Re: humans vs bots
Then we all need to start fighting back against that. There have been plenty of corporate giants that ended up digging their own grave. Look at RCA...
1315 publicly visible posts • joined 6 Jun 2013
...from the Cathode Ray Dude about the insane attempts made by PC manufacturers about 20 years back to try and "quick start" their PCs by bundling whole other operating systems with them, and in some cases even resorting to some utterly insane "bury flash memory inside the motherboard" approaches. All this because Vista was so huge compared to XP and spinnys were so slow it took forever for the damn thing to start on the PCs of the day - and that went double with all the shovelware and other garbage manufacturers insisted in sticking on there that you had to scrape off before a new PC was even slightly usable.
20 years is about right for a generational cycle and everything old to be new again, right? I hear My Chemical Romance are back on tour...
...not that I don't agree with the principle, but surely anything that's not really obviously AI generated - things that are intended to mislead - well, misrepresentation can already be a criminal offense under certain circumstances. "That's AI generated!" "No it isn't." "Is." "Prove it" and... how, exactly? How do you prove that an image has been doctored?
I'm glad they're trying, but I see this getting messy.
...isn't so much the idea that it should exist - in many respects I think it's a good idea - it's what they'll try and use it for once they have it.
If I could link up my passport, driving licence, national insurance card into a single thing that seems like it could be very helpful. But what we're talking about there is a form of identity credential. A very powerful one.
I would not want to be handing that over to... well, sort of anyone really. It's something that only the government should be in a position to issue, and only the government should be in a position to see. I mean, they already are. You don't need to worry about the government holding that sort of information about you. It's them that issued it to you in the first place. As an exercise in tidying up complex sprawling backend systems therefore I'm sort of fine with it.
But if I'm not happy handing it over to any non-governmental entity in that case it becomes immediately useless as a proof-of-age or whatever token. I am not giving something with that much personal information that's impossible to revoke (Looking at you biometrics) to any private company, who will inevitably - regardless of what their "privacy policy" says - attempt to monetise it.
I'd want to see laws around some sort of key-exchange mechanism which makes this information completely *unavailable* to 3rd parties except in the form of some sort of well documented API where I have to explicitly consent to each individual attempt to access it - and they don't even get to see what it says for the most common uses like proof of age. You query the endpoint, I get one of those "Hey, this business wants to access the 21 or above" endpoint, do you agree?" notifications, and I can approve it or refuse it.
That sort of thing is complicated tho. There's a whole stack of infrastructure required to make something like that work, and I don't really trust the government to do that without it accidentally going 500% over budget and only being delivered after the underlying technology has already been proven obsolete.
so... yeah, ok, no. We're just better off without this.
...don't. I'm sorry I even wrote it now."
Probably my favourite. It's in the manpage for a quick and dirty "Serve absolutely anything - binary, text, the output of an executable - anything - wrapped in HTTP over an arbitrary network port" tool.
IIRC I wrote it because I was trying to get a file into a really early Windows95 VM by abusing the fact that IE4 or whatever ancient version it was would treat anything that arrived as a result of an HTTP request as a file if it didn't know what else to do with it and try just offer to write it to disk, thus neatly side stepping the fact that I didn't have scp or nfs or anything sensible to move files around with.
Oh god so much this. People bitching about battery life and yet every phone is advertised as being "0.1mm thinner than last year!" while at the same time having a massive camera bulge that makes it about a half inch thick *anyway* so why not just make the whole bloody thing that thick and fill the space with a decent battery?
As for the default "all devices must have at least a 6" screen diagonal" now - it's infuriating. If you have small hands it;s basically impossible to buy a phone these days where one can reach the entire keypad one-handed leaving you resorting to the sort of two-thumb typing that blackberries were actually good at and touch screens sodding aren't. Bring back the Xperia Compacts!
...my problem with it is that people seem to think it's some kind of magic bullet that means you don't have to understand or be good at something to be able to do it.
This is not true.
While I can get AI to spit out bash scripts and simple python programs at a speed I could never possibly produce them myself, the only reason I know they're any good - and oftentimes the first few revisions really aren't any good - is because I could, in theory, have just sat down and written them myself. The difference is that I can get the AI to do it over a dozen iterations and gentle corrections in a couple of hours, where as I'd have had to spend several days typing it all out, writing tests, doing the tests, refactoring... end result is I'm somewhat more productive than I was before I had it in terms of getting things done on time.
What I'm not is a better programmer.
What I'm also not is enjoying the code reviews, because clearly there are plenty of people who are happy to have the AI write their code for them and submit merge requests without having read it themselves. The number of things I have to reject because - yes, they pass the pipeline - but dear god look at the code man, can you not see how dangerous that would be if you let it lose on unsanitized user input rather than the well understood environment of our CI?
So... Yeah. I guess I agree with Linus. AI has a place. It can be useful. But like most useful tools it has downsides - and that's before we get anywhere near the ethical or financial arguments, both of which I think are certainly due consideration and I don't think I've really hammered out my own complete opinion on either yet.
"Look you're British, rein your expectations in a bit."
Maybe once we've gotten the hang of regular IT projects like running a post office without sending innocent people to prison, or a council SAP system that doesn't bankrupt a city of over a million people we can start thinking about the big things.
I have a particular personal gripe about people (and linters! especially sodding linters) who try and make themselves arbiters of what is and is not acceptable code.
Did it compile? Then it's correct. The compiler* sets the standard for what is and is not valid in any given language. Anyone who insists that you're not allowed to
function(){ tiny one line function }
because it has to be
function(){
tiny function that's now three lines long in the diff
}
or some bullshit like that can fuck all the way off. It compiled. It was correct. Stop complaining about it!
*or interpreter, if that's the sort of language you're using.
"Good methods might not really exist. We should consider some options, but I think one of the most important things we must admit first is that, if we fail to find good methods, that still doesn't make bad methods good."
This right here. This "Cure" is way more dangerous than the disease.
Vista, ME and 8 were all better for the user than 10 is. I mean, they were awful buggy crashy pieces of crap, but they didn't set out to randomly move things around and hide them from you between one release and the next. You could learn to live with them once you understood their quirks and flaws. 10 appears to treat the person in front of the keyboard as a direct adversary to be thwarted at all costs. It felt like dealing with a really obstructive supermarket. Just when you think you've worked out where the popcorn lives they reorganize the place and now you can't find it again. Or you've finally found a really good own-brand alternative to something and out of absolutely nowhere they just stop selling it.
I can't speak to 11, because I rage-quit Windows about 8 years back and have been entirely happy with that decision.
Yeah, see, this right here. We simply don't believe the "You can't have online games without us!" argument because we're old enough to remember when we did have online games without dedicated servers that could be revoked with no comeback getting in the way.
...someone decided to outsource the IT department - we'd gone cloud native y'know, there's really no need for a dedicated team of people to run hardware on site, every dev could look after their own cloud etc. etc. etc.
Fortunately for me as the team lead of said IT department at the time I saw this coming and jumped ship to one of the dev teams several months before the IT department got cut in a decidedly sideways "promotion" that left me gainfully employed when the restructure kicked in. For months afterwards no one revoked my access to the server room or IT build space - not least because the IT team had been responsible for the system that controlled site access and there was no one to do it.
Since there was a bunch of hardware that really couldn't move to the cloud (Not least the afore-mentioned site access control that was hooked up to the actual doors, but also things like the phone system, the CCTV system, a few random bits of testing hardware plus some big Windows boxes to run some weird proprietary software that required a hardware licence dongle) and someone needed to look after it I continued to just let myself in and fix things whenever someone reached out to me on slack and pointed out that some box or other needed rebooting or what-have you.
Now via a weird quirk of the org-chart the IT team actually reported through the information security team, and eventually a new head of infosec was appointed who - arguably correctly - promptly pulled my access. I wouldn't have minded but he was really quite aggressive about it and made the whole thing extremely confrontational by virtue of not telling me first and the first I knew about it was showing up to work one day and being locked out of my office before being dragged up to HR and made to hand over my keys.
Fair enough, here you go, I'll be working from home from now on then if you don't need me here.
I think it took about a week for the first thing to go wrong that no one on the outsourced team knew how to fix and I started getting the "Can you go and look at this box for us?" messages.
Nope. Sorry. Not my job. Not been my job for months. Couldn't if I wanted to.
The same head of infosec who pulled my access got involved and wanted to know why I was being so "obstructive" to his team. I pointed out that I literally could not do what they wanted because he had taken my access card and keys away and he ought to get my replacement to deal with it. Turns out I was never replaced because he had decided that us "cable pluggers" weren't required in this modern could-centric world.
Boy have you missed the point.
What hardware is this perfect firewall of yours running on? Is it one that has an OOB management chip hiding inside it, secretly inspecting network packages as they pass through? It is? How do you know that the firewall didn't quietly pass anything directed to a list of TLA owned IP addresses straight out to the internet before passing it to the firewall software who's rules promptly dropped it?
I dunno about "bad" but by the time we got to Chapter-house it had gotten seriously, seriously weird and tbh kinda disturbing.
Edit:
"I am not willing to risk my cerebral cortex on their combined efforts, and I very strongly advise that nobody else does either."
I, unfortunately, have read some of those, and from a position of superior knowledge I strongly support Liam's position. Those are hours of my life I'm not getting back.
To be fair to it, systemd is not pointless. It does a bunch of things that are not trivial to do any other way. It has uses.
If it's good at any of those things - or for that matter even better than the alternatives that ended up stillborn because all the effort went into systemd - is a different matter entirely.
When was it, do you think, that technology moved from doing useful things that made our lives easier, into this highly financialised sort of arrangement where it's all about "adding value" but no one seems to be able to tell what the value is?
2000: We've got a phone where you can get email! Me: Awesome - I can be on call and not have to carry my laptop around with me. That's very helpful.
2026: We've got glasses that record everything that you see and upload it to the mothership for analysis! Me: ...um... how does that help me?
Not mine it's not. It's a thing I carry around so I can keep in touch with people. Mostly via SMS. I actively avoid using it for reading the web (because it's bad at it, it's too small) or doing email (can't type on it) or watching video (too small again).
My phone is a phone. If you replaced it with the one I carried in 2005 it would make very little difference to me. The centre of my digital life is my desktop PC and I just really don't see that changing.
...based on my interactions with Gemini and other LLMs.
They tend to start well, but if you spend too long on one thing weird errors start creeping in until you get to a point where they enter a total state of meltdown. You'll fix a bug and they'll re-introduce it next iteration and then profusely apologise, and come up with some totally bogus fix that technically fixes the problem, but makes something else worse rather than go back to the way things were before. They seem to have an almost pathological aversion to regressions. Trying to get them to undo something they've done wrong and just put it back the way it was is damn close to impossible. They'll come up with increasingly insane explanations for why you can't do that - despite the fact that you just did that it was right there. 5 prompts ago. See? In that output there? You did the thing. That version worked fine. Now put it back the way it was! ...nope. Won't put it back. Can't put it back, and if you the user try and force it by saying "here's the block of code I want you to replace that with" it'll accept it, and then randomly revert it a few prompts later.
These things can save vast amounts of time typing out boilerplate and being effective copy-paste machines with a nice autoformatter. If you want a simple bit of bash or python or C that hundreds of people will have written something just like it before, but taylored very slightly to your specific circumstances then Gemini is excellent at that. But when you start getting into anything more than - in my experience - about a thousand lines of code, it all goes off the rails.
It's absolutely not a full sized keyboard. That stupid squished up arrow block and home+end block is just horrible and probably the worst thing about trying to do serious work on a laptop type keyboard.
It's also not a proper full travel mechanical - which admittedly your co-workers might appreciate compared to the mighty noise that a lot of those make - but that's part of what I like about them. There's really strong tactile and audible feedback when typing.
You can't use it while traveling - on the train for example - or in meetings where you need to bring your data and talking points with you, but when it is tied to your desk it doesn't have enough clout to be a proper workstation.
I am strangely attracted to the idea - probably because I have all kinds of pent up 80s nostalgia for the form factor - but looked at rationally I just can't see a use for this that isn't better served by something else.
...Since the second half of December 2025, agentic development workflows have accelerated sharply.
As per the article.
So they're admitting they did this to themselves, and that the tools they're using aren't up to the job. So... Stop using them please? ...and stop trying to force us to use them while they're at it?
...I need a key file that exists on my device that anyone who manages to somehow gain access to my device will be able to use.
Unless I put a password on the key file at which point I now need a password and some device - which is basic 2fA which I already have for most services that support it.
Except I presume people are expecting me to use a password/pin/pattern/thumbprint/faceprint to unlock the device, which is considered "as good" as a password per key file, except clearly it's not because now one only needs defeat one password/pin/pattern/thumbprint/faceprint to get access to all my key files as opposed to having to break each password separately.
How is this better again?
Never mind stranger danger, think of the "I've fallen off my bike and messed up my leg and I can't walk the rest of the way home" call, or the "The bus did that thing where it bypassed our village again and now I'm 5 miles up the road in a massive rainstorm" call. Both of which I actually made (On my massive Nokia brick) when I was at school in the 90s.
People seem to forget that the countryside exists and kids live there.
...that said I can't imagine any 2020s parent allowing their kids to make their own way to school the way I did in the 90s.
You don’t need no gun control, you know what you need? We need bullet control! We need to control the bullets, that’s right. I think all bullets should cost five thousand dollars… You know why? Cause if a bullet cost five thousand dollars there would be no more innocent bystanders!
...with their shitty packaging.
Now, MSI packages are actually good*. They have mechanisms to do sort of everything you might want from a package manager. The MSI system is not an inherently bad system.
BUT MICROSOFT DON'T ALWAYS BLOODY USE IT!
Admittedly it's been the better part of a decade since I was a Windows admin on the regular, but back then it was obvious that MS often don't follow their own guidelines for packaging. This is insane. Even when they do use it they don't always use it properly; only filling out the "install" part of the package manifest so that it either doesn't remove cleanly or doesn't upgrade cleanly.
Not only is it not really better, but in a lot of cases it's actively worse - because by containerizing it they introduced a bunch of the problems and overhead we all know and love Snap packages for. And just like Snap vs deb you find yourself in the same situation where bits of your system are installed using one, bits are installed using the other, and then the update system has to manage both and sometimes they're properly independent from one another and then sometimes they're not because something packaged one way has a dependency on something packaged the other way and if we're updating this we also need to update that, but that's a separate kind of update process so hold on a second while I spawn a new process thread to deal with that and try and resolve the dependency hell...
It didn't need to be like this if they'd just insisted in the MSI spec that you simply were not allowed to leave bits of it out, and then ensured that everyone who worked there actually used the sodding thing instead of wrapping self extracting .exe files.
*or at least not actively bad.
This is because people like you and I see failures as something to be avoided. Something goes wrong -> That is Bad. It's a very simple workflow. Sadly this is not how most businesscritters see it.
As long as there's someone else to blame, and therefore sue, and therefore it doesn't in the long term cost too much, they actually don't care about things not working. Everything can be reduced to a profit/loss calculation, at which point the cloud can start looking really, really good. I mean sure, it doesn't always work and it exposes us to enormous risk, but if it does fail there's someone else to blame and in the mean time it's really cheap in up-front costs which will make things look nice in the quarterlies. Collect bonus. Move on.
Oh god, yes, the academic data. I once got emailed - back when I was working for The University - a file named "output.o" by a desperate graduate student who had been sent it by their tutor with no explanation for how to read it, just before said tutor went on holiday.
On closer inspection it was about 15M of text data in the format [$long_floating_point_number,$longer_floating_point_numer] 12 times per line, over and over and over. I have no idea what this could possibly represent and neither did the poor grad student.
I'm guessing it was the output of some script or other in a totally custom format that the professor had made up on the fly. When faced with something like that you're screwed.