Re: Short cuts make long delays, but at least nothing shorted out
Maxim 48: "If it ain't broke, it hasn't been issued to the infantry."
70 publicly visible posts • joined 19 Jul 2024
those "warfare semi-automatic guns" you're talking about are civilian-market versions of military rifles, yes. But the market for them is mostly veterans. They get to keep up their practice at the range with a version of the rifle they've become so familiar with, except chambered for a civilian round and with no burst or auto mode. They are civilian-grade guns cosplaying as military hardware so that folks already familiar with the real thing can shoot something that handles like what they are used to, but they "look scary" so... Personally I always preferred a magazine-fed bolt action but to each their own
A nationally enforceable magazine capacity limit would be useful here, true. I don't see any reason for a 30-round box on a civilian rifle. Even in a self-defense scenario, if you need 30 rounds in a civilian defense situation then you're one of: (a) a piss-poor shot, (b) in such deep shit that the extra ammo won't help, (c) trying to play Rambo instead of firing defensively to get yourself the hell out of there.
Let me ask a question about your time in the USA: How many did you know, at least seen in person, which should NOT own a gun. But you couldn't do anything about it. I am not against guns in USA, it is the lack of control.
Ok, amongst the folks that I actually met in person:
More folks that shouldn't own guns than I could count. Heck, some of them I wouldn't trust with a butter knife.
Folks that shouldn't own guns but did anyway? A few..
One was a cranky old geezer who never left his small ranch except to go to church. So no danger to anyone who wasn't trespassing and annoying his livestock (although he'd be much more likely to just set his dogs on them than reach for his old 12-gauge)..
Another was a former colleague who I was surprised and a little horrified to meet at the range one weekend. Second time I saw him there I reported him for safety violations that scared the shit out of me to watch. That got him banned from the range. The range owner was also the county sheriff and saw the video of his behavior on the range that day. Blacklisted from ever getting a carry permit ("May Issue" state,) and with that showing up in his background checks he'd not get any job that required him to be armed. So while he could legally own it, he couldn't legally carry it concealed, it had to stay in its case unless it was at home or at the range. With the nearest range he wasn't banned from a two-hour drive away, he lost interest in gun ownership pretty quick.
Almost all the others were irresponsible or downright unsafe hunters. Their rep gets around the hunting community pretty quick and almost all of them run afoul of Fish&Game eventually.
There are actually controls in place already, they just need to be enforced - eg by closing the gun show loophole for background checks (without screwing it up for legitimate private sellers.)
Speaking as a Brit who lived 20 years in America, I think we need a little perspective on the whole "American Gun Culture" meme.
Let's start with a basic fact about the USA. It's big. A lot of it is still very wild. Even in the bits that aren't truly wild, there's a lot of it where your nearest neighbor might be two miles away. So, starting from there lets look at some natural results of that.
Hunting is a very big thing in most of the USA. It's managed, controlled, most states set quotas for how many does and bucks can be taken each hunting season to control the deer population in the (human-created) absence of their natural predators, wolves and cougars. If there are hunters, there will be rifles. I'll freely admit that I put one or two deer in my freezer every year courtesy of the .30-06 rifle that lived in the safe in my study.
Those wild or almost-wild spaces do have a human population. When you live in those places you almost always have a shotgun for vermin control and maybe the occasional bird for your table. This isn't a uniquely American thing, even in the UK "I'm a farmer" will usually be considered "sufficient reason" to apply for a shotgun certificate by the local plod. Then there's the fact that, in the USA, a lot of those sparsely populated bits of landscape are shared with wildlife that wouldn't object to discovering what you taste like and are more than capable of pursuing that interest if given the opportunity. There are countless wild and beautiful places in the USA to visit, but in many of them wandering the landscape unarmed would be almost stupid.
So, we have an unavoidable presence of firearms in huge parts of an even bigger nation. And if they are there, they will be everywhere. City-based hunters will be taking their guns home. Folks with nefarious intent aren't going to gripe about driving a few hours to a rural area where buying a firearm is normal and won't raise any curious or official eyebrows. So the cops are armed, because the criminals are. And, because the criminals are, you find folks in a situation like mine where a data center is in a bad part of town and all the local uglies knew that anyone coming in out of hours likely had "useful" access to banking servers... It wasn't mandatory, but my employer strongly encouraged folks who might respond out of hours to acquire a carry permit and a sidearm.
And then we get to the simple fact that recreational marksmanship is fun. Putting a couple of boxes of rounds down the range regularly and watching your skills improve is immensely satisfying, like succeeding in any sport is.
That's the vast majority of gun owners in the USA right there. Folks like me who have never pointed a gun at anything that wasn't a paper target, a clay pigeon or something I intended to eat. Unfortunately the American population is as big as its landmass, so it doesn't take a significant percentage of idiots or criminals to have a LOT of people prepared to acquire weapons and misuse them. And they are the ones that are loud and make the news.
I no longer own any of the ones I had, by the way. I sold them before coming back to the UK because I didn't need them for my lifestyle here. The only one I might have wanted to keep was my shotgun, but as a pump-action that held more than two shells it would have been problematic to import. So I sold that one over there and got myself a nice double-barrel for lots of clays and the occasional pheasant that makes the mistake of deciding that destroying my garden is a wise course of action.
During the pandemic I needed a quick supplement to my cash flow so took a homeworking stint taking support calls for a significantly sized ISP. Even though I hadn't sat a helldesk in decades.
Between the flood of talking elderly customers through how to reset their email password I got the occasional knowledgeable guy on the other end. Maybe 2% of calls. A significantly higher proportion were folks who'd watched a youtube video and had tried something "clever" and were burbling what was clearly BS at me.
The calls where the guy on the other end knew his arse from a hole in the ground were far and away the easiest. They'd describe the problem clearly, give me what data they had already, couple of quick diagnostics to confirm and done.
Problem was, the toolset was set up to enforce following the script. - they'd ding you on every step you didn't check off as complete and without the completion code for the entire workflow you literally couldn't raise a ticket to a "higher-level" tech or dispatch an engineer. My solution was to learn the toolsets decision tree. Took me about two shifts. So I'd gather data, know which diagnostic tools to run to confirm/deny my suspicions, then click through that decision tree in about 20 seconds to get to the solution I already knew was correct.
The youtube sysadmin who'd screwed up his setup was almost as easy. I could work that decision tree to "remotely fac-reset their router" in my sleep.
I can understand why they enforced the script. My colleagues didn't have my decades of professional geekery behind them. The best you could say for a lot of them was that they (theoretically) met the requirement in the job ad for "basic computer proficiency." Without the script they'd be as much use as a chocolate teapot. They'd run the script, click on the buttons in the diagnostic tools with no understanding of what was actually being tested, then bash in the code it returned in the same state of blissful ignorance.
I was actually a customer of the same ISP too, so when I had to call in I didn't bother telling them the details of a problem I'd already diagnosed, because I knew they wouldn't (couldn't?) get it. I knew the exact words to say to quickly guide them through that same decision tree on the shortest route to get to where I knew they needed to be.
One very annoying colleague got his entire cubicle filled with packing peanuts and encased in pallet-wrap. We stopped short of slapping a faked-up shipping label to "Anywhere but here" on it, thinking that was a little too snarky.
And then there was the ongoing institutional prank, like when a developer checked in some code that bombed the nightly he came in the following morning to be greeted at his desk with a life-sized cardboard cutout of a barely-dressed Kate Moss, her decency only preserved by the sign she was holding. We had modified the sign to read "I broke the build" and he had to keep it in his cube until somebody else messed up and he could pass it on.
Don't we all know how that happens...
Security controls, change control checkpoints, business controls, folks almost always find workarounds without even realising that they are doing it.
The more critical the control, the more essential it is that the process is designed in such a way that compliance is the path of least resistance. Because if it isn't, even in the most idiot-proof process the universe will eventually build a good enough idiot to break it.
It's not just the idiots with their nonsense diagnoses, they are also poisoning the well for when folks like us need to call a support line. They are conditioning helldesks everywhere to just lump our (somewhat) more credible analyses in with that lot. Like trying to find a story that's not written by AI on youtube.
... when if you got into a system you'd have a good nosey around, do no damage then claim your bragging rights by leaving the sysadmin an email from themselves saying "Gotcha. Here's how to patch the hole you left open."
Some guys preferred to do something "amusing," particularly if the opposition had previously got at them or if that "nosey around" showed they "had it coming." But even that was things like mirroring all the screens or making every printer randomly produce questionable ASCII art instead of the requested print job.
Aside from the sheer scope of the outage, I think I actually have to give them some props here.
I defy any rational techie to even think about denying that mitigating the React bug, on a service that ubiquitous, is a code red, all hands on deck priority. It's serious, disclosed and in active use by threat actors. The moment your admins think they have a patching strategy, the pressure to deploy it yesterday, if not sooner, is immense. You're going to do testing, of course, but under that kind of pressure you're going to do the big tests, the obvious ones, but the risk of missing edge cases is real, and - in practical terms at least - largely unavoidable. I think it's highly likely that the process was accelerated to the point that it satisfied their change control requirements by a hair at most.
And this time they got bit. It happens, any of us that claim not to have been there is probably* either fibbing or new to the industry.
OK, so the change has now blown up in their faces. Again, we've all been there. Now they are in the balancing act between continuing to work on fixing it and pulling the eject handle and backing out the change. The folks on the sharp end are aware how long it will take to do the rollback, they can only speculate (intelligently perhaps, but still only speculate) how long the fix will take. Given how long it went from first awareness of the problem to service restoration I'd say they got that balance about right.
* ok, if you CAN honestly say it's never happened to you after you've been around the data centre for a decade or two I'm not going to call you a liar, just suggest you might want to buy a lottery ticket or two, given your luck.
Honestly, the person who set that shit up should've been fired. And they should've, at the very least, documented it or have put it inside a basement far away with signs saying "Beware of the leopard".
I see your HHGTTG and raise you a Pratchett...
If you put a big red button in a distant cave with a sign saying "End of the world switch, do not touch!" the paint wouldn't even have time to dry.
If the hyperscalers routinely apply US law to their data privacy practices and are not prepared to guarantee non-transfer of data that is legally protected under UK law outside the jurisdiction of that law then, honestly, they should not be permitted to operate IN the UK. Practically, of course, that will never happen. We need them and their infrastructure and the only practical place to get it is from the existing US-dominated providers. And these mega-corps will not give a damn about "regulations," they will only care about laws that have actual teeth.
When GDPR was introduced, the big data companies paid attention because violations carried enforceable fines of a significant percentage of their annual revenue. Something similar is needed for enforcing respect for the UKs data privacy rules and preventing the application of laws like the Patriot Act that have no applicability in the UK to UK citizens data.
And I'm not talking about programming languages, where most of us are fluent in half a dozen or so.
1: Regulatorian: This is the language of politicians and lawyers. It sets the mandates on banks, hospitals, schools etc. It contains nuances and terms of art that sometimes make a word mean something totally different to what you would infer if you heard it in general conversation.
2: Beancounterese: Spoken by accountantrs, salesmen and middle manglement. It sounds very similar to regulatorian but is sufficiently different in some of its meanings that it's as big a gulf as between old scots and english.
3: Geekian: The language of hard science, mathematics, real-world realities and the only one to use when specifying what a programmer needs to code. Because they will code what you tell them to, and it will work the way this language describes it.
The same word can mean different things in these three languages.
We have to be fluent in all three to accurately interpret requirements and predict what the emerging software will look like, to take error logs and demonstrate to (sometimes hostile) manglement what corrective action is needed and where it needs to be applied.
I buy a newspaper, but I tear out the classifieds and shred them unread. - Have I infringed copyright? Of course not.
To counter the argument "but you've already paid for it there, the website is free and the ads pay for it!" let's do the same with a free community paper I take home from the stack in the coffee shop. There, like the website, they make their only income from the ads. Just as before I detach the ad section and shred it unread. Still not copyright infringement.
The advertisers pay for the websites to serve their content. There is no requirement on me to render it unless I choose to.
Why do all those newfangled systems today no longer allow what could be done back in the mid-90s?
Because in the process of making everything "smart" the folks specifying and coding it have made ever-more-basic assumptions about how the workflows "should" operate, how the user "should" want to do things and how the admins "should" need to configure it.
Rather than give a wide menu of possibilities from which to build a configuration that fits its intended purpose on THIS site in THIS corporate culture.
And the customers manglement go along with it because that way they can employ more (cheap) trained monkeys and fewer (expensive) knowledgeable and professional sysadmins.
Good NCOs are the glue that holds any military together. ANY officer, newly hatched or vastly experienced, who doesn't immediately think twice on hearing a respectful "Are you sure about that, Sir?" from their senior NCO is overconfident. Even if they decide to go ahead anyway, they at least need to look again for anything they might have missed.
In principle, I like the idea of age checks to protect children from seeing pornographic content online. It’s easy to say that it should be up to the parents to carry out this protection, but many parents don’t have the capability to do so.
In my case, I treated my adolescent kids with the respect they merited and simply told them "Every device you can log into on my network goes through MY firewall to access the net. Yes, I know VPNs exist but I'm the sysadmin as well as the network admin and I'm already inside that loop. I am not blocking anything from you, not filtering anything apart from sites I filter for everybody because I know they are malicious. Just remember that I could, theoretically, go back and see everywhere you go from my net. It would be a pain in the arse so I will be extremely pissed off at you if you give me a reason to go look. Don't give me that reason."
Oddly enough, they ended up learning better digital hygiene than 90% of their peers and never did give me any reason to go check up on "where they'd been."
On local media and capable of a bare-metal restore to exactly how the system was when you backed it up. OS, apps, documents, settings - ALL of it. Drive died? Swap it, boot from a USB stick and restore your backup. That SHOULD be it, done. None of this crap about backing up only your documents, forcing you to recover first the vanilla OS, then reinstall all your other apps one by one, redo all your personal customised bits (see the long list of registry tweaks elsewhere in the comments, as an example) and then, finally, you can restore your backed up files and start to use them again...
Hell, have an option in your backup software to make your backup media bootable into PXE with a built-in option to "restore this backup"
If you can have file-level granularity on what you want to restore, that's honestly just gravy. If it can't save your arse when the magic smoke gets out, it's not really a backup.
MS backup hasn't done full system image backups for ages and it sucks.
The more panicky the phone call, text chat or email, the nastier you know this particular user to be, the more you wrap yourself in zen-like calm before heading out to confront 'em.
Only listen with half an ear to their spew of "having tried everything" unless you pick up on something specific they say they've done.
Once they wind down or run out of breath, a pleasant smile and "OK, let me have a look and I'll see what I can do."
Then fix it. Whatever it is.
They look a lot worse to any witnesses when all their ire runs off you like water off a duck's back.
And, as a bonus, it winds them up beautifully when all their sound and fury clearly signifies nothing.
Given the superiority of the GM system, it's a no-brainer for them to continue to use it. After all, "If it ain't broke..."
The arguments for expanding that system's use to other areas are also valid, given that it's a system in actuality, not vapourware. Implementation at other trusts should be hugely simplified (and cheaper) as a result.
There are the inevitable questions about whether the system would scale nationally, but it doesn't have to, at least not immediately. Even if each regional trust is running as a standalone environment, it would still benefit the residents of that area. Initially the system only has to scale to a regional trust, and it's already proven in one of the three most populous regions in England.
National integration as an end-state is obviously the goal, but wouldn't it make more sense to invest the effort and Taxpayer's money in creating an integration platform that operates as a distributed architecture, allowing all these (now-similar) regional nodes to interoperate? No great central repository of data (avoiding the whole single point of compromise risk) but distributed data, with each node being able to request data from a patient's "home" node if they are treated there, and to request a permanent transfer when somebody relocates and registers with a GP in their new area. That would almost certainly end up as a much simpler, cheaper, both more effective and less disruptive path to the goal of national health record availability than the big bang of trying to create and implement something entirely new.
HA! Another one who did an "Elite-like" setup for fun and found they liked it so kept it for everything :) Not just the keyboard's backlight but as a low-eyestrain dark theme.
Even ported it to my work PC until manglement decided they wanted conformity and demanded IT force a corporate theme on every desktop with GPOs.
(and yes, a decade on, I'm still enjoying the game as much as I enjoyed the BBC Micro version)
At the place I work, even though we use 2FA across the board, when we reset passwords we will only provide the reset password to the users manager or to their manager. If both are unavailable, the user is SOL until one of them returns, because providing a password directly to the impacted user is strictly verboten. This is to provide an extra level of protection against precisely this kind of social engineering.
"Ok it will be reset within a few minutes and your manager will contact you with your new temporary password..." - It not only breaks the social engineering, but it exposes when one of our Helldesk guys got caught and we play back the call looking for anything they missed. Not to hammer the guy, but to learn the enemy's tactics.
At one point I was asked to make some minor tweaks to some code written by one of these untrained types. I griped, hard, about being assigned the task, I knew what I was going to get - spaghettified, uncommented, the kind of mess that's been related in all the other comments here.
To my surprise this was not the case. The comments in the code were comprehensive and clear enough that the source truly was entirely self-documenting. Even where I found myself raising an eyebrow and thinking "I wouldn't have done it like that," what the original author had done, and why, was perfectly clear. Making the required tweaks was an absolute doddle as a result.
A few months later, I actually met the original author and complimented him on it. His response?
"Oh, I know I'm not a good programmer so I lay out the entire design as comments first, then just write the functional bits to match what the comments say it should do, one piece at a time."
"Keep doing that." I replied. "It works."
And based on the current behaviour of the administration, they'll lose the case, the courts will order them to rescind the tariffs and they'll ignore the court order, wipe their collective arse on the Constitution again and nothing will change.
The only upside to this is that civil litigation can't land you in jail but contempt of court can.
There was one time when a long-standing bug in a different package, developed in-house but by another team, had gone unfixed for so long it had whiskers.
Then a colleague of mine realised that as a "fresh set of eyes" on it, each of the two teams led the code review of the other's changes. So, every time his code called the other package's API he'd have a comment along the lines of "They still haven't fixed this. Apply dodgy workaround AGAIN."
After a month or two of this it was suddenly fixed - and in the comments of their code when we reviewed it was "Shut up, [colleague's name]"
"You can loath AD simply because evidently you don't have to manage large organizations security. It's not rare that even organizations using a lot of Linux servers rely on Windows Domain Controllers."
You may not be aware of this, if you are an AD specialist, but there's almost nothing you can do in AD that you can't do with tools available on linux or, indeed, any *nix out there. Since I have experience of managing both environments as a standalone ecosystem and many hybrid ones, I'm aware of both. The mixed environments tended to vary as to which one held the master back-end, often based upon what they set up originally. These design decisions tend to persist more tenaciously as the size of the org increases.
My current gig is at a corp with around 12000 workstations and servers to manage - it's windows-centric but with a significant number of linux servers alongside the windows ones. Here it's all done with AD. At my last position at a somewhat larger corp, their history was all unix servers of one flavour or another although the workstations were mostly windows and they now have a significant number of windows servers as well. Their infrastructure is now linux-based with AD taking a much more secondary role. Managing the two is different but they both work just fine and achieve the same things for their respective infrastructures.
Not just the wrong question in the sense of "you should be asking something else" but also "wrong" in a very fundamental way. It rests on the whole "Only criminals have anything to fear from this" fallacy that has been part of the authoritarian playbook utilised by everyone from the Gestapo to the Stazi, to the KGB - and ever more increasingly by the right wing of our supposed democracies, as they seem to be sliding more and more into the authoritarian space.
I've actually nothing against conservatism - a healthy democracy NEEDS a conservative voice participating in it to curb the sometimes-excessive enthusiasm of a diverse bunch of "progressives" and I can see that even though I probably fall into the progressive camp myself. But when the politics starts to slide into authoritarianism and autocracy, then I'd suggest that there are freedoms involved that should be "conserved"
There's a reason experienced typists handled early versions of wordstar and wordperfect so seamlessly.. The workflow was set up to mimic that of a typewriter. You'd set up all your document formatting before you started typing. with the settings all named the same as typists were familiar with. This design choice was why many typists stuck with wordperfect even when wordfperect was behind the curve and very late to the WYSIWYG party. Even after WP ended up with WYSIWYG (in version 4, I think it was?) you could still use that same workflow with the same keyboard shortcuts to access it while Word made this "typist's workflow" not impossible but harder to access.
Exactly this.
outsourcing between business entities has its advantages (IMHO far outweighed by the risks, but as I said, that's just my opinion) but when government - ANY government - tries it, it always ends up with crappy service and the business entity sucking so hard on the teats that the poor cash-cow practically turns inside out.
"The question is whether the cost to the rest of us exceeds the benefit to society by making some criminals easier to catch."
If that were the only cost to consider, it would be a no-brainer. Those of us that know what we're doing can always implement E2EE when communicating with somebody else who knows what they are doing. Losing the consumer-level services doesn't impact that comms channel at all, beyond a little mild swearing at having to set it up ourselves.
Except if "the cost to the rest of us" includes - as it does - the cost to everybody else, who have to rely on the consumer level services because they don't have the tech chops to set it up themselves, then that cost balloons massively as soon as one backdoor is compromised. It doesn't even require all the crims to be that tech-savvy either. The tech-savvy criminal will code up a one-click tool to open the compromised backdoor and then make their cash at once-removed by selling that tool to the folks that actually use it to extract private data for their schemes.
"If it can be compromised, sooner or later it will be" is a statement that is as fundamentally true as rule 34 of internet pr0n.
And so when we compare that risk to society at large vs the risk that "some criminals" will perhaps escape because they weren't made "easier to catch" it's pretty obvious which way the scales will tip.
As a sysadmin at a major hospital I was responsible for the care and feeding of a PDP-11 that ran a horrendous flat-file database of text documents (histopathology and cytopathology reports) that needed a regular archiving procedure run to avoid it filling up. That archiving process had a maximum document size smaller than the primary database for some insane reason so occasionally on the larger documents it would puke and stall the entire batch. The largest documents to be found on that system were autopsy reports. When the archive job stalled, I had to identify the document responsible and manually break it into two linked documents, with the requirement that I didn't break it up in such a way that reading either half standalone could give apparently incorrect info. This was why they wanted a techie from a bioscience background, which was why I got the job.
Every document that stalled the process I had to actually read and understand before choosing where to split it.
Twice, I encountered the report on the death of somebody I knew in this process.
But we did tell you so.
"We" in this case being sysadmins, net admins, in-house developers and all the other folks who actually understood these risks way better than the suits in the executive suite.
We warned about it when IT operations were outsourced the first time around. Upper manglement ignored the warnings because it saved money, forgetting the adage that you can only have two of fast, cheap or good. And when you outsource it, you don't get to pick which two any more. A lot of enterprises got burned. I, personally, made a fair stack of cash contracting on projects to bring operations back in house at a few places.
We warned about it again when the idea about saving money by outsourcing the storage infrastructure to the cloud came along. We got exactly the same pushback from exactly the same folks. It was cheaper so they were going to do it. All our arguments about putting our company's crown jewels in somebody else's control fell on deaf ears.
Along came the rapacious feature creep of software "subscriptions" and SAAS, where you weren't even running your applications locally, and we warned about that too.
So yeah. We told you so. And we don't even get to enjoy the schadenfreude, because we're working our behinds off trying to mitigate these risks and fix the mess - those of us that didn't just shrug our way into retirement or weren't laid off as part of the "savings."
Even today, perl has its place - because of the "regex thingy"
Peri is simply easier than almost anything else when it comes to sometimes-quite-complex regexes. I doubt that makes me a "genius" at parsing them, but I've never had a problem with script readability because of it.
I can't count the number of places I've been when test restores weren't done.
I admit I've even been guilty of it myself once. I had quite a nice little network at my home office, all of it built from previous-generation hardware purchased cheap from various employers. Along the way I'd also managed to acquire a license for a decent network backup suite and a small broken-but-repairable 5-slot DLT robot.(repairing that thing was pure geeky heaven and generated a significant amount of smug when successful, but that's another story) and a couple of boxes of unused tapes that the company just wanted gone - Full backup of my entire network on the first Sunday of each month and incrementals every Sunday night thereafter. Just needed to remember to swap the tapes out each month.
I did partial test restores to one machine per month, just enough to make sure my backups were ok... Then I acquired another server, set it up, got it into the backup system... and had a brainfart and didn't add it to the script that kicked off after each full backup and selected the next machine on its list, restored the backed-up /etc to /backup-verify/etc and emailed me if the directory contents differed.
Of course, we all know which machine decided to commit spectacular suicide and its backup then proved to be garbage.
"So, did the gun match the bullets ? Did it ? Because I believe we have actual reliable technology to determine that. And if it did, then screw this facial recog stuff, that guy is guilty."
IANAL but my understanding of US rules of evidence is that if the search warrant was invalid, nothing they found under it is admissible - I believe they call it "fruit of the poison tree."
So no evidence derived from the gun recovered in an improperly issued search warrant can be used. I believe that there are nuances around it but as a rule of thumb "If they wouldn't have it without the improper search, it doesn't exist for the purposes of investigating or prosecuting the case"
That's why the screwup by the cops is so egregious, because even if the guy IS guilty, now no evidence recovered in the search, or anything derived from it, is usable at all and if they want to prove their case they have to do it without any of that. They'd have been better off not making the invalid search in the hope that they could gather a VALID lead to that evidence later, one that wouldn't poison the results of recovering it.