* Posts by karlkarl

1357 publicly visible posts • joined 10 Apr 2015

Page:

NASA missions are being delayed by oversubscribed, overburdened, and out-of-date supercomputers

karlkarl Silver badge

So I am assuming the NASA GPU Hackathon is just a ploy to bring your own GPUs so that they can steal them?

https://www.nas.nasa.gov/hackathon/

Trying out Microsoft's pre-release OS/2 2.0

karlkarl Silver badge

Re: Pints' on me Brian

Track down Peter Howkins (Flibble) and see if he is interested.

https://www.marutan.net/

He managed to herd all the weird cats in order to get CDE released open-source after all!

No App Store needed: Apple caves, will allow sideloading in EU

karlkarl Silver badge

Still full of DRM, at every bloomin turn.

Consumer landfill crap.

Linux 6.9 will be the first to top ten million Git objects

karlkarl Silver badge

Re: I wish Linux followed NetBSD build system structure.

Its frustrating because most downstream distros support *less* architectures than the more wholistic NetBSD (kernel + userland).

Apple's had it with Epic's app store shenanigans, terminates dev account

karlkarl Silver badge

Re: iOS App stores should not be under Apple control

I think you misunderstood.

You can update your app as much as you want. If the Apple license / DRM stops it working, there isn't much you can do as a developer.

All the passion in the world can't guess Apple's private key...

karlkarl Silver badge

Re: iOS App stores should not be under Apple control

Many developers simply refuse to engage with iOS until the developer DRM is removed. It kills passion and innovation knowing that your program won't work in ~10 years and all that hard effort will be wasted.

So once this backwards crap is resolved and iOS joins the rest of the world in "non-locked down" platforms, it will be better for consumers too.

Supermium drags Google Chrome back in time to Windows XP, Vista, and 7

karlkarl Silver badge

> Once again, we urge you not to run Windows XP on any internet-connected computer in the third decade of the 21st century

I would suggest not running *any* Windows computer directly connected to the internet. Make sure it is behind a strict proxy (the inbuilt firewall is too limited) to ensure the contained spyware (official and 3rd party) is contained and controlled.

So ultimately, Windows XP or Windows 11 makes little difference. Especially since neither should be logged into via the admin account.

Musk joins OpenAI lawsuit queue, says there's nothing 'open' about it

karlkarl Silver badge

Re: Nothing has achieved AGI yet.

Indeed.

I think any company who purchases it will be disappointed. They are buying air (or at least a glorified noise generator).

Dell promises 'every PC is going to be an AI PC' whether you like it or not

karlkarl Silver badge

"every PC is going to be an AI PC"

It really wont be for those that are effectively offline or behind correct firewalls.

Unless big strides are made for localhost LLMs hahahaha!

Musk 'texts' Nadella about Windows 11's demands for a Microsoft account

karlkarl Silver badge

Don't Musk's Tesla cars also require the same kind of DRM-esque bullshite?

Ubuntu, Kubuntu, openSUSE to get better installation

karlkarl Silver badge

Re: Not mentioned....but should be......

To be fair, I use almost identical commands to install Fedora from the livecd:

# dnf install --installroot=/mnt ...

Installers are obviously great but I wish they were *much* more transparent about what random commands they are running on your machine, as well as any needless cruft being generated.

AI is going to need a global investment, just maybe not $7T, says OpenAI CEO Sam Altman

karlkarl Silver badge

> "If I had to sit there and correct every mistaken report, I would not be able to do my job, but people say all sorts of things,"

Ironically his business is going to greatly contribute the the randomly generated noise of this sort on the internet.

AI is at best, a glorified "search box" and at worst, a glorified "noise generator". I don't think there is any evidence that this will change any time soon.

Top five reasons to move from CentOS to RHEL (according to Red Hat)

karlkarl Silver badge

Re: CERN

In some ways I bet they regret decommissioning Scientific Linux.

The commercial world simply couldn't be trusted to provide in the end.

The successor to Research Unix was Plan 9 from Bell Labs

karlkarl Silver badge

Plan 9 is an interesting OS.

I wasn't convinced by the "mouse first" approach. The CLI has a lot to offer still in comparison. However we did get later ports of Vim to it.

The git9 client is great as a project. In many ways it is so much cleaner than the upstream Git. They took a better approach than just a bunch of messy compatibility layers unlike Windows's Cygwin/Msys2 based approach.

For me, last time I checked, Plan 9 lacked mmap which is a little bit of a deal breaker for a lot of software to be ported.

I found drawterm to be very good for remote desktop. It even has a built in 9fs client in it so your disks are instantly accessible. That said, the vncs VNC server was also very straight forward to get working and slightly more portable across platforms.

Since I run my web browsers in a Windows VM anyway, I probably could use it day to day. I just prefer the command line for what I do.

Microsoft Publisher books its retirement party for 2026

karlkarl Silver badge

> the future is uncertain for those wondering what to do with all those .pub files.

Darn, what else will I use to accidentally open up my public SSH key on Windows?

Moving to Windows 11 is so easy! You just need to buy a PC that supports it!

karlkarl Silver badge

Windows as early as 95 actually had this as an accessibility option. It used an XOR buffer to allow pixels to quickly be updated rather than needing a full fat compositor.

Weirdly, many user studies, including the classic one carried out by Sun Microsystems on beginner users, actually suggested the opposite, "moving things" on the screen overwhelms the user and distract from the actual process. Operations being done atomically ended up with less cognitive load.

My personal hunch is it depends on the specific user. The only thing I know is that users tend to be experienced for a much longer duration than they are inexperienced.

karlkarl Silver badge

RDP can work offline, in the same way that X11, Wayland, NFS can.

A local-only UNIX domain socket forwarded through the VM boundary offers best performance. If the OS / VM doesn't support that, then just a simple local-only network interface.

karlkarl Silver badge

The trick is to use an open-source VM. Not specifically for "freedom" and all that, but mostly because there is very little money in maintaining support for older systems, so commercial offerings tend to ignore the use-case.

Open-source projects do a better here, partly due to the guys contributing to it typically have a passion for vintage stuff but also because they don't need to offer a "guarantee" it works, so don't try to remove it as soon as possible.

karlkarl Silver badge

> Aside from sounding like it's ridiculously and needlessly complicated

Running a VM is really not complicated. Especially one running an older OS with no DRM present.

> I always find it fascinating. In the gaming world, people want MORE antialiasing

The gaming world is really quite irrelevant. For people who work with computers, check out a project such as Spleen:

https://github.com/fcambus/spleen

Which is the default terminal font in OpenBSD (and now, FreeBSD). You will see that a lot of effort is put into raster fonts which look good without anti-aliasing. If everyone wanted blurry fonts, this project wouldn't be quite so popular would it? It is one of those things you have to try for yourself to realize that it is calmer on the eyes and clearer to provide a slight boost to efficiency. Does it look "good"? Who gives a crap?

> Unless you're conflating ClearType and AA with the GUI as a whole

The font rendering is pretty terrible for the eyes since Window XP but really I was referring to the naff effects enabled by the compositor. When a window closes, I want it to be gone. I don't want some childish fade or transparency effect wasting my time. Again, does a window suddenly disappearing look good? Again, who gives a crap?

But ultimately this is all personal preference. It is a known fact that not everyone likes the nonsense that Microsoft forces on their consumers. I just happened to have jumped off that treadmill at ~2000. Some guys do it earlier, some guys do it at Windows 7.

karlkarl Silver badge

> I'd be willing to bet 10% of your income that if you loaded Win2K into a VM

I do all of my office work on an offline Windows 2000 VM (and RDP). The office 2007 ribbon was a broken concept and adobe pdf tools were better around this era.

I recommend (everyone) giving it another shot.

> It still had a shitty 2D software rendered GUI instead of a 3D hardware accelerated one like every version of Windows since Vista

Removing the blurry effects on any version of windows (under accessibility settings) make any version of windows feel more snappy and responsive. This just comes default in 2000. It also means RDP can work more effectively in enterprise / office settings.

karlkarl Silver badge

Can they shut up, release security backports for Windows 2000 and be done with it?

Drowning in code: The ever-growing problem of ever-growing codebases

karlkarl Silver badge

Some of them I publish and release, but so many people are only interested in "newer" stuff that it often isn't worth the time.

I rarely offer patches back upstream. Mostly because they wouldn't accept them, gutting build systems, ripping out features is not the direction that these projects go. Hence the article on "bloat" I suppose ;)

Some of my work to replace the Gtk in GtkRadiant with FLTK got upstreamed in FLTK itself though (Fl_Flex): https://github.com/fltk/fltk/blob/master/FL/Fl_Flex.H

I only really care about Windows 7 and OpenBSD platforms. Linux tends to build due to the simplification of the software but I only get the urge to give it a shot once in a blue moon. Part of my work on Half-Life was to *remove* Android support because I find it a mess and was damaging the codebase.

karlkarl Silver badge

One of my personal hobbies is gutting older software projects, cleaning them up and simplifying them, getting rid of the bloat.

I have a number of carefully maintained ports of i.e:

- Blender (2.49)

- Half-Life

- GtkRadiant

- Abiword / Gnumeric

- GTK+2 (sits on SDL2)

- CDE

- DOSBox (direct framebuffer)

Often it is ripping out random dependencies for trivial stuff, replacing GUI libraries and swapping terrible build systems.

Open-source tends to grow and grow. Sometimes it is good to take a step back, have a think about the role of the core software and ultimately to make the codebase a joy to maintain.

I get the temptation though. When a project is public, you are always stretched between keeping the project "correct" and focused vs trying to appease and make everyone happy.

Closure of Windows 10 upgrade path still catching users by surprise

karlkarl Silver badge

Agreed.

I simply don't engage with DRM. It is so easy to avoid these days.

Mozilla CEO quits, pushes pivot to data privacy champion... but what about Firefox?

karlkarl Silver badge
Pint

> Could it have more to do with browser's ever-increasing irrelevance

Firefox was as irrelevant back in 2010 as it is irrelevant now.

Gosh, I wish all software could remain as irrelevant as Firefox.

Here's to the next decade of Mozilla Firefox! Well done, keep going and you will win by default! ------------------>

Rust can help make software secure – but it's no cure-all

karlkarl Silver badge

Re: FALSE

Imagine you have 2 unsafe sections:

1) One that allocates some data and passes it out as a wrapped object.

2) Another that causes the data to dangle.

You are passing that wrapped object around potentially the entire codebase. I would now say that 100% of the Rust codebase is compromised. The issue is the order / state of calls that ultimately called that 2) section. This is likely why Rust is finding it hard to establish a common GUI library; those things are very risky with regards to dangling data when wrapping and making bindings.

Yes, this is better than an entire codebase of C that could also be compromised but that loss of direct interop with C may not be worth the difference. Or at the very least, C++ is a pretty strong compromise between the two.

karlkarl Silver badge

> Because one "unsafe" block can break the guarantees of safe code around it.

Indeed. The unsafe block isolates the "entry" and "exit" points for i.e memory bugs to manifest themselves. However the bug is likely not at these points but within the wider software design.

This is further exacerbated by the fact that almost all Rust software relies on one or more *-sys crates linking it to the underlying OS (almost exclusively a C API). So there is many of these unsafe points that people don't even know about; so how can they know to be careful?

Rust is close to safe. Closer than Ada in my experience, especially where C boundaries are concerned. However in more use-cases than we like to admit, keeping with homogenous C (or even C++ with direct interop with C) is more appropriate.

IBM pitches bite-sized $135k LinuxONE box for smaller biz types

karlkarl Silver badge

"The company said that customers who move Linux workloads from an x86 server setup to an IBM LinuxONE 4 Express could save over 52 percent on total cost of ownership over five years."

Just... How?

Windows 3.11 trundles on as job site pleads for 'driver updates' on German trains

karlkarl Silver badge

Re: Improvement?

For ISA bus and parallel ports, then DOSBox (or another VM/emulator tech) really is your only choice.

ISA and parallel at least are very tricky to track down on "modern" hardware.

karlkarl Silver badge

Re: Improvement?

> replacement hardware -- Windows 3.1 doesn't run well on modern hardware

Since DOSBox is so portable, and Windows 3.1 is so light, this setup could run on much more hardware than current Windows.

Cory Doctorow has a plan to wipe away the enshittification of tech

karlkarl Silver badge

Enshittification is a great term.

For almost any dead tech product, you can go through screenshots of their releases and see the exact stages of enshittification occurring.

Oracle quietly extends Solaris 11.4 support until 2037

karlkarl Silver badge

Re: I guess no further extention to the deadline possible

I don't disagree.

But the Lindy Effect is currently suggesting that Solaris 10 will likely be the last remaining OS on earth, outliving all of us.

I mean, seriously, if I knew the lifespan that Solaris 10 was going to have in 2005, I would have absolutely prioritised it over even Linux. 20 years is almost an entire career length.

The real significance of Apple's Macintosh

karlkarl Silver badge

Re: In the early versions of MacOS it just blew up..

Indeed. Though, I would have thought that in a "debug runtime" build expensive checks could have been done to alert the programmer of this error. This wasn't done and I am slightly surprised it wasn't.

Same with C++ actually, in debug builds, similar to Microsoft's debug iterators we could:

- Lock operator-> to prevent use-after-free of "this"

- Lock operator[] to prevent invalidation of elements during access

- Return some lock passthrough on operator T& to prevent use after free during use as a parameter reference

Yes, it would run slowly (possibly slower than ASan) but in this kind of debug build who cares? It would be a really good compromise.

karlkarl Silver badge

I had a few very good fundamental ones I used as part of my thesis. I will try to dig them out. However for a quick immediate example, this article explains it quite well:

https://oberoidearsh.medium.com/dangling-pointers-tombstones-and-lock-and-keys-f6bd0791810f

This is quite good too, following this reference leads down a more academic path / approach.

https://books.google.com/books?id=To3xpkvkPvMC&pg=PA392

karlkarl Silver badge

Sure, its relating to this:

https://en.wikipedia.org/wiki/Classic_Mac_OS_memory_management

"To solve this, Apple engineers used the concept of a relocatable handle, a reference to memory which allowed the actual data referred to be moved without invalidating the handle. Apple's scheme was simple – a handle was simply a pointer into a (non-relocatable) table of further pointers, which in turn pointed to the data."

(You are right though, this doesn't *have* to be C but there was an API that provided this at the time. It is a technique that can be used by most languages utilizing pointers)

Basically, pointers went "through" a validation table containing "validity" records (formerly called "tombstones") to check memory errors.

Think along the lines of struct Employee ** pointing to a struct Allocation * and thus passing through to the first member (void *).

struct Allocation {

void *data;

size_t size;

int valid;

};

I attempted to implement my own standalone approach here:

https://www.thamessoftware.co.uk/forge?repo=stent

It has been moderately successful with a websocket proxy / web server I wrote using it here:

https://www.thamessoftware.co.uk/forge?repo=wsstent

But... can't seem to find a way to comply with strict aliasing rules.

karlkarl Silver badge

Behind that interface is an even cooler thing. A really portable approach to dealing with memory safety in C. Their pointer tombstone approach was quite inspired for the time.

(that said, it likely violated strict aliasing rules).

Linus Torvalds flames Google kernel contributor over filesystem suggestion

karlkarl Silver badge

Re: A better long-term approach...

Yes it is. Its all of ours to weaponise. So long as software freedom is our aim.

Have a read up on the recent article on the ways that the free software community is approaching this:

https://www.theregister.com/2023/12/27/bruce_perens_post_open/

karlkarl Silver badge

Re: A better long-term approach...

To isolate them from the wider community. Yes.

Then if they can no longer contribute, they will lose relevance and market-share. It is an open-source weapon that really can keep commercial machines in line.

Native Chrome arrives fashionably late to the Windows on ARM party

karlkarl Silver badge

Its amazing how much of a big deal "an ARM version of a program is" on Windows.

On BSD / Linux, you just kind of expect it. Whether the C or C++ becomes Intel or Arm machine code out the other end is barely exciting.

The Land Before Linux: Let's talk about the Unix desktops

karlkarl Silver badge

> but Linux has already become a top end-user operating system, thanks to Android and Chrome OS

Isn't Chrome OS still a tiny minority? Some schools were tricked into buying a fleet and a some elderly use it as a web browser thin client perhaps?

Saying that Android is providing Linux to the masses is very true but I personally find it a little bit... sad. I feel Linux can do so much better than that as a proper general purpose operating system. And this is wasted when misused as an over-engineered phone OS.

Wait, security courses aren't a requirement to graduate with a computer science degree?

karlkarl Silver badge

There is more to computers than the network. (Contrary to Sun's slogan ;)

An offline solution is pretty darn secure.

The rise and fall of the standard user interface

karlkarl Silver badge

Since CDE / Motif is the last GUI system that was ever standardized by IEEE. We should all just keep using that.

Or we could... wait for another to become standardised *snigger*.

Russia takes $13.5M bite out of Apple over in-app purchases

karlkarl Silver badge

Re: Apple in Russia?

It exists to extract money from people without giving them much in return.

So I can exactly see how a US company has the relevant blessings to extract money from the Russian people.

Eben Upton on Sinclair, Acorn, and the Raspberry Pi

karlkarl Silver badge

I like the Pi but using a low powered (by design) hardware to use in order to emulate another platform is ultimately a pretty daft idea.

So, are we going to talk about how GitHub is an absolute boon for malware, or nah?

karlkarl Silver badge

Re: GitHub

Not just source, I used to dump loads of random binaries on there too (basically a poor man's dropbox).

That said, I haven't engaged with GitHub since they forced consumer 2FA. It was finally the kick up the butt I needed to jump back to my own infrastructure. Everything is just cleaner and more fun again now.

Microsoft suggests command line fiddling to get faulty Windows 10 update installed

karlkarl Silver badge

> When did Windows turn into Linux?

Windows as a whole never did. But certainly the GUI experience regressed to such an extent since Windows XP that it is now on par with the Linux offerings.

A true race to the bottom!

Microsoft prepares Visual Studio 2013 for retirement

karlkarl Silver badge

Indeed. As I mentioned "offline backups just feel like a random cludge of files with no real versioning"

karlkarl Silver badge

I can't recall, is 2013 the last one with proper deterministic install media via DVD before it went, "just grab a bunch of random crap off the servers and call it an install"?

These days even the VS offline backups just feel like a random cludge of files with no real versioning.

What if Microsoft had given us Windows XP 2024?

karlkarl Silver badge

It would just be funny to see people say how "modern" it looks. It would be good proof that people's opinions of aesthetics are entirely what Microsoft dictates.

It would bubble down to something far more important though; we would have a good push to get rid of defective shite like Gnome 3+ and KDE 4+ and go back to something more useful.

Asahi's Fedora remix dazzles and baffles on Apple Silicon

karlkarl Silver badge

Re: Why?

> I guess all those apps I have installed, not from the App Store, must have been compiled on my machine then...

> They're all signed (an ad hoc signature is fine). This is not necessarily a bad thing, but some people are concerned that over the long run Apple will seek to "iOS-ize" macOS and have only known signatories, certified by Apple.

Its worse than that. Those apps, not from the App Store have been signed by a certificate that the authors have had to beg Apple for (and pay for the privilege). This is the same as Microsoft's Windows RT developer license.

As I mentioned, try compiling a binary on your machine and try running it on your second machine. I bet you can't run it without also begging Apple (and paying) for the privilege. Basically the developers who write software you enjoy using are being forced through stupid hoops by Apple.

Page: