senior executives got involved. First he tried the "I never told him to do that" approach, except there were several members of the brass who'd been present to see him to just that. They didn't understand the implications of the orders, but they remembered damned well that he'd not only ordered me to do it, he'd done so repeatedly.

Likewise, the "well, he misunderstood" argument went nowhere, because I'd repeatedly asked for clarification, he'd provided it, and it matched letter for letter what I'd done, and what had caused the situation we were in.

But the chef's kiss was his statement that "if I was really telling people to do things that stupid, people would be complaining all the time", apparently unaware that at least six team members (although I was not one of them) had made formal complaints both to management, and in two cases to HR, about being backstabbed exactly like this. When they checked, there were something like 38 such complaints over a period of 3 years.

Unsurprisingly, in the next re-org shuffle about six weeks later, he was moved into the newly-created "Special Projects" group where he would be leading the team (which at the moment was just him) in said special projects, which were yet to be defined. Internally, this was later referred to as the "Ice Floe" group, named after the practice of some Inuit tribes to put their sick and elderly members who were a drain on the tribe onto an ice floe so that they'd float away and die.

Modernize, shomdernize, as long as it runs

I did OS/2 applications work on contract from about 1990 (with OS/2 v1.1) to 1992 (with OS/2 v1.3 and 2.0). I leveraged those OS/2 skills to get contracts at companies in the back half of the 1990s, although sadly, that often meant decomissioning OS/2 in favour of something else (Linux, Windows NT 3.1 or 3.5, or even... Windows 95).

I still work with companies that have an installed base of OS/2 deployments 25 year later. They are usually in turnkey solutions for control systems. eComStation and now Arca Noae have been lifesavers.

Lots of companies have clients with OS/2 software that can't be replaced or upgraded. Often the original vendor no longer exists. If it does exist, it often doesn't have the source for the legacy system, and even if they do, they don't have the skills to maintain it, let alone upgrade it. You want an upgrade to our 25 year old OS/2 product? Here's our Windows 10 version.

Unfortunately, for some clients, that's not an option. They're perfectly happy to keep running the original code, which works as well today as it did in 1997. That's especially true when talking about PCs that weren't networked. Unfortunately, when that 1997 era Pentium Pro dies, finding a replacement that will even load OS/2 is the problem.

Sure, there are virtual boxes, but a lot of OS/2 device drivers are fussy about running in VMs.

That's where Arca Noae has been a godsend. I've had customers with $25M installations with six OS/2 boxes that had a box die, then another, then another, and then another. The cost of upgrading the software is often quoted at $1M or more, when all they need are four replacement PCs that can run OS/2 v4 (or v3, or even v2.1).

No version of OS/2 will install on a modern PC. It doesn't know about USB, or SATA, or pretty much any other innovation that took place after 1996.

But an Arca Noae installation that gets the customer up and running again is worth its' weight in gold. Does it handle quad cores properly? We don't care. As long as it runs as fast as that 1998 era Pentium, and it can handle the custom device driver for the weird 128 port output controller, it spares the customer spending $1M in needless upgrades.

I can't see myself ever doing any new development in C/Set2, and I doubt many others do, either. But just keeping existing OS/2 installs alive makes Arca Noae a viable business.

And then there was OS/2

The UK has bins, the US prefers trashcans, and computers like their /bin

Due to the various look and feel lawsuits, after macOS put their "trashcan" on the desktop, Microsoft had to use "recycle bin" for theirs, to be different.

So, when IBM released OS/2 v2, they could use neither. So they settled on "shredder". The desktop was complete with a paper shredder WAV file that made the appropriate grinding noises when any object was dropped on it.

Since IBM was new to the desktop, or at least they believed their customers were, they populated it with all sorts of instructional games and the like to start with. They included a chess game (GNU chess, which they neglected to credit), and an entertainment program called Neko that helped users get used to the mouse interface.

However, many of the studious, steadfast IBM employees had little use for humour, and their first reaction was to remove it from the desktop. However, this was a system object, not a user object, and so it couldn't actually be deleted.

This led to much amusement as the OS/2 bug reports contained several variations of "I dropped the cat into the shredder but it didn't die". This became even more amusing when Very High Level Executives visited the site. These were not just Executives but Senior Executives. They were far too Senior to be familiar with (ugh) PC desktop software.

And one of these Senior Executives either was on, or his wife was on, one of the state ASPCA (American Society for the Prevention of Cruelty to Animals) board of directors. The phrase "dropped the cat in the shredder and it didn't die" was not as amusing to the Senior Executives as it was to the rank and file, oddly enough.

Been there, done that

No problem. The US tech had simply grabbed the disk from the PC running the testing and copied it to the other computers. "Obviously it worked because they are all up and running," he said.

I had a somewhat similar experience, although thankfully lesser in scope.

As in the article, I was working on a testbed simulator for a safety-critical system that was currently running in production. To be clear, this was a simulator to test an upgrade which had not been deployed yet.

We had multiple client X machines that communicated with server Y. Many of the X machines were mobile, and field testing of the upgrade showed that many X machines intermittently either lost communication, or had communication errors (duplicated, out of sequence, or lost messages). Sometimes.

The stationary clients were fine, but no one could determine the cause of the intermittent failure. The systems integrator said it was the hardware, the hardware vendor showed the hardware tests that proved the communications systems worked when stationary, so it had to be the mobility aspect, the mobility supplier ran pings and traceroutes showing it couldn't be them, so obviously, it had to be our proprietary protocol. Our protocol guys pointed out that we didn't even know if an X was mobile or stationary, so how could the protocol fail only on the mobile clients?

To investigate the problem, I essentially wrote a customized ping command to test. My ping didn't use ICMP, though, it used our proprietary protocol. I made it a master/slave application, where the master would timestamp a message packet, give it a master sequence number, and send it to the client. The client would in turn timestamp the arrival time, give it a slave sequence number, and send it back. When the master received it, it would log it to disk. The operators could configure the protocol's send rate, packet size, etc.

The idea was to try it in the field, identify which clients/mobility locations were the problem, and play with the frequency and protocol payload to narrow down what was going on.

Of course, with all the time stamping and optimizations going on, there were several issues (time syncing was particularly troublesome), and although version 1 worked as a proof of concept, they wanted more features, so I got a budget for version 2.

And then a field report came in. My tool had reported a catastrophic loss of communication in an entire range of client machines where there was actually no problem with the actual upgrade application. So obviously my tool was crap, the customer had no faith in us because we had no clue what we were doing, how could we be trusted with safety-critical systems, etc. I had endangered the entire multi-million dollar project, we would all be out on the street, our children would starve, and all because of me. The project manager wanted blood, specifically mine.

Hmm. Send me the logs, I said.

Looking at the logs, I noted that the master and slave packet formats were slightly different, which made no sense, because masters and slaves were paired. I was extending the packet format for version 2, but it hadn't been released yet. This looked almost like a version 1 master trying to communicate with a version 2 slave. So I asked them to confirm the MD5 checksums of the master and all slave versions in the field.

Sure enough, there was a version 1 master, and the offending slaves didn't match version 1. What on earth had happened?

It turned out that the project manager had been in our test lab, and seen a couple of the machines where I'd been testing version 2. He liked what he saw, so he made a copy of the software (the unreleased, work in progress software, which was being debugged) from the lab machine, and when he returned to site, he gave it to the field testers, who then ran with it.

Not surprisingly, when this came to light (not to the customer, at least), there was a reshuffling, and there was a new project manager.

I promptly added two more features to the software. First, during the initial handshake, masters and slaves verified that they were each running compatible versions, or they stopped talking. Secondly, I added a time/date check, and the software would only work for 5 days after compilation, after which it would post an onscreen message saying it was expired beta software. I disabled that check in the formal release, but it put a stop to project-ending reports from the field.

I didn't need to do it. Management did it for me.

In the early 1980s, I was on contract for a Fortune 500 company. My manager was notoriously, almost pathologically miserly. So while other teams had IBM ATs (80286 machines with 20MB hard disks) or at worst IBM XTs (8088 PCs with 10MB hard disks), our team had IBM PCs (8088 machines with no hard disk, only floppy drives), in order to save money.

Floppy disks cost about $3 each, or $25 for a box of 10. That was if you got the good ones (Dysan or Maxell), and of course, we didn't. Our manager got something like 10 boxes of BASF for $89. As you can imagine, these disks were cheaper for a reason, and data loss was very high.

Being the paranoid sort, I kept redundant backups. My source and tools grew to about 8 diskettes, so I just allocated one box to the project. One at home, and one at work, plus the original source meant I was using 3 boxes.

When my contract ended, in my exit interview, I turned over all 3 boxes. My manager very angrily said "so you're the one who's been hogging all the disks". He summoned another team member, and handed him the two boxes of backups, and ordered him to reformat them and put them in the general diskette pool, "where they should have been in the first place".

I left the company, and life went on.

A few months later, I got a call from them. The manager had gone to another office to install the component that I'd been working on. Rather than "waste" two more floppies and make copies of the executables to install, he took the entire box of disks, including the source and tools, with him. So when he slipped on the ice getting out of the streetcar, dropped the box of disks, and they were promptly run over by a car, that meant everything I'd worked on for them for over a year was lost. Source, executable, tools, documentation, everything.

Did I... by chance... happen to have any copies, even partial ones, at home?

I pointed out that would have been a violation of NDA (it would have), so no, I didn't.

Fortunately, for me, if not the company, I still had the copy of my exit interview, where the manager had put in writing the fact that I'd been "wasting" valuable floppies by "hoarding" floppy disks from the rest of the team. So if they wanted to claim that I'd been the one at fault for the loss, I had proof that I'd provided complete source when I left, and they'd accepted it.

The person one the one sounded like I wasn't the first person to tell him that, and this wasn't the first time this had happened.

I had a customer with almost the exact opposite

We have a couple of graphic designer/artist/musician types as clients. They're brilliant in their fields, but as far as computers go, they require that their computers be, to quote one of them, "blonde proof".

Last year, one of them had their external USB backup disk die on them. So, they got a shiny new 4TB MyBook as a replacement, and they said things were good.

A few months later, things were less good. Their C: drive was starting to die, making hideous belt sander noises, and losing critical files. But when they tried to back up to the USB disk, it was out of space. Panic ensued, as the primary disk wasn't backed up, and the backup disk had no space.

I had them send me screenshots, but the external disk was 4TB, had 3.98TB free, and they seemed to be able to copy some files over, but only about 1% of them.

So, I went over and checked it. My first assumption was that it was formatted as FAT32, and any files over 4GB wouldn't fit. Nope, that wasn't it.

I had her show me exactly what she did, so I could see if it was a PEBCAK error, as it usually was.

She double clicked on the icon of the USB disk, which opened a Windows Explorer window, as excepted. She then dragged a folder from her desktop to it. And sure enough, an "insufficient space" error, with a Windows error code number appeared.

She, of course, was panicking that she was going to lose years of work, and rightly so. I tried various things, but there was no issue. I put the external USB on my laptop, and there was no problem writing to it. The disk wasn't write protected, it didn't need administrator rights, it wasn't FAT32, and in fact it could copy about 300MB of files because it gave the "insufficient space" error. What the hell was it?

So, I researched the Windows error number. Strangely, it was not a file system error, it was a OneDrive error code. WTF? She wasn't even using OneDrive. Or was she?

Sure enough, OneDrive was enabled. She hadn't configured it; she had no idea what OneDrive even was. This was one of those "it came that way when I bought it" things. Either the box store had configured Windows for her, or it was done when she set up Windows the first time. Since she was a "click yes to everything" type user, it would be whatever Microsoft sets as defaults.

In the end, it turned out to be one of the more malicious things Microsoft has done. When OneDrive is enabled in Windows 10, when you copy from one drive to another using Windows Explorers, it backs up the destination disk on OneDrive. It's totally seamless and transparent.

Of course, if there isn't space on OneDrive, the copy is aborted. And that also aborts the local copy to the USB disk, too.

Yes, that's right. Her backup was failing because she was trying to back up about 2TB of data to a 4TB disk, but OneDrive only had 5GB or so, so Windows would only allow her to copy 5GB to the disk.

This idiocy could be bypassed by using the command line, or a third party tool, or another file manager, but this was her workflow.

By logging her out of OneDrive, and disabling, and then removing OneDrive so it didn't restart at boot time, she was actually able to use her 4TB disk.

Another "improvement" that makes things worse.

People went to digital photography to get AWAY from this

Back in the 1990s and early 2000s, there was a "think of the children" panic in Canada, and crusaders went on the tear to get the police and government to "do something" to stop it.

In the middle of this climate, I know of three cases where people ended up getting visited by police investigating them for alleged child pornography.

One case was a Japanese anime, as in, a cartoon, with no actual humans being filmed, let alone children.

The other two were the result of photo development. Those old enough to remember actual film cameras know that unless you had a darkroom, chemicals, and skill, you needed to go to a photo developer to convert your raw film into actual snapshots. Camera stores did it, of course, as well as specialty outlets like Fotomat, but one of the most common photo development places was, oddly enough, the pharmacy. And it was pharmacies that called the cops on two people getting their photos developed.

The first case showed the shocking picture of a nude 5 year old boy with his pants on the sidewalk with a scantily clad 3 year old girl next to him. In other words, a 3 year old girl snuck up on her big brother and pants his swimsuit on him. Mom happened to be taking pictures of her kids in the pool, and couldn't resist getting a snap of her kids pranking each other.

The second case was similar, with a grown woman in a bathtub with a 2 year old boy, who decided to make an obscene gesture to shock his mommy just as Daddy walked in. In other words, a typical "Jim, get in here and see what your son is doing" family moment.

Fortunately, in both cases, the police officers were parents themselves and not idiots, and when they visited the families and saw that the kids photographed were the children of the photographers, they realized that the photo developers had completely over reacted. But as you can imagine, those families stopped sending their film out to be developed, and went to digital photography.

Now, you don't even have to drop your film off to have busybodies report you to the cops, your camera vendor will do it as soon as you take your picture.

There's no way that this won't be abused, both by companies, and governments.

Shut up and take my money

If you're tempted, the handset will be available in March. Bus tickets and c are sold separately.

Could someone explain to a poor Canadian what the "c" in "and c" refers to? Probably a party drug reference, I'd guess, but I don't know the lingo.

Oddly enough, I still have my Nokia 5130, the thicker, cheaper version of the 5310. It was my daily driver from 2009-2010, and it's been a solid backup performer/emergency car phone since. The battery just died in January via overcharge, and I got a replacement Feb 2nd. Instructions said to charge it up and let it run down full three times. I've done so, having charged it on Feb 2nd, Feb 17th, March 2nd, and now... today.

Yes, two weeks plus a day or three on a charge. Ah, the good old days.

The only real complaint was that the PC/Phone software is, by modern standards pretty horrific. And as I recall, it could only take 1GB or 2GB microSD cards; 4GB or larger aren't recognized. So a 32GB capacity would be welcomed.

As I recall, it happily supported Bluetooth, but no wifi. There wasn't much in the S40 that needed the net back then, to be honest. This was not a phone to email or browse on. And there was no GPS for similar reasons.

The camera was tolerable for car accident photos, but little else (there was a green tinge on everything), but it ran for weeks on a charge, got reception when no other mobile would, matched MP3 players that cost more than the phone did, and being Nokia, a friend's daughter put hers in the wash cycle and was annoyed that the 5 key was "a bit fiddly" after the wash, but that was about it.

I can definitely see a market for this.

OS/2 all over again

When IBM tossed OS/2 onto the scrap heap of history, there was considerable hue and cry from the OS/2 user base, asking (or in the case of Team OS/2, demanding), that IBM release OS/2 as an open source project.

That was 20 years ago, and it still hasn't happened. It didn't happen then, and it won't happen now.

All other considerations notwithstanding, there are one major reason IBM didn't open source OS/2, and why Microsoft won't open source Windows 7.

Never mind the proprietary intellectual property that's embedded in all that code, which would have to be audited before release. Ignore the fact that they'd be giving away a code base that cost hundreds of millions of dollars to develop, over the past forty years. That doesn't even matter.

What does matter is that Microsoft doesn't own Windows, any more than IBM owned OS/2. Oh, it owns a huge percentage of it, to be sure. It's developed it over the past four decades, and the overwhelming majority is no doubt Microsoft spawned, for good or ill. But there are components in there, lots of components, that were purchased with conditions from other vendors, and/or licensed. Microsoft would have to either get free and clear title to the dozens (hundreds? thousands?) of licensed pieces of code that they've acquired and incorporated into Windows over the years before they could release it to the public.

That would be a major cost, and that's not even counting the potential legal exposure if they got sued for releasing something that they didn't have full legal claim to release. Parts may wind up on github, but lots of things are tied up in cross licensing agreements that will never be released to the light of day.